By: Giselle C. Matlis, Research Assistant
As artificial intelligence (AI) and machine learning (ML) become more popular in the healthcare and medical device industry, their use could irrevocably change the way data is processed and lead to a greater understanding of medical information. Although there are many benefits, this technology could also have unintended consequences.
To this end, FDA along with the United Kingdom’s Medicines and Healthcare products Regulatory Agency (MHRA) and Health Canada released 10 guiding principles for good machine learning practice for medical device development.
These 10 “commandments” of AI will promote safe, successful, and high caliber use of these tools in the medical device world.
The Ten Regulatory Commandments of AI are:
- Thou Shall Ensure That Multi-Disciplinary Expertise Is Leveraged Throughout the Total Product Life Cycle
- Thou Shall Ensure That Good Software Engineering and Security Practices Are Implemented
- Thou Shall Ensure That Clinical Study Participants and Data Sets Are Representative of the Intended Patient Population
- Thou Shall Ensure That Training Data Sets Are Independent of Test Sets
- Thou Shall Ensure That Selected Reference Datasets Are Based Upon Best Available Methods
- Thou Shall Ensure That Model Design Is Tailored to the Available Data and Reflects the Intended Use of the Device
- Thou Shall Ensure That Focus Is Placed on the Performance of the Human-AI Team
- Thou Shall Ensure That Testing Demonstrates Device Performance during Clinically Relevant Conditions
- Thou Shall Ensure That Users Are Provided Clear, Essential Information
- Thou Shall Ensure That Deployed Models Are Monitored for Performance and Re-training Risks are Managed
This collaborative guideline will assist developers in making the right choices when employing these unique technologies in the medical device field. Furthermore, they will encourage the use and growth of these remarkable tools in the medical device world to change it for the better.