Wavelet Transform-Based Virus Detection System for Screening Viral Epidemics using Medical Images

Backgrounnd: This research presents a Wavelet Transform-based Virus Detection System for rapid identification of viral infections in medical images, including CT scans, MRIs, and X-rays. The main objective of this innovative research initiative is to architect a system that integrates wavelet transforms for the purpose of scrutinizing various medical images (such as CT scans, MRIs, and X-rays) to identify viral infections that may emerge during epidemics. The primary focus revolves around establishing a streamlined pipeline for extensive virus detection, capable of efficiently and accurately identifying viral infections by leveraging wavelet-derived features through sophisticated feature extraction methods and other cutting-edge techniques. Methods: This unique system efficiently harnesses the power of Wavelet Transform to extract features at multiple resolutions from medical images, facilitating the comprehensive capture of both intricate and more general details pertaining to viral anomalies or lesions. Further enhancing this process, these wavelet-based features undergo rigorous evaluation using advanced feature selection methods like Principal Component Analysis (PCA) and Independent Component Analysis (ICA), ensuring that only the most pertinent features are incorporated into the classification process. To accurately discern infected from non-infected areas within medical images, state-of-the-art machine learning algorithms, including the Convolution Neural Network (CNN) based Improved AlexNet deep learning classifier, are employed. Results: The evaluation of the system is carried out through the assessment of standard metrics such as accuracy, sensitivity, specificity, and F1-score as part of its optimization for real-time deployment. This research presents a Wavelet Transform-based Virus Detection System for rapid identification of viral infections in medical images. The system achieved an accuracy of 97.8% and significantly reduced diagnosis time compared to traditional methods. Results indicate that it enhances early detection capabilities, proving essential for effective management during viral epidemics.