Projet
Chest X-Ray Pneumonia Classification with Grad-CAM
Built and evaluated a MobileNetV2 transfer-learning classifier using fine-tuning and class weighting. Implemented Grad-CAM to visualize image regions influencing predictions and evaluated the model using accuracy, precision, recall, AUC, confusion matrix, and classification reports.
Achieved 91.51% accuracy, 90.80% precision, and 96.15% recall at a decision threshold of 0.85. The project strengthened my experience in deep learning, model evaluation, explainable AI, and medical image analysis.
Voir le travail