Normalize Foundational Analysis case ordering

foundational_analysis/cases/case01/case01_01.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_01.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 11.81567963841674,
+                    "tolerance": 1.181567963841674
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_02.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_02.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 9.26960186255792,
+                    "tolerance": 0.926960186255792
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_03.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_03.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 11.932442037349787,
+                    "tolerance": 1.1932442037349789
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_04.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_04.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 5.878455525394873,
+                    "tolerance": 0.5878455525394873
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_05.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_05.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 7.919897825685701,
+                    "tolerance": 0.7919897825685701
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_06.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_01.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 13.560618171116563,
+                    "tolerance": 1.3560624385726978
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_07.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_02.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 12.075784079421299,
+                    "tolerance": 1.2075783110068763
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_08.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_03.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 12.149602175318517,
+                    "tolerance": 1.2149599789598067
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_09.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_04.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 14.652083592026083,
+                    "tolerance": 1.4652082512405658
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_10.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_05.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 13.166214068045193,
+                    "tolerance": 1.3166212097326537
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_11.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-01.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-01.cnt",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 4.836610733624954,
+                    "tolerance": 0.4836610733624954
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_12.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-02.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-02.cnt",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 6.833748557175907,
+                    "tolerance": 0.6833748557175907
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_13.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-03.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-03.cnt",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 7.544371519964062,
+                    "tolerance": 0.7544371519964063
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_14.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-04.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-04.cnt",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 13.23650829225669,
+                    "tolerance": 1.3236508292256692
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_15.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-05.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-05.cnt",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 9.461418170390266,
+                    "tolerance": 0.9461418170390266
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_16.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_01.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 25.60839634754504,
+                    "tolerance": 2.560839634754504
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_17.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_02.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 26.963042816030004,
+                    "tolerance": 2.6963042816030005
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_18.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_03.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 58.63472717305611,
+                    "tolerance": 5.863472717305612
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_19.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_04.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 21.885057050117236,
+                    "tolerance": 2.188505705011724
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_20.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_05.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 36.11783061563106,
+                    "tolerance": 3.6117830615631057
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_21.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_01.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 41.33018074506845,
+                    "tolerance": 4.133018074506846
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_22.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_02.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 18.30015063387382,
+                    "tolerance": 1.830015063387382
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_23.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_03.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 23.478518914155703,
+                    "tolerance": 2.3478518914155706
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_24.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_04.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 40.95759770354007,
+                    "tolerance": 4.0957597703540065
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case01/case01_25.json

