AI-Powered Microscopic Diagnostic Techniques for <i>Candida albicans</i> Detection: A Systematic Review.

Abstract

<h4>Background</h4>Artificial intelligence (AI) powered technologies can help detect <i>Candida albicans</i> (<i>C. albicans</i>) infections, which are a public health challenge due to increasing incidence rates and conventional therapy resistance.<h4>Purpose</h4>This review explores recent advancements, methodologies, and clinical implications in the AI-driven microscopic detection of <i>C. albicans</i>.<h4>Materials and method</h4>A literature search was conducted across multiple databases, including PubMed, Scopus, Embase, Web of Science, and Google Scholar. Following a thorough review of the retrieved articles, 7 studies were selected for inclusion in this review.<h4>Results</h4>This review analyzed 7 studies that employed AI and machine learning (ML) to detect the presence of <i>C. albicans</i>. The most commonly used dataset for detecting <i>C. albicans</i> through AI was microscopic images. Two studies employed time-lapse microscopy, and another study used the microorganism's smell fingerprint or volatile organic compounds with an impressive accuracy of 97.70%. The accuracy of detecting <i>C. albicans</i> through AI using microscopic images ranged from 63% to 100% depending on the model used.<h4>Conclusion</h4>AI can improve the detection of <i>C. albicans</i> infections. It can enhance the accuracy, speed, and efficiency of detection, providing clinicians with invaluable support in identifying infections earlier, optimizing treatment strategies, and ultimately improving patient outcomes.