Artificial intelligence (AI) and machine learning continue to show their potential in various industries and applications. Recently they have proven up to the challenge of transforming and disrupting healthcare, but there are significant challenges to the adoption of these technologies. MedTech DIVE brought this interesting topic to our attention in their article, “FDA AI-machine learning strategy remains work in progress.”
Among those hurdles are the lack of large, high quality and well-curated data sets and the inability to explain “black box” approaches, as well as social biases in the data that do not benefit care for all patients and could exacerbate health disparities, to name a few.
The Federal Department of Agriculture (FDA) has been exploring what good practices look like for algorithm design and development, not to mention testing and training. The agency is considering a total product life cycle-based approach to regulating medical devices that leverages self-updating algorithms.
For the best results with machine learning, the data sets must be of the right quality for training. More data doesn’t always mean better.
Melody K. Smith
Sponsored by Data Harmony, a unit of Access Innovations, the world leader in indexing and making content findable.