The use of Smartphone Audio and Artificial Intelligence to diagnose several health diseases

Abstract

Artificial intelligence (AI) has made significant progress in various fields, including healthcare. However, its implementation faces challenges such as precision, resilience, and interpretability. AI algorithms rely on signal manipulation and analysis, which can convey visual or auditory data. While image-based methodologies are increasingly recognized, AI applications in healthcare using audio data remain limited. Current audio methodologies often prioritize speech data analysis due to the intricate interaction between anatomical structures and neurological processes. This research project aims to investigate the viability of using audio recordings from smartphones for diagnosing various health conditions, such as COVID-19, TB, stroke, Vitamin D deficiency, hypertension, diabetics, sleep disorders, and mental health diseases. Early detection is crucial for effective management and treatment, and traditional diagnostic techniques are expensive and lengthy, making them inaccessible to a large population, particularly in low-income countries.