Leveraging HPC with GAN-CNN Integration for Enhanced Tuberculosis Detection using Synthetic Image Generation

Tuberculosis (TB) remains a significant global health challenge, especially in remote areas with limited medical resources. Early and accurate diagnosis is crucial to control its spread. This project harnesses high-performance computing (HPC) to address two key issues: managing unreliable medical data and improving TB diagnosis. Given the imbalance between healthy and TB-affected samples, Generative Adversarial Networks (GANs) are used to generate synthetic images, enriching the dataset and making it more diverse. This helps train Convolutional Neural Networks (CNNs), which are highly effective in medical image analysis. By leveraging HPC, we accelerate the training of CNNs on large, augmented datasets, significantly reducing processing time while maintaining accuracy. This system improves TB detection by combining GAN-enhanced data and CNN models, enabling faster and more reliable diagnosis.