Publication

Abstract : The psychological well-being of individuals with vitiligo, a persistent skin illness marked by prominent white patches on the skin and a lack of pigmentation, is severely compromised. Conventional methods for classifying the quality of life impairment and predicting the psychological effects in patients with vitiligo often fall short of capturing the complex, non-linear connections seen in the data. This paper explores the use of deep learning models to overcome these limitations. We developed neural network architectures that are used to predict Depression, Anxiety, and Stress Scale (DASS) scores using regression models and classify Dermatology Life Quality Index (DLQI) scores using classification models. DLQI scores are divided into three impact categories by the categorization model: “Low,” “Moderate,” and ‘Migh.” Continuous DASS scores are predicted using the regression model. Our results demonstrate that deep learning models can effectively capture complicated patterns in the data, achieving above 98% accuracy in classification and a MSE of 1.6 in prediction. Both classification performance and prediction accuracy are much higher with this than with traditional methods. Deep learning architectures therefore provide a solid basis for understanding and treating the psychological impacts of vitiligo and may lead to enhanced patient management and targeted therapy.