Machine Learning is becoming a powerful approach for integrative analysis of whole slide histology images. More recently, convolutional neural networks (CNNs) have been used heavily on histopathology images for predictive tasks like, tumor classification, mutation prediction and classification of cancer subtypes. But for these work target prediction have been slide level labels instead of fine-grained labes from small group of cells. This model called ST-Net consist of a deep learning algorithm that combines spatial transcriptomics and histology images to capture high-resolution gene expression heterogeneity. The model is trained on 30,612 spots in 68 breast tissue sections from 23 patients with breast cancer. The whole-slide images being very large (10,000 × 10,000 pixels) cannot be directed used as inputs to CNN, hence patches of 224 × 224 pixel are extracted from the whole-slide images centred on the spatial transcriptomics spots. All experiments are performed using the PyTorch machine learning library.
Input variables : Whole-slide histopathology image
Output Variables : Gene expression
| Statistical | : | Somers D | Accuracy | Precision and Recall | Confusion Matrix | F1 Score | Roc and Auc | Prevalence | Detection Rate | Balanced Accuracy | Cohen's Kappa | Concordance | Gini Coefficent | KS Statistic | Youden's J Index |
| Business | : | Population at High Risk of Disease | Risk by Geography | Risk by Demographics | Risk by Clinical Parameters | Optimized Hospital Resource Utilization | Decreased Cost of Care | Decreased Patient Visits |
| Infrastructure | : | Log Bytes | Logging/User/IAMPolicy | Logging/User/VPN | CPU Utilization | Memory Usage | Error Count | Prediction Count | Prediction Latencies | Private Endpoint Prediction Latencies | Private Endpoint Response Count |
Visit Model : github.com
Additional links : nature.com
| Model Category | : | Public |
| Date Published | : | June, 2020 |
| Healthcare Domain | : |
Life Sciences
Provider |
| Code | : | github.com |
| Health Risk Management |
| Health Risk Prediction |