carywoods commited on
Commit
2c07438
Β·
verified Β·
1 Parent(s): 61464b5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +94 -0
README.md CHANGED
@@ -1 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: gpl-3.0
3
+ language:
4
+ - en
5
+ metrics:
6
+ - accuracy
7
+ pipeline_tag: feature-extraction
8
+ tags:
9
+ - medical
10
+ ---
11
+ # 🧾 Model Card: Maternal Morbidity Risk Classifiers (Indiana 2010–2020)
12
 
13
+ **Model Name:** `maternal-morbidity-classifier-suite-indiana-2010-2020`
14
+ **Version:** 1.0
15
+ **Developed by:** Dr. Cary Woods, HarnessAI
16
+ **License:** [GNU General Public License v3.0](https://www.gnu.org/licenses/gpl-3.0.en.html)
17
+ **Release Date:** May 2025
18
+ **Trained on:** 813,837 public birth and infant death records from Indiana (2010–2020)
19
+
20
+ ---
21
+
22
+ ## πŸ”¬ Research Context
23
+
24
+ These models were developed as part of the research paper:
25
+ **"Predictive Modeling of Maternal Morbidity: Insights from a Decade of Regional Birth Data (2010–2020)"**
26
+ Published on ResearchGate, May 2025.
27
+ πŸ“„ DOI: *[insert DOI here]*
28
+
29
+ The study investigates the use of machine learning to predict maternal birth complications using administrative birth record data. The work focuses on model interpretability, sensitivity to rare events, and the challenges posed by missing socioeconomic indicators in public health datasets.
30
+
31
+ ---
32
+
33
+ ## πŸ“Œ Overview
34
+
35
+ This release includes three supervised learning classifiers trained to predict maternal morbidity from Indiana birth records: Logistic Regression, Random Forest, and Gradient Boosting. Each model uses the same engineered feature set and preprocessing pipeline. The classifiers were trained to identify rare maternal complication outcomes and are optimized for high recall.
36
+
37
+ ---
38
+
39
+ ## πŸ“Š Intended Use
40
+
41
+ - Triage support for public health researchers and analysts
42
+ - Educational use in public health informatics curricula
43
+ - Demonstration of risk modeling in imbalanced health datasets
44
+ - Not for direct clinical deployment without validation
45
+
46
+ ---
47
+
48
+ ## πŸ“ˆ Performance
49
+
50
+ | Model | Precision | Recall | F1 Score | ROC-AUC |
51
+ |---------------------|-----------|--------|----------|---------|
52
+ | Logistic Regression | 0.75 | 0.70 | 0.72 | 0.81 |
53
+ | Random Forest | 0.80 | 0.77 | 0.78 | 0.86 |
54
+ | Gradient Boosting | 0.83 | 0.80 | 0.81 | 0.89 |
55
+
56
+ Gradient Boosting showed the highest overall performance. Logistic Regression offers strong interpretability, and Random Forest provides a reliable non-linear baseline.
57
+
58
+ ---
59
+
60
+ ## πŸ§ͺ Limitations
61
+
62
+ - Omits social determinants like race, education, and income
63
+ - Reflects administrative data only (not clinical records)
64
+ - Binary outcome may oversimplify severity levels
65
+ - Developed on Indiana data; needs regional validation
66
+
67
+ ---
68
+
69
+ ## πŸ”„ Reuse & Redistribution
70
+
71
+ This model suite is released under **GPL v3**. You may reuse, modify, and redistribute it, provided derivative works maintain this license and include attribution.
72
+
73
+ ---
74
+
75
+ ## πŸ“¦ Files Included
76
+
77
+ - `model_lr.joblib` β€” Logistic Regression classifier
78
+ - `model_rf.joblib` β€” Random Forest classifier
79
+ - `model_gb.joblib` β€” Gradient Boosting classifier
80
+ - `test_models.py` β€” test script
81
+ - `README.md` β€” Usage guide
82
+ - `LICENSE.txt` β€” GPL v3 license
83
+
84
+ ---
85
+
86
+ ## πŸ”¬ Research Context
87
+
88
+ These models were developed as part of the research paper:
89
+ **"Predictive Modeling of Maternal Morbidity: Insights from a Decade of Regional Birth Data (2010–2020)"**
90
+ Published on ResearchGate, May 2025.
91
+ πŸ“„ DOI: [http://dx.doi.org/10.13140/RG.2.2.26163.13608](http://dx.doi.org/10.13140/RG.2.2.26163.13608)
92
+
93
+ The study investigates the use of machine learning to predict maternal birth complications using administrative birth record data. The work focuses on model interpretability, sensitivity to rare events, and the challenges posed by missing socioeconomic indicators in public health datasets.
94
+
95
+ ---