International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 03 | Mar 2025
p-ISSN: 2395-0072
www.irjet.net
AI-BASED DISEASE DETECTION USING MACHINE LEARNING AND CONVOLUTIONAL NEURAL NETWORKS Dr. K. Soumya1, Gorla Sukanya2, Jakkula Madhurima3, Jeeri Alekhya Reddy4, Kallakuri Vyshnavi5, Dr. S. Pallam Setti6 1 Assistant Professor, Dept. of Computer Science and Systems Engineering, Andhra University College of
Engineering for Women, Andhra Pradesh, India
2-5B.Tech, Final year Student, Andhra University College of Engineering for Women, Andhra Pradesh, India 6Founder & CEO, Dr Pallam Setti Center for Research & Technology, A-hub, Andhra University, Visakhapatnam,
Andhra Pradesh, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Artificial Intelligence serves as a vital
predictions. This disease detection system aims to serve as an effective decision-support tool for individuals seeking initial medical evaluations.
healthcare component by detecting diseases with excellent precision. Biomedical researchers utilize Convolutional Neural Networks (CNN) with machine learning techniques to extract image features and sequential data through this AI-Based Disease Detection system which leads to improved model performance. The system requires different types of data input such as images and patient records and timeseries measurements for chronic kidney disease and Parkinson’s disease diagnosis. This hybrid approach which uses various neural networks ensures robust feature learning, temporal pattern recognition and optimization making it suitable for real world diseases and clinical applications. Experimental results demonstrate reduced false positives, improved accuracy, and better generalization across datasets. The AI system enables swift early diagnosis of diseases, feature selection and enhances medical care while facilitating better deliverance of health results to patients with applications of deep learning.
The core technology behind this system uses deep learning algorithms, particularly Convolutional Neural Networks (CNNs) and various Machine Learning(ML) algorithms. CNNs are employed for image-based diseases. On the other hand, Machine Learning algorithms are utilized for analyzing sequential medical data, such as patient health records, time-series data which are critical for conditions like chronic kidney disease, and Parkinson’s disease. The user-centric design ensures that individuals with minimal technical knowledge can benefit from AIdriven diagnostics, making healthcare more reachable. Despite the advancements in AI and machine learning for disease detection, most existing systems are limited to identifying a single disease. This lack of versatility forces individuals to rely on multiple tools for different conditions, leading to increased complexity and inefficiency in the diagnostic process. Additionally, many AI-based diagnostic applications are not user-friendly, making them difficult to access for individuals without technical expertise. So, this project aims to address these challenges by developing an AI-based disease detection system using Machine Learning with CNNs through a userfriendly web application.
Key Words: AI in healthcare, disease detection, CNN, chronic kidney disease, Parkinson’s, medical diagnostics, machine learning, early diagnosis, feature extraction, deep learning.
1.INTRODUCTION In past years, use of Artificial Intelligence (AI) in healthcare has completely changed how illnesses are identified and treated. For better treatment, an early and precise diagnosis is essential, but traditional diagnostic techniques are costly, time-consuming procedures, and depend on availability of medical professionals or assistance with specialised medical knowledge.
Most existing AI-based diagnostic systems are designed to detect only one specific disease, requiring separate tools for different conditions. This fragmentation increases the complexity for healthcare professionals and patients, making the diagnostic process inefficient. Existing techniques often struggle to integrate different types of medical data (e.g., images, clinical text, lab results) effectively. This limits the ability to provide a comprehensive diagnosis based on multiple data sources. Many AI-powered healthcare tools are designed for medical professionals who have technical knowledge, making them inaccessible to the general public, especially in rural areas.
To address these challenges, our project, "AI-Based Disease Detection System" aims to develop a comprehensive, user-friendly web application that facilitates the early detection of Chronic kidney disease and Parkinson’s disease. The application allows users to input relevant medical information which is then processed by specialized AI models to provide accurate
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 781