Neofect Succeeds in Developing 'AI for Stroke FMA Prediction'... Culmination of 3 Years in Government Pan-Ministerial R&D

Neofect, a digital healthcare company specializing in artificial intelligence (AI) and robotics, has achieved results in the 'Government Pan-Ministerial Medical Device Research and Development (R&D)' project.

On January 29, Neofect announced that it has developed an 'AI-based Clinical Functional Assessment Prediction Model for Stroke Patients.'

The government’s pan-ministerial medical device R&D project is a national initiative aimed at revitalizing the healthcare industry and securing leading-edge medical device technologies for the future. Through the development of advanced medical devices, the project seeks to foster the domestic medical device industry and strengthen health security response capabilities.

This development was carried out jointly with a research team led by Professor Jang Wonhyuk of the Department of Rehabilitation Medicine at Samsung Medical Center over approximately three years, from April 1, 2023, to the end of last year.

The predictive AI was trained on more than 2,700 real-world clinical data cases collected from institutions such as the National Rehabilitation Center and Samsung Medical Center. This model serves as a core algorithm for advancing the Fugl-Meyer Assessment (FMA), a global standard indicator for evaluating motor function in stroke patients.

While the traditional FMA relied on therapists 'observing' patients' movements for assessment, this new technology collects and analyzes quantitative data-such as movement distance, smoothness, and angles-through wearable devices. This approach reduces subjective judgment and evaluation variability among therapists, implementing a more standardized assessment system. The technology is currently in the confirmatory clinical stage, and the company aims to commercialize it within the first half of this year, after clinical completion.

With this achievement, Neofect is expected to accelerate its transformation from a rehabilitation medical device company to an AI digital rehabilitation company based on clinical data. The company cited 'Clinic' as the cornerstone of this model’s development. For the past decade, Neofect has been operating Clinic, consistently securing, accumulating, and training on longitudinal clinical data related to rehabilitation medical device training and assessment.

Clinic has long accumulated real-world medical databases (DB) from hospitals such as Stanford Children's Health, Boston Children's Hospital, Spaulding Rehabilitation Hospital, and Barrow Neurological Institute. Domestic hospitals are also building their own databases using Clinic in accordance with relevant regulations.

The company explained that as the database grows, the algorithm becomes more sophisticated, and its utility in clinical settings increases. As prediction accuracy and standardization improve, hospitals have greater justification for adoption and operation, which in turn leads to more usage data-reinforcing a virtuous cycle of 'data accumulation → performance advancement → broader adoption.' The rapid evolution of AI algorithm technology has also accelerated development speed.

Going forward, Neofect plans to integrate AI-based assessment technology into its hardware products and provide subscription-based (SaaS) services and data-driven software for hospitals.

A company representative stated, "In the long term, we plan to build an AI-based digital rehabilitation platform that integrates patient information, training data, and treatment outcomes, and transition to a subscription economy model centered on software and data-driven recurring revenue. Through this, we aim to drive innovation in the silver care market by combining smart technologies such as AI and robotics into care services."

This content was produced with the assistance of AI translation services.