AI predicts contralateral knee osteoarthritis
Researchers from Seoul National University Hospital (SNUH) have developed an AI model for predicting the likelihood that patients with unilateral osteoarthritis would develop the same condition in their other knee.
They trained a machine learning-based algorithm on two large-scale arthritis tracking cohort datasets in the United States. It considered nine variables: gender, ethnicity, body mass index, severity of arthritis in both the arthritic and contralateral knees, pain and function index of both arthritic and contralateral knees, lateral joint space narrowing of the contralateral knee, and history of meniscectomy surgery of the contralateral knee.
The model was then used to analyse approximately 1,353 cases of unilateral osteoarthritis that were followed for four to five years from the arthritis datasets.
A major finding of the study, published in Wiley’s Journal of Orthopaedic Research, was that a reduction in the lateral joint space of the contralateral knee is a major risk factor for osteoarthritis of the contralateral knee, increasing the risk by 4.5 times.
The study also demonstrated for the first time that the severity of arthritis and the pain and function index of the unilateral knee arthritis are greatly associated with the development of osteoarthritis in the opposite knee.
Additionally, the model demonstrated 69% accuracy in identifying patients with a high risk of developing contralateral knee arthritis.
“This study is the first to develop a machine learning model that predicts the possibility of developing contralateral knee osteoarthritis. It is expected to help develop personalised treatment plans that comprehensively consider both knees in the future,” said study lead Du Hyun Ro, who is also a professor of Orthopaedic Surgery at SNUH.
Online resource on experimental traditional meds data launched
The South Korean Ministry of Health and Welfare (MOHW) has recently piloted a new service providing online access to experimental data on traditional Korean medicine.
The pilot service launched on the website of the Korea Institute of Oriental Medicine seeks to consolidate scattered experimental data on traditional Korean medicine and make it available to the public, especially to researchers.
Following this pilot phase, it is planned for the service to provide detailed information on herbal medicines and introduce a network pharmacology-based analysis feature by next year.
Meanwhile, the MOHW is also exploring the development of an AI model to predict the efficacy of herbal medicines and an interactive chatbot service.
Seoul’s Gangdong district to offer digital chronic disease management service
Kakao Healthcare, the digital health unit of technology conglomerate Kakao Corporation, will be powering the delivery of personalised health management services in the Gangdong district of Seoul, capital of South Korea.
The company entered into a partnership with the Gangdong district government to establish a data and AI-driven chronic disease management system. They will also conduct research to enhance Kakao Healthcare’s digital diabetes application, as well as promote awareness and education around blood sugar management.
The partnership is reportedly the first of its kind for any local government unit in the country.
