NNXG: Privacy based image processing in Pneumonia Detection from Chest X-ray using Modified Neural Network Architecture and XGBoost

Pneumonia stands as a prevalent affliction claiming lives of both the young and the elderly worldwide, particularly prominent in underdeveloped regions. Its impact resonates through families, encompassing generations. This respiratory ailment ensues from microbial invasion of the lungs, necessitating prompt diagnosis for effective treatment. While conventional methods like chest X-rays aid in identification, challenges arise in discerning subtle anomalies or differentiating them from similar conditions. Hence, the integration of computer-aided diagnostic tools becomes imperative. Herein, we propose an innovative approach to pneumonia detection, amalgamating traditional deep learning networks with an XGBoost Classifier. Moreover, we advocate for the incorporation of digital watermarking technology to safeguard medical image privacy, integrity, and management. The retrieval algorithm ensures seamless access to concealed data when needed. Extensive validation of diverse algorithms based on performance and reliability culminates in conclusive findings and recommendations for future exploration.