AI-Driven Multi-Disease Detection System Using Deep Learning for Brain Tumor and Pneumonia

Abstract

This paper presents an intelligent and automated system for detecting brain tumors and pneumonia using advanced deep learning and image processing techniques. The proposed system is designed to analyze medical images such as brain MRI scans and chest X-ray images to identify the presence of diseases accurately. Convolutional Neural Networks (CNNs) are employed with transfer learning to improve classification performance while reducing training time. Various image preprocessing techniques, including resizing, normalization, noise reduction, and contrast enhancement, are applied to improve the quality of input images. Additionally, data augmentation methods are used to increase dataset diversity and enhance model generalization. The trained models are integrated into a Flask-based web application, allowing users to upload images and receive real-time predictions. The system produces results in a clear and understandable format, indicating whether the image is normal or affected. By automating the detection process, the system reduces diagnosis time, minimizes human error, and supports early medical intervention, making it highly beneficial in modern healthcare environments.

Introduction

This work presents an AI-based medical image analysis system designed to improve early detection of brain tumors and pneumonia using deep learning.

The motivation stems from limitations in traditional diagnosis, which relies on radiologists and is often slow, inconsistent, and less accessible in rural areas. To address this, the proposed system uses Convolutional Neural Networks (CNNs) to automatically analyze MRI and chest X-ray images for disease detection.

The study identifies key gaps in existing research, including single-disease models, lack of real-time deployment, limited usability, and poor explainability. Most current systems are also standalone and not integrated into practical healthcare platforms.

The proposed solution integrates multiple CNN models into a unified web-based system built with Flask. It includes preprocessing (noise removal, resizing, normalization), transfer learning for better accuracy, and a layered architecture that processes images and outputs predictions in real time through a user-friendly interface.

Implementation uses TensorFlow and Keras with data augmentation techniques to improve robustness. The system is evaluated using standard metrics like accuracy, precision, recall, and F1-score, showing strong performance and fast response time in real-world conditions.

Conclusion

The proposed AI-based system for brain tumor and pneumonia detection represents a significant advancement in medical image analysis. By combining deep learning techniques with image processing and web-based deployment, the system provides an efficient and reliable solution for disease detection. The high accuracy and real-time capabilities of the system make it suitable for practical healthcare applications. The system reduces dependency on manual diagnosis and minimizes the risk of human error, thereby improving diagnostic efficiency. Its user-friendly interface ensures accessibility for both technical and non-technical users. Overall, the project demonstrates the potential of artificial intelligence in transforming healthcare and provides a foundation for future research and development in this field.

References

The proposed AI-based system for brain tumor and pneumonia detection represents a significant advancement in medical image analysis. By combining deep learning techniques with image processing and web-based deployment, the system provides an efficient and reliable solution for disease detection. The high accuracy and real-time capabilities of the system make it suitable for practical healthcare applications. The system reduces dependency on manual diagnosis and minimizes the risk of human error, thereby improving diagnostic efficiency. Its user-friendly interface ensures accessibility for both technical and non-technical users. Overall, the project demonstrates the potential of artificial intelligence in transforming healthcare and provides a foundation for future research and development in this field.

Copyright

Copyright © 2026 K Tejaswi, B Gowthami , G Sneha , N Jagadeswari. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.