Image for representative purposes only.
| Photo Credit: Getty Images/iStockphoto
The Clinical Decision Support Systems (CDSS), an AI-powered tool for standardised data capturing and consistency, is set to be deployed at various public and private hospitals.
CDSS is an initiative through the Ayushman Bharat Digital Mission (ABDM) in collaboration with the All-India Institute of Medical Sciences (AIIMS) and the Centre for Chronic Disease Control (CCDC). It provides free digital assistance to doctors for faster, evidence-based decision-making.
A Health Ministry official said the system is integrated the national telemedicine service eSanjeevani and its differential diagnosis AI model has been trained on most common symptoms and diseases observed during eSanjeevani teleconsultations such as respiratory tract infections, gastritis, fever, diabetes etc.
He said CDSS has two components – patient assistance form (PAF) at spoke level (Ayushman Arogya Mandirs) and the differential diagnosis feature at hub level (primary, secondary and tertiary care hospitals).
PAF, validated by AIIMS Rishikesh, is an intuitive feature that enables mid-level healthcare providers to record patient complaints. It is available in 13 languages, adhering to Systematized Nomenclature of Medicine – Clinical Terms international standards. The PAF, based on patient data, recommends relevant specialty departments where the patient should be referred for teleconsultation. It helps in reducing probability of errors in patient data capture with appropriate referral from spoke level.
Differential diagnosis utilises patient details captured in PAF at spoke level and recommends three differential diagnoses to doctors at hub. It reduces the probability of missed or delayed diagnosis. Recommended diagnoses are helpful for hub doctors in providing appropriate clinical decisions on treatment and patient advice.
“eSanjeevani teleconsultation data has been utilised to train the system. Since CDSS integration in April 2023 till November 2025, 282 million eSanjeevani consultations have benefited,’’ the official said.
Additionally, as part of the rollout, all States and Union Territories have been advised to update and get approved software under ABDM to enable the integration.
The official said PAF requires minimal inputs from health providers at spoke level such as symptom and its attributes. Once symptoms are selected, the system guides the health provider through sequential multiple-choice questions to choose details (duration, severity, location etc.) mitigating manual typing errors.
“The AI model analyses structured patient data that include age, gender, symptoms, attributes and suggests probable differential diagnoses to doctors. With high accuracy in its top three recommendations, it acts as a smart filter that reduces doctors’ workload. Importantly, doctors maintain complete control, they can accept or reject AI suggestions, and their feedback continuously improves the system. AI doesn’t replace clinical judgment; it is enhanced by incorporating the Human-in-the-Loop approach,’’ added the official.
Published – December 30, 2025 09:30 pm IST
