Assessing the Validity and Impact of Remote Digital Image Reading in Fungal Diagnostics

Abstract

Mycological diagnostics play a crucial role in patient management and treatment of invasive fungal infections. Despite the significant global burden of fungal diseases, awareness and diagnostic capabilities in mycology laboratories lag behind other microbiological disciplines. Mycological diagnostics often require microscopic analysis of clinical samples and culture. The interpretation of microscopy requires extensive expertise in clinical mycology. This study aimed to explore the feasibility of remote digital reading for preliminary identification of fungi. In this study, five mycology-trained participants were asked to analyze a total of 474 images divided into three main groups of yeasts (73 images), filamentous fungi (341 images), and direct fluorescent microscopy from clinical samples (60 images). The accuracy of the assessments varied, with an average correct decision rate between 78 and 93% across the three image groups. Individual participant's performance showed a mean accuracy rate ranging between 76 and 92%. A significant difference was observed in the assessment accuracy across specimen groups and among individual participants (p < 0.05). However, there was no significant interaction effect between participants and image group (p = 0.118). In conclusion, telemycology offers a promising alternative to standard microscopy diagnostics of fungal infections, especially in settings where skilled mycologists are lacking, including low- and middle-income countries.