AI-Based Medical Assistance via Mobile Devices and Enhanced CNN for Diagnostic Imaging

Abstract

The COVID-19 pandemic exposed critical flaws in global healthcare diagnostics and emergency response systems, highlighting the urgent need for practical solutions. This dissertation addresses these challenges through three applied contributions incorporating artificial intelligence to enhance medical diagnosis of diseases (COVID-19 and brain tumors) via image classification, and crisis communication in both routine and emergency situations. The first contribution, presented in Chapter 3, is an Android-based mobile application called Help Pro: SOS Application. It allows users to send essential information including geographic location, identification, and type of emergency (vehicle problem, ambulance, blood donation, police, or firefighter) to preselected contacts or responders with a single touch. The app is optimized for limited-connectivity environments, with evaluation focusing on usability, response time, and system stability, aiming to minimize delays and support timely interventions. Chapter 4 presents the second contribution: a deep learning model named CXR-DNN, built on the EfficientNetB7 architecture for detecting COVID-19 in chest X-ray images. The model was trained and validated on multiple datasets and evaluated using key metrics such as accuracy, precision, recall, and F1-score. The study also addresses fairness, robustness, class imbalance, hardware limitations, and the transparency of AI decision-making. The third contribution, in Chapter 5, introduces a CNN-based joint diagnostic framework for classifying brain tumors using CT and MR images, employing MobileNetV3 over a six category dataset. Experimental results show that combining both imaging modalities significantly improves classification accuracy in neuro-oncology.

Chapter 6 summarizes results and proposes future research directions, including expanding diagnostic tools to additional illnesses and enhancing the SOS application. This dissertation combines artificial intelligence with healthcare and emergency management to deliver practical solutions with meaningful real-world impact.