Software as a Medical Device (SaMD)
What Is SaMD?
Software as a Medical Device (SaMD) is software intended to be used for one or more medical purposes that performs these purposes without being part of a hardware medical device. Examples include:
- AI-powered diagnostic decision support software
- Software that analyses medical images for diagnostic purposes
- Clinical calculators used to determine treatment dosing
- Software that monitors physiological parameters and alerts clinicians
- Software used to plan surgical procedures
Software that is an integral part of a hardware device (device software / embedded software) is regulated as part of that hardware device, not as a separate SaMD.
Current Regulatory Status in Indonesia
Indonesia does not yet have a dedicated, published SaMD-specific regulation. SaMD is currently regulated under the existing medical device framework (Permenkes 62/2017) on a case-by-case basis, applying general classification rules to determine whether a particular software product is a medical device and, if so, what class it falls into.
The Indonesian SaMD regulatory framework is actively developing. Manufacturers bringing AI/ML-based or complex SaMD products to Indonesia should engage with their LAR and directly with Kemenkes for the current position on specific products.
SaMD Classification Principles
In the absence of Indonesia-specific SaMD guidance, Kemenkes applies general classification criteria to SaMD, informed by:
| Reference | Relevance |
|---|---|
| IMDRF SaMD N10:2013 | Definition and categorisation of SaMD |
| IMDRF SaMD N23:2015 | Quality management system for SaMD |
| IMDRF SaMD N41:2017 | Clinical evaluation for SaMD |
| GHTF/SG5/N6:2012 | Clinical evidence for medical devices |
| ASEAN AMDD classification rules | Applied by analogy for ASEAN-market SaMD submissions |
Key classification question for SaMD:
- What is the intended clinical purpose?
- What is the severity of the condition the software is intended to address?
- What is the consequence of an incorrect output from the software?
Higher-severity intended use combined with high consequence of error → higher class.
Practical Considerations for SaMD Registration in Indonesia
- Engage early: Contact your LAR and potentially Kemenkes directly to seek informal guidance on classification before preparing a full dossier
- Document the software's medical purpose precisely: Ambiguous intended use descriptions are a common source of SaMD registration queries
- IEC 62304 compliance: Software lifecycle documentation per IEC 62304 is expected for SaMD
- Cybersecurity: For internet-connected SaMD, a cybersecurity risk assessment is increasingly expected
- AI/ML: For continuously learning or adaptive AI algorithms, document how algorithm updates will be managed and whether updates trigger re-registration