The AI-based application can diagnose 12 high level dermatological disorders, further classifiable into 90 secondary conditions, with high accuracy.
Artificial intelligence (AI) is actively being explored as a potential tool in the field of Dermatology, to diagnose and provide accurate cure to skin disorders, including hair and nails. D. Y. Patil University, Department of Dermatology, in association with California-based Polyfins Technology Inc, is evaluating the efficiency of AI tools in an OPD clinic for diagnosing skin and hair problems in India. The project is being headed by Dr Sharmila Patil, in an effort to increase early-detection of skin diseases and to effectively increase clinician capacity for treatments.
The Intelligent Solution
Along with Polyfins Technology, the team has formulated an AI-enabled application called Tibot.ai. Tibot uses deep learning technology and machine learning tools to predict top 3 skin conditions the user may be suffering from. It will also give a confidence score for each analysis, helping users to determine more accurately which condition or ailment requires to be diagnosed with priority.
As inputs, the app utilises pictures of the skin area and responses given by users to questions generated by the app relevant to the issue. Using deep-learning neural networks and image analytics, the program can evaluate and detect skin conditions.
Tibot tracks upto 12 high-level common skin conditions, including alopecia, bacterial skin infections, benign tumors, eczema, fungal infections, immunological skin disorders, malignant tumors, psoriasis, skin infestations, viral infections and pigmentation disorders. These conditions are further classified into 90 low level secondary classification as the app proceeds in the investigation.
The app also provides a ‘Knowledge Base’ to users, which provides context and detailed information about the condition diagnosed. It further details about the seriousness of the condition, whether to meet the physician and possible course of action to be followed.
AI apps like Tibot leverage the medical knowledge it has been trained on by comparing the symptoms with the problems. Learning algorithms become more precise and accurate as they interact with training data, thereby allowing medical institutions to gain unprecedented insights into diagnostics.
