This repository contains Projectwork ⬇️ undertaken for the partial fulfillment of Deep Learning Module and ETCS Creditpoints @OpenCampus.sh.
Objective : To train and explore Deep learning models for Medical Radiology Assistance ( Classification and Detection of Thorax Diseases) using Chest X-ray datasets.
-
Datasets
-
Implementation
-
References
-
NIH Clinical Center Chest x-ray datasets y | National Institutes of Health (NIH)
- ⬇️ Download Here
⚠️ Incase of warning with Image folder !! : runpython Utils/download.py
Status /Progress
-
Exploration of Datasets
-
Exploartion of Application Domain and Challenges [Concepts: @AI for medical Diagnosis (coursera)]
- Class Imabalance Problem
- Patient Overlap
- Explore Evaluation Metrics
-
Data Processing
- Data Preprocessing
- Image Processing
- Image Data Generator and Loaders
-
- Densenet-121
-
Model Evaluation
- Subset Dataset
- Full Dataset
-
Local to Colab setup
- [1. ] ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases
- [2. ] CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays [arXiv:1711.05225v3 [cs.CV] 25 Dec 2017]
- [3.] Densely Connected Convolutional Networks [arXiv:1608.06993v5 [cs.CV] 28 Jan 2018]
- [4.] Coursera - AI for medical Diagnosis
- [5.] Chest xray12 dataset problems : Single Radiologist Label Accuracy