CaxtonEmeraldS commited on
Commit
029d259
·
verified ·
1 Parent(s): f2fefe2

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +115 -0
README.md ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ language:
4
+ - en
5
+ library_name: keras
6
+ tags:
7
+ - regression
8
+ - tensorflow
9
+ - keras
10
+ - colourimetric-sensing
11
+ - biosensing
12
+ - cholesterol
13
+ - research-demo
14
+ ---
15
+
16
+ # Cholesterol Concentration Predictor — ANN Model
17
+
18
+ ## Model summary
19
+
20
+ This model repository accompanies research on the use of artificial neural networks (ANNs) and an Au@CeO₂ (gold–cerium oxide) nanozyme-based colourimetric sensor for cholesterol estimation. The corresponding **Gradio application** enables users to compare ANN predictions with a linear-regression baseline using mean RGB colour measurements.
21
+
22
+ **Publication:** Govindharaj, P., Alungal, N., Emerald, C. S. and Kannan, S. (2026). *Au@CeO₂ nanozyme based smart colourimetric sensor for cholesterol: A neural network powered point of care solution model.* **Biochemical Engineering Journal**, **225**, 109908. DOI: [10.1016/j.bej.2025.109908](https://doi.org/10.1016/j.bej.2025.109908).
23
+
24
+ **Links:** [Associated publication](https://doi.org/10.1016/j.bej.2025.109908) · [Interactive Gradio Space](https://huggingface.co/spaces/CaxtonEmeraldS/CholesterolConcentrationPredictor) · [Researcher's website](https://caxtonemerald-s.github.io)
25
+
26
+ ## Intended uses
27
+
28
+ - Research and educational exploration of ANN-based colourimetric regression.
29
+ - Comparing predictions under different neural-network configurations **in the companion application**, to the extent that their trained files are available in its archive.
30
+ - Reproducing and examining the method when the original measurement protocol, preprocessing and training artefacts are available.
31
+
32
+ **Not intended for diagnosis, screening, monitoring, treatment selection or any other clinical use.** This repository does not establish medical-device validation.
33
+
34
+ ## Data and features
35
+
36
+ The application uses **three numerical inputs**: mean red, green and blue intensities. Each is expected to be in the range **0–255**. These values should come from appropriately collected colourimetric measurements, rather than arbitrary photographs or uncontrolled lighting conditions.
37
+
38
+ The application code uses this transform for inference:
39
+
40
+ ```python
41
+ x = [[R / 256.0, G / 256.0, B / 256.0]]
42
+ ```
43
+
44
+
45
+ ## Architecture and model coverage
46
+
47
+ The application describes a **3–N–1** feedforward ANN: three RGB features, one hidden layer of variable width `N`, and a single output. Approximately **2,500 trained model configurations**, varying activation function, random seed and hidden-neuron count.
48
+
49
+ The Gradio Space locates its model files under:
50
+
51
+ ```text
52
+ Models.zip
53
+ └── Models/
54
+ └── <activation>/
55
+ └── <seed>/
56
+ └── model_<number_of_hidden_neurons>.keras
57
+ ```
58
+
59
+ Configuration from the repository name or the publication's best-performing model can be inferred.
60
+
61
+ ## Baseline and evaluation
62
+
63
+ The application implements a linear-regression comparator using normalised RGB inputs:
64
+
65
+ ```python
66
+ linear_raw = (
67
+ 1.9221 * (R / 256.0)
68
+ - 1.3817 * (G / 256.0)
69
+ + 1.4058 * (B / 256.0)
70
+ - 0.1318
71
+ )
72
+ linear_mM = max(0.0, linear_raw) * 50.0
73
+ ```
74
+
75
+ The publication's abstract reports:
76
+
77
+ | Measure | Study-reported value |
78
+ |---|---:|
79
+ | Best validation MSE | 0.0577 (hard shrink, 13 hidden neurons) |
80
+ | Lowest training MSE | 0.0637 (SELU, 7 hidden neurons) |
81
+ | Median training MSE | 0.0896 |
82
+ | Median validation MSE | 0.0769 |
83
+ | Reported colourimetric linearity range | 0–50 mM |
84
+
85
+
86
+ ## How to use
87
+
88
+ **Recommended:** open the [companion application](https://huggingface.co/spaces/CaxtonEmeraldS/CholesterolConcentrationPredictor), supply calibrated mean RGB values, choose an available activation, seed and hidden-neuron count, and compare the outputs.
89
+
90
+ **For programmatic use:** examine the repository's actual files and loading instructions for the specific saved Keras model. The general TensorFlow/Keras pattern is:
91
+
92
+ ```python
93
+ import numpy as np
94
+ import tensorflow as tf
95
+
96
+ model = tf.keras.models.load_model("path/to/model.keras", compile=False)
97
+ x = np.array([[R / 256.0, G / 256.0, B / 256.0]])
98
+ raw = float(np.asarray(model.predict(x, verbose=0)).reshape(-1)[0])
99
+ predicted_mM = max(0.0, raw) * 50.0
100
+ ```
101
+
102
+ Replace `path/to/model.keras` with the downloaded model path.
103
+
104
+ ## Limitations, safety and licence
105
+
106
+ - Experimental results depend on sample preparation, sensor chemistry, calibration, lighting, camera characteristics, image processing and domain shift.
107
+ - Predictions are **not** calibrated confidence intervals and should not be used as medical measurements.
108
+ - Clamping negative outputs to zero does not correct model error or guarantee validity.
109
+ - The publication reports a range of 0–50 mM; extrapolation and comparisons outside its study conditions require separate validation.
110
+
111
+ ## Citation
112
+
113
+ Govindharaj, P., Alungal, N., Emerald, C. S. and Kannan, S. (2026). Au@CeO₂ nanozyme based smart colourimetric sensor for cholesterol: A neural network powered point of care solution model. *Biochemical Engineering Journal*, 225, 109908. https://doi.org/10.1016/j.bej.2025.109908
114
+
115
+ **Contact:** [Caxton Emerald S — professional website](https://caxtonemerald-s.github.io) · [ORCID](https://orcid.org/0000-0002-7763-3987)