A Web-Based Explainable AI Framework for Unified Detection of Lung-Related Diseases from Chest Radiographs
An explainable AI platform that classifies chest radiographs into four conditions — COVID-19, bacterial pneumonia, viral pneumonia, or normal — and reveals the visual evidence behind every prediction with Grad-CAM heatmaps and per-class confidence scores.
4
Lung conditions
Grad-CAM
Visual evidence
OOD-aware
Safety checks
How It Works
Three steps from radiograph to explainable result.
Upload a Chest X-ray
AI Analysis & Safety Check
Diagnosis with Grad-CAM Evidence
What We Detect
COVID-19
Pneumonia caused by SARS-CoV-2, often presenting as bilateral, peripheral ground-glass opacities.
Viral Pneumonia
Lung inflammation from a viral infection, typically showing diffuse, interstitial patterns.
Bacterial Pneumonia
Bacterial lung infection that commonly appears as focal, lobar consolidation.
Normal
No signs of the targeted lung conditions are detected in the chest radiograph.
What is Grad-CAM?
Grad-CAM (Gradient-weighted Class Activation Mapping) overlays a heatmap on the radiograph, highlighting the regions that most influenced the model's prediction. This shows the visual evidence behind every result — not just a label — supporting transparent, trustworthy interpretation.