Dynamic Personalized Optimization: An AI Functionality Framework for Digital Therapeutics

TL;DR

This paper introduces a new AI framework called Dynamic Personalized Optimization to improve real-time, personalized digital treatments using patient data and AI models.

Contribution

The novel contribution is the DPO framework, which integrates AI functions for real-time personalization in digital therapeutics.

Findings

DPO uses predictive AI models to adapt treatments based on patient responses.

LLMs can support DPO by processing complex data formats for better personalization.

DPO addresses current limitations in real-time personalization within digital therapeutics.

Abstract

Dynamic Personalized Optimization (DPO) is introduced as a conceptual framework that defines core artificial intelligence (AI) functions required to deliver real-time, personalized, and optimized treatment in digital therapeutics (DTx). DPO continuously refines therapeutic strategies by integrating patient data, treatment content, usage feedback, and status measurements to provide real-time, personalized treatment. Using predictive AI models, DPO adapts treatment approaches based on patient responses, thereby improving therapeutic effectiveness. Furthermore, this paper explores the potential role of large language models (LLMs) in supporting DPO by processing diverse and complex data formats. By addressing current limitations in real-time personalization within DTx, DPO establishes a structured, AI-driven approach to delivering tailored digital interventions. This framework ultimately aims to enhance treatment efficacy and improve patient engagement.

Figures