International Research Journal of Education and Technology

MediScan AI helps solve India's healthcare problems by making medical help easier to get, especially in villages where there are very few doctors and people wait too long for diagnosis. The system integrates two independent modules: a Natural Language Processing (NLP)-based symptom chatbot that engages users conversationally to collect symptoms (fever, cough, fatigue) and predict probable conditions using trained ML classifiers, and a Convolutional Neural Network (CNN) for X-ray abnormality detection targeting pneumonia, lung infections, and skin diseases with high precision through image preprocessing and deep feature extraction. Deployed via the Django web framework on accessible hardware (Intel i3, 8GB RAM, Windows 11), MediScan provides real-time preliminary diagnostics through a responsive HTML/CSS/JavaScript interface, eliminating specialist dependency. This modular architecture enables flexible usage—patients can utilize either symptom analysis, X-ray interpretation, or both sequentially. The enhanced report rigorously evaluates system performance through comprehensive testing (unit, integration, system, acceptance), benchmarked against existing AI medical tools like Ada Health and CheXNet. Key contributions include scalable healthcare AI deployment in resource-constrained settings, reduced clinician workload by 60-70% for preliminary screening, and alignment with India's National Digital Health Mission (NDHM). Future enhancements target multilingual NLP and mobile deployment for broader rural penetration. 

Keywords: MediScan AI, NLP chatbot, CNN X-ray analysis, healthcare accessibility, Django deployment, preliminary diagnostics, pneumonia detection, symptom analysis, medical AI, rural healthcare.