Revista de métodos de diagnóstico médico
Acceso abierto

abstracto

Synergizing Machine Learning and Blockchain for Pioneering Early Disease Detection: A Focused Study on COVID-19 Diagnosis

Jayendra S. Jadhav

Early disease detection plays a pivotal role in modern healthcare, as it can substantially influence patient prognosis, healthcare costs, and overall public health. Machine learning algorithms serve as indispensable tools for uncovering subtle patterns, trends, and predictive factors present in complex medical data sources, such as patient records, diagnostic images, and genomic information. The integration of machine learning with Blockchain technology presents a substantial opportunity for transformative advancements in healthcare. This document examines several machine learning techniques such as LR, RF, GB, SVC, and GNB. It showcases their remarkable effectiveness in analysing symptoms for accurate disease detection, with COVID-19 serving as a primary case study. The application of cross-validation offered a sophisticated analysis of the performance capabilities, revealing that the Random Forest and Gradient-Boosting models are particularly effective, striking a vital balance in their metrics, which is vital for the reliable detection of diseases at their beginning. In addition, these models, with their significant accuracy (0.91) and precision (0.92), affirmed its status as an exceptional tool for the early identification of diseases. Ultimately, the combination of machine learning and Blockchain technologies significantly bolsters healthcare systems’ ability to detect and manage diseases early, enhancing our understanding of diseases and guiding public health measures and strategies.