Cloud Health Diagnostics: AI Medical Scan Classifier

This project implements a deep learning pipeline for early-stage medical diagnosis, deployed using AWS SageMaker. It first detects the scan type (X-ray, CT, or MRI), then performs a preliminary classification using trained CNN models. The system is designed to assist medical professionals and patients by offering early insights ahead of formal diagnosis.

Tools and Libraries:

• Python, AWS SageMaker, TensorFlow, Keras
• Deep Learning Models: CNN, DenseNet121, ResNet50, VGG16, VGG19
• Pandas, NumPy, OpenCV, Matplotlib, Seaborn

Key Features:

• Classifies input images as X-ray, CT, or MRI before diagnosis
• Runs inference using pre-trained convolutional architectures
• Achieves 67–83% accuracy across different image types and model variants
• Enables early insights for quicker intervention in resource-limited settings
• Deployed on AWS SageMaker for scalable and low-latency cloud access

Future Goals:

• Deploy GUI for interactive upload and result display
• Integrate patient history metadata for better diagnostic accuracy
• Expand dataset across more conditions and imaging modalities
• Enable API access for hospital management systems

Outcome:

Built as a scalable, assistive diagnostic tool to support remote and urban clinics alike. The system speeds up the diagnostic process by automating early scan interpretation and reducing initial workload for radiologists.