@@ -0,0 +1,27 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_05.edf",
+        "instruction": "Please first extract only the EEG channels from the raw signal, and then apply a 0.5-40Hz FIR bandpass filter to these channels. After filtering, obtain each channel's Alpha-band power and total filtered-signal power by integrating the PSD over frequency. Calculate Alpha relative power separately for each channel, then average the channel-wise ratios and report the final percentage value clearly in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the 'Alpha Relative Power Percentage' from the agent's report.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units (%).\n4. The Key in the JSON must be EXACTLY \"alpha_ratio\".\n5. The Value must be a FLOAT (e.g., if the report says 25.5%, output 25.5).\n6. If the value is missing or cannot be found, return {\"alpha_ratio\": null}.\n\n### OUTPUT TEMPLATE\n{\"alpha_ratio\": <float>}",
+        "metrics": [
+            {
+                "metric_id": "alpha_ratio_accuracy",
+                "type": "numeric_check",
+                "target_key": "alpha_ratio",
+                "weight": 100,
+                "params": {
+                    "gt_value": 18.87056907818833,
+                    "tolerance": 1.8870569078188328
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_01.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_01.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.344264,
+                    "tolerance": 0.234426
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 3.603543,
+                    "tolerance": 0.360354
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_02.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_02.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.950857,
+                    "tolerance": 0.095086
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.989856,
+                    "tolerance": 0.098986
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_03.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_03.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.168239,
+                    "tolerance": 0.116824
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.684525,
+                    "tolerance": 0.168453
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_04.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_04.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.161963,
+                    "tolerance": 0.116196
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.076704,
+                    "tolerance": 0.20767
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_05.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "ISRUC_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "ISRUC"
+    },
+    "agent_input": {
+        "data_path": "data/core/ISRUC_05.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.669702,
+                    "tolerance": 0.06697
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.774508,
+                    "tolerance": 0.077451
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_06.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_01.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.293048,
+                    "tolerance": 0.029305
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.232857,
+                    "tolerance": 0.023286
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_07.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_02.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.323811,
+                    "tolerance": 0.032381
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.202792,
+                    "tolerance": 0.020279
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_08.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_03.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.264357,
+                    "tolerance": 0.026436
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.244232,
+                    "tolerance": 0.024423
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_09.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_04.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.408242,
+                    "tolerance": 0.040824
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.212913,
+                    "tolerance": 0.021291
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_10.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "BCIC2020-3_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "BCIC2020-3"
+    },
+    "agent_input": {
+        "data_path": "data/core/BCIC2020-3_05.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.184103,
+                    "tolerance": 0.01841
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.181707,
+                    "tolerance": 0.018171
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_11.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-01.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-01.cnt",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.372668,
+                    "tolerance": 0.037267
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.702498,
+                    "tolerance": 0.17025
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_12.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-02.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-02.cnt",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.048704,
+                    "tolerance": 0.10487
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.825685,
+                    "tolerance": 0.282569
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_13.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-03.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-03.cnt",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.618508,
+                    "tolerance": 0.061851
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.884954,
+                    "tolerance": 0.288495
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_14.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-04.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-04.cnt",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.826021,
+                    "tolerance": 0.082602
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.513809,
+                    "tolerance": 0.151381
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_15.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "SEED-V-05.cnt",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "SEED-V"
+    },
+    "agent_input": {
+        "data_path": "data/core/SEED-V-05.cnt",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.554786,
+                    "tolerance": 0.155479
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 7.43622,
+                    "tolerance": 0.743622
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_16.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_01.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.904181,
+                    "tolerance": 0.090418
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.133355,
+                    "tolerance": 0.113335
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_17.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_02.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 3.290876,
+                    "tolerance": 0.329088
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.536704,
+                    "tolerance": 0.15367
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_18.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_03.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 6.319008,
+                    "tolerance": 0.631901
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.504666,
+                    "tolerance": 0.150467
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_19.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_04.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.254156,
+                    "tolerance": 0.125416
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 5.571346,
+                    "tolerance": 0.557135
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_20.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "Mumtaz2016_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "Mumtaz2016"
+    },
+    "agent_input": {
+        "data_path": "data/core/Mumtaz2016_05.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 3.528239,
+                    "tolerance": 0.352824
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.723892,
+                    "tolerance": 0.272389
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_21.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_01.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_01.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.952141,
+                    "tolerance": 0.295214
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.846158,
+                    "tolerance": 0.084616
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_22.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_02.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_02.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.128332,
+                    "tolerance": 0.112833
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.040372,
+                    "tolerance": 0.104037
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_23.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_03.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_03.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.035565,
+                    "tolerance": 0.103557
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.463546,
+                    "tolerance": 0.046355
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_24.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_04.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_04.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 2.892534,
+                    "tolerance": 0.289253
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.24287,
+                    "tolerance": 0.124287
+                }
+            }
+        ]
+    }
+}

foundational_analysis/cases/case02/case02_25.json

@@ -0,0 +1,37 @@
+{
+    "meta_info": {
+        "case_id": "MentalArithmetic_05.edf",
+        "bench_subset": "NeuroBench-Core",
+        "difficult": 1.0,
+        "original_dataset": "MentalArithmetic"
+    },
+    "agent_input": {
+        "data_path": "data/core/MentalArithmetic_05.edf",
+        "instruction": "Please select all available prefrontal EEG channels (FP*/AF* channels) from the raw signal; if the recording has no FP or AF channels, use F3/F4 as frontal proxies. Apply a 1-40Hz FIR bandpass filter and estimate the PSD using the Welch method. Obtain theta [4,8)Hz, alpha [8,13)Hz, and beta [13,30]Hz band power by integrating the PSD over each frequency band. First sum each band power across all selected channels, then calculate ABR = summed alpha power / summed beta power and TBR = summed theta power / summed beta power. Clearly report both final values in your response."
+    },
+    "eval_config": {
+        "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract two values from the agent's report:\n1) Alpha/Beta Ratio (ABR)\n2) Theta/Beta Ratio (TBR)\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks (e.g., no backticks or json tags).\n3. DO NOT include any conversational text, explanations, or units.\n4. The keys in the JSON must be EXACTLY \"abr\" and \"tbr\".\n5. Both values must be FLOAT.\n6. If one value is missing, return it as null.\n\n### OUTPUT TEMPLATE\n{\"abr\": <float|null>, \"tbr\": <float|null>}",
+        "metrics": [
+            {
+                "metric_id": "prefrontal_abr_accuracy",
+                "type": "numeric_check",
+                "target_key": "abr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 0.996986,
+                    "tolerance": 0.099699
+                }
+            },
+            {
+                "metric_id": "prefrontal_tbr_accuracy",
+                "type": "numeric_check",
+                "target_key": "tbr",
+                "weight": 50,
+                "params": {
+                    "gt_value": 1.322632,
+                    "tolerance": 0.132263
+                }
+            }
+        ]
+    }
+}