Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,46 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
metrics:
|
| 4 |
+
- accuracy
|
| 5 |
+
- f1
|
| 6 |
+
- precision
|
| 7 |
+
- recall
|
| 8 |
+
- roc_auc
|
| 9 |
+
tags:
|
| 10 |
+
- cancer
|
| 11 |
+
- MLP
|
| 12 |
+
- multi-classification
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# Serum-MiR-CanPred: An Artificial Intelligence-Driven Framework for Pan-Cancer Prediction Using a Minimal Set of Circulating miRNA Biomarkers
|
| 16 |
+
|
| 17 |
+
## Dataset
|
| 18 |
+
- Name: GSE212211, GSE113740, GSE211692, GSE164174
|
| 19 |
+
- Source: [GEO Database] (https://www.ncbi.nlm.nih.gov/geo/)
|
| 20 |
+
- Preprocessing: #TODO
|
| 21 |
+
|
| 22 |
+
## Training Details
|
| 23 |
+
- Number of Layers: 3
|
| 24 |
+
- Units: 512
|
| 25 |
+
- Batch Size: 256
|
| 26 |
+
- Epochs: 100
|
| 27 |
+
- Dropout Rate: 0.4
|
| 28 |
+
- Learning Rate: 0.00032506541805349084
|
| 29 |
+
- Hardware: NVIDIA GeForce RTX 3050
|
| 30 |
+
|
| 31 |
+
## Performance
|
| 32 |
+
The model was validated on 4055 samples (20% of the dataset)
|
| 33 |
+
| Metric | Score |
|
| 34 |
+
|--------|-------|
|
| 35 |
+
| Accuracy | 96% |
|
| 36 |
+
| Precision (weighted avg) | 96% |
|
| 37 |
+
| Recall (weighted avg) | 96% |
|
| 38 |
+
| F1-Score (weighted avg) | 0.96 |
|
| 39 |
+
|
| 40 |
+
Classification Report and confusion matrix can be found in the repository.
|
| 41 |
+
|
| 42 |
+
## Citation
|
| 43 |
+
If using this model, please cite:
|
| 44 |
+
|
| 45 |
+
## License
|
| 46 |
+
MIT License
|