FACES: A Deep-Learning-Based Parametric Model to Improve Rosacea Diagnoses

Rosacea is a chronic inflammatory skin disorder that causes visible blood vessels and redness on the nose, chin, cheeks, forehead. However, visual assessment, current standard method used to identify rosacea, often subjective among clinicians results in high variation. Recent advances artificial intelligence have allowed for effective detection of various diseases with accuracy consistency. In this study, we develop new methodology, coined “five accurate CNNs-based evaluation system (FACES)”, classify rosacea more efficiently. First, 19 CNN-based models been widely image classification were trained tested via training validation data sets. Next, five best performing selected based accuracy, which served as weight value FACES. At same time, also applied majority rule detect rosacea. The exhibited performance FACES was superior individual terms sensitivity, specificity, precision. particular, sensitivity highest, specificity precision higher than most models. To improve our system, future studies must consider patient details, such age, gender, race, perform comparison tests between model clinicians.