Skip to main content

Ai Ml Devices

Official sources: CDSCO; ABDM — NHA; IMDRF

Overview​

AI/ML-based medical devices represent a growing category under CDSCO regulation. These devices use artificial intelligence or machine learning algorithms for diagnosis, prognosis, monitoring, or treatment recommendations.

Current Regulatory Status​

As of 2026, CDSCO classifies AI/ML devices under the existing Medical Devices Rules 2017 framework. Specific AI/ML guidance is under development but not yet published. Manufacturers should reference IMDRF guidance on AI/ML-based Software as a Medical Device (SaMD).

Key Regulatory Considerations​

  • Algorithm transparency and validation: Manufacturers must demonstrate algorithm performance, including sensitivity, specificity, and clinical validation across diverse datasets.
  • Risk-based classification: AI/ML devices are classified as Class A, B, C, or D based on risk and intended use.
  • Post-market surveillance: Enhanced monitoring required for devices using adaptive or learning algorithms.
  • Cybersecurity: AI/ML devices must address data security, model integrity, and protection against adversarial attacks.