Can Artificial Intelligence Support Patient Education In Scabies? A Comparative Analysis Of Large Language Model Responses

This study evaluated the accuracy, reliability, quality, and readability of AI chatbot responses to scabies-related questions using three models: ChatGPT-5.2, DeepSeek, and Claude Sonnet 4.5. ChatGPT-5.2 excelled in information quality and readability, while DeepSeek achieved the highest global quality and accuracy scores but had higher hallucination rates. Claude Sonnet 4.5 scored lower across most metrics. The findings indicate variability in the quality of responses among the models, emphasizing that AI chatbots should be used as supportive resources rather than replacements for professional medical advice.