# RehabGen's AI Personalization Enhances Patient Engagement in Rehabilitation

> RehabGen transforms everyday objects into rehabilitation tools, providing personalized exercise recommendations for patients recovering from upper-limb injuries. This generative AI system bridges the gap between clinical instructions and home-based recovery, enhancing patient independence and engagement.

**Source**: news.engineering.asu.edu | **Published**: 2026-11-14 | **Type**: research

## Key Facts

- RehabGen's AI-driven personalization could disrupt traditional rehab, enhancing patient engagement.
- 57.6% expert preference for RehabGen suggests competitive edge over static rehab solutions.
- Trust in AI recommendations may lead to increased patient adherence, improving recovery rates.
- Future scalability hinges on addressing privacy, regulatory, and safety concerns in AI applications.
- Collaboration with clinicians ensures RehabGen's relevance, balancing innovation with professional oversight.

## Summary

A collaboration involving Arizona State University (ASU), the Korea Advanced Institute of Science and Technology (KAIST), and Columbia University has led to the development of RehabGen, a generative AI system designed to personalize rehabilitation exercises for patients recovering from upper-limb injuries. By leveraging patient-specific data, including symptoms, goals, and images of their home environments, RehabGen aims to create tailored exercise recommendations that incorporate everyday objects. This innovation is significant as it addresses the gap in rehabilitation support that patients often face when transitioning from clinical settings to home exercises.

The need for such a system stems from the challenges patients encounter when completing rehabilitation independently. After receiving an exercise plan from a therapist, patients may struggle with remembering instructions or adapting exercises to their available resources. Traditional printed instructions and static exercise videos do not account for the dynamic nature of a patient's recovery journey, which can involve changing symptoms and abilities. RehabGen seeks to fill this void by analyzing both clinical assessments and the patient's surroundings to suggest practical exercises that utilize common household items.

Heejin Jeong, an assistant professor at ASU, played a pivotal role in defining the parameters that RehabGen must consider to make effective recommendations. The system's design reflects insights gained from direct interactions with rehabilitation professionals and patients, ensuring that it accounts for individual capabilities and available equipment. For instance, if a patient with wrist mobility issues shares an image of their kitchen table, RehabGen could identify a water bottle and propose a series of exercises that involve reaching, grasping, and lifting it. This contextualized approach aims to enhance the relevance and applicability of rehabilitation exercises.

Despite its promise, RehabGen is currently a research prototype and should not be viewed as a standalone medical tool. Rehabilitation professionals must still evaluate the appropriateness of the suggested activities, including specific movements and resistance levels. The involvement of clinicians throughout the development process has been crucial. Their feedback helped refine the system, ensuring that it aligns with real-world rehabilitation practices and safety standards. 

The collaborative effort has produced a model that has shown promise in initial evaluations. In a study involving 23 rehabilitation experts, the AI-generated recommendations were preferred over baseline systems in over half of the comparisons, particularly for their personalization and clarity regarding exercise purposes. Patient feedback also indicated increased trust and perceived usefulness of the system. However, the current iteration does not monitor movement quality or account for pain and fatigue, raising concerns about its readiness for routine clinical use.

Looking ahead, the potential for RehabGen to transform home rehabilitation is substantial. As the system undergoes further testing with diverse populations and develops protocols for clinician oversight, it could significantly enhance patient engagement and recovery outcomes. The integration of AI in rehabilitation could lead to more personalized and practical home exercise programs, ultimately fostering a collaborative environment among patients, clinicians, and intelligent systems. 

As the field of digital health continues to evolve, the successful implementation of RehabGen could signify a shift toward more individualized care models in rehabilitation. This aligns with broader trends in healthcare that prioritize patient-centered approaches and leverage technology to enhance treatment efficacy. The ongoing research into AI’s role in monitoring recovery and engagement will further refine these systems, potentially setting new standards for rehabilitation practices.

## Entities

- **Products**: RehabGen
- **Technologies**: generative AI, vision-language model
- **People**: Heejin Jeong, Youjin Sung, Cindy Ivy
- **Organizations**: Arizona State University, Korea Advanced Institute of Science and Technology, Columbia University, Northern Arizona University

## Key Concepts

personalized rehabilitation, home rehabilitation, AI in healthcare, expert-in-the-loop development, patient-centered care, rehabilitation professionals, clinical assessments, exercise recommendations

## Definitions

- **RehabGen**: A generative AI system designed to provide personalized rehabilitation exercise recommendations based on a patient's symptoms, goals, and environment.
- **expert-in-the-loop development**: A development process that involves rehabilitation professionals in defining requirements and validating system outputs to ensure safety and feasibility.
- **vision-language model**: An AI model that integrates visual and textual information to understand and generate recommendations based on images and descriptions.
- **generative AI**: A type of artificial intelligence that can generate new content or recommendations based on input data.
- **patient-centered care**: An approach to healthcare that emphasizes the needs, preferences, and values of patients in their treatment.

## Use Cases

- Providing personalized exercise recommendations using everyday objects
- Supporting rehabilitation professionals in developing home exercise programs
- Enhancing patient engagement and trust in rehabilitation
- Facilitating remote rehabilitation monitoring
- Improving the quality of rehabilitation exercises through AI analysis

## Frequently Asked Questions

**What is RehabGen?**

RehabGen is a generative AI system that offers personalized rehabilitation exercise recommendations based on a patient's specific circumstances, including their environment and clinical assessments.

**How does RehabGen work?**

RehabGen analyzes images of a patient's surroundings and combines this with their symptoms and goals to suggest suitable rehabilitation exercises using everyday objects.

**Who developed RehabGen?**

RehabGen was developed by researchers from Arizona State University, the Korea Advanced Institute of Science and Technology, and Columbia University, with contributions from rehabilitation professionals.

**Can RehabGen replace a therapist?**

No, RehabGen is intended to support rehabilitation professionals by providing exercise suggestions, but it should not replace professional judgment or oversight in patient care.

**What are the limitations of RehabGen?**

RehabGen currently cannot monitor movement quality or assess patient pain and fatigue, and it requires further testing to ensure safety and effectiveness in diverse populations.

## Links

- [Read on Welcome.AI](https://welcome.ai/content/rehabgens-ai-personalization-enhances-patient-engagement-in-rehabilitation)
- [Original source](https://news.engineering.asu.edu/2026/11/bringing-rehabilitation-exercises-home/)
- [RehabGen](https://welcome.ai/company/rehabgen): Featured company

---

Source: Welcome.AI | https://welcome.ai/content/rehabgens-ai-personalization-enhances-patient-engagement-in-rehabilitation