Skip to main content

Digital medicine and the curse of dimensionality


December 14, 2021

Discussion Topic

A patient’s health state can be characterized by a multitude of signals from many different data modalities. This high-dimensional, personalized data stream aggregated over patients’ lives has spurred interest in developing new clinical AI models. One of the rate-limiting factors in developing AI models that generalize to real-world scenarios is the very attribute that makes the data exciting—their high-dimensional nature.

At DiMe’s #AskMeAnything Journal, author Visar Berisha, PhD led a discussion on how “the curse of dimensionality” can doom models to failure, even when they seem to work well during development. We explored the key highlights of his Nature publication “Digital medicine and the curse of dimensionality“, Visar also provided some suggestions on how to develop clinical AI models that are more likely to fare well during prospective validation.


Read the article

Listen to the discussion

View the presentation slides


Join our next project

Help streamline the path to regulatory and commercial success to optimize health outcomes for the greatest number of patients

Join the Integrated Evidence Plans project

Join us
Not today