A Novel Machine Learning-based Diagnostic Algorithm for Detection of Onychomycosis through Nail Appearance

Onychomycosis is the most common nail fungus disease in clinical practice worldwide, caused by localization of various fungal agents, including dermatophytes, on nail. The tests traditionally used for diagnosing onychomycosis are native examination, histopathological examination with periodic acid Schiff (PAS) staining, and culture. There no gold standard method disease, diagnosis process time-consuming, costly, quite laborious. Today, new technologies needed to detect via AI-based ML reduce clinician laboratory-induced error rate increase diagnostic sensitivity reliability. present study aimed design a decision support system help specialist doctor toenail artificial intelligence-based image processing techniques. images were taken any camera initially from individuals referred clinic. divided into 12 RGB channels. Three hundred features removed each channel as 25 time domain. best selected through feature selection algorithms next step performance number features, models created algorithm classification. average values all proposed models, accuracy, sensitivity, specificity, 89.65, 0.9, 0.89, respectively. successful model-created specificity 97.25, 0.96, 0.98, Although method, according findings obtained study, has many advantages compared literature, it can be diagnosis.