AI-Enhanced Personalization in Digital Therapeutics: A Framework Prioritizing Patient Safety

1

Digital therapeutics (DTx) provide evidence-based software interventions for various health conditions, including ADHD and chronic pain.
2

AI-driven personalization in DTx poses clinical challenges, particularly regarding patient safety and the unpredictability of complex AI systems.
3

The SAFE_DTx framework prioritizes intermediate feedback signals to optimize DTx while maintaining explicit medical safety constraints.
4

SAFE_DTx separates clinical surrogate prediction from predefined safety-constraint filtering to enhance patient safety in DTx.
5

Current practices favor AI-enabled therapies with human oversight rather than fully autonomous DTx to mitigate safety risks.

1