The All-India Institute of Medical Sciences (AIIMS) is collaborating with Google to develop a groundbreaking artificial intelligence (AI)-based dermatology app, aiming to make skin healthcare more accessible and inclusive across India. Expected to launch next year, the app—named Indus Derma—is designed to help both patients and non-specialist healthcare providers identify and manage common skin conditions using AI-powered image analysis.
How the App Works
Users can simply take a photo of a skin rash or discoloration and upload it to the app, providing basic details such as age, gender, symptom duration, and whether the area is itchy or painful. Advanced computer vision models, integrated with large vision-language AI systems, will analyze the image, compare it against extensive dermatology databases, and suggest a likely diagnosis.
The app will also recommend the next steps, including whether urgent referral to a dermatologist is needed, whether the condition can be managed locally, or if further investigations are required. In addition, it will provide general treatment advice, lifestyle recommendations, and preventive skincare guidance.
Dr. Somesh Gupta, professor of dermatology at AIIMS, emphasizes, “This is a decision support tool, not a replacement for doctors. Any medication suggestions generated by the AI must be verified by a qualified medical professional.”
Why AI Suits Dermatology
Dermatology relies heavily on visual pattern recognition, making it particularly compatible with AI technology. “Most skin conditions are diagnosed based on appearance. AI models, especially computer vision systems, perform exceptionally well in this domain,” Dr. Gupta explains. Even general-purpose AI systems like ChatGPT, Gemini, and DeepSeek are reported to match dermatologists in identifying common skin conditions, while specialized models such as Google’s MedGemma can sometimes perform even better under controlled conditions.
Addressing India-Specific Challenges
A major limitation of existing global AI dermatology models is their training on predominantly white-skinned patients, reducing diagnostic accuracy for Indian users. To overcome this, AIIMS and Google are developing Indian foundation models to retrain AI systems with images of Indian skin tones and region-specific dermatological conditions. “Skin color significantly influences visual diagnosis. Local data is essential,” Dr. Gupta notes.
Accuracy and Validation
Currently under development and validation, the app aims for an accuracy rate of 80–90%, considerably higher than the approximately 40% accuracy achieved by non-specialists managing skin conditions. Regulatory approvals and extensive testing are ongoing before the public release.
Who Will Benefit
While experienced dermatologists may not rely on the app, it promises significant support for general physicians, healthcare workers, and patients in areas with limited specialist access. The AI can help users understand the severity of a condition, determine the need for urgent care, and provide guidance on basic management steps.
For patients, the app will generate easy-to-understand explanations about conditions, their causes, and general care tips. For doctors, the AI’s suggestions can be reviewed and integrated into clinical decision-making where appropriate.
Dr. Gupta anticipates the app’s release may be slightly delayed, with a realistic launch timeline set four to five months later than initially projected. “The AIIMS-Google dermatology app could revolutionize skin healthcare in India, delivering expert guidance to millions currently lacking access to specialized care,” he says.
