ADVANCING AI FOR COMPUTER-AIDED DETECTION SYSTEMS IN CHEST X-RAY DIAGNOSTICS
DOI:
https://doi.org/10.62647/Abstract
Chest X-ray diagnostics play a crucial role in detecting and diagnosing various respiratory conditions, including pneumonia, tuberculosis, lung cancer, and COVID-19. Radiologists analyse X-ray images to identify abnormalities, but the process is time-consuming and prone to human error. The history of chest X-ray diagnostics has evolved from manual interpretation to computer-assisted detection methods, enhancing accuracy and efficiency. Early prediction models relied on statistical techniques and rule-based systems, which lacked adaptability to complex patterns. Before AI advancements, diagnosis depended on radiologists' expertise, with methods such as manual film analysis and basic image processing techniques. These approaches required significant time, leading to delays in treatment. Several early computer-aided detection systems provided limited support, but they lacked precision and required constant human intervention. The increasing demand for rapid and accurate diagnoses, combined with the limitations of manual analysis, has driven the need for an AI-powered approach. Human dependency, misdiagnosis due to fatigue, and inefficiency in handling large volumes of data are significant challenges in conventional diagnostic methods. Machine learning models trained on vast datasets can accurately classify chest diseases, reducing diagnostic errors and improving patient outcomes. Integrating deep learning techniques with a web application enables automated, real-time predictions, allowing users to upload X-ray images and receive instant analysis. This system enhances accessibility and supports healthcare professionals in making informed decisions. By leveraging convolutional neural networks, image preprocessing, and predictive analytics, the proposed system significantly improves diagnostic accuracy, speed, and reliability, minimizing the burden on medical professionals while ensuring early disease detection.
Downloads
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.