8/27/2024

We are pleased to announce publication of the HL7 Informative Document: AI/ML Data Lifecycle Edition 1 – US Realm on 27 August 2024.

Under sponsorship from the US Department of Veterans Affairs, as the project facilitator and Editor-in-Chief, our Founder, Dr. Mark Janczewski, was honored to have worked with a team of dozens of volunteer clinical and administrative SMEs and software developers for over a year on this vital project.

This is the first ever specification from Health Level Seven (www.HL7.org) addressing Artificial Intelligence (AI) in healthcare systems and was conducted under the auspices of the AI Focus Team (AI FT), part of the Electronic Health Record (EHR) Workgroup (WG) within HL7. As the world’s largest and ANSI-accredited health IT standards development organization, HL7 is responsible for interoperability standards for healthcare across paradigms: messaging, clinical documents, Clinical Decision Support (CDS) tools, and application programming interfaces (APIs). Over the past 40 years, HL7, along with other complementary standards (e.g., DICOM, NCPDP, SNOMED, ICD, CPT), has introduced patterns of encoding, knowledge representation, and standards-based coding to health information.

AI is a rapidly evolving technology that presents significant potential to improve healthcare delivery yet needs to have standards to minimize potential harm.  For HL7, this is a new area of activity. This Publication includes recommendations and guidance to software developers to promote the use of standards to improve the trust and quality of interoperable data used in AI models.  Standards are needed for the development and implementation of AI systems in healthcare to ensure that the data used to train and receive output from these systems are of consistently high quality, interoperable (uses data that involve standard terminologies), transparent, and ethically sound, and used for the purpose intended (e.g. "answers the question").