AI-based analysis and severity evaluation of Parkinson’s illness. Credit score: Nature Communications (2025). DOI: 10.1038/s41467-025-63025-w
Globally acknowledged figures Muhammad Ali and Michael J. Fox have lengthy suffered from Parkinson’s illness. The illness presents a posh set of motor signs, together with tremors, rigidity, bradykinesia, and postural instability. Nevertheless, conventional diagnostic strategies have struggled to sensitively detect adjustments within the early levels, and medicines focusing on mind sign regulation have had restricted medical effectiveness.
Just lately, Korean researchers efficiently demonstrated the potential of a expertise that integrates AI and optogenetics as a software for exact analysis and therapeutic analysis of Parkinson’s illness in mice. They’ve additionally proposed a method for creating next-generation customized remedies.
A collaborative analysis group, comprising Professor Gained Do Heo’s group from the Division of Organic Sciences, Professor Daesoo Kim’s group from the Division of Mind and Cognitive Sciences, and Director Chang-Jun Lee’s group from the Institute for Primary Science (IBS) Heart for Cognition and Sociality, achieved a preclinical analysis breakthrough by combining AI evaluation with optogenetics.
Their work concurrently demonstrated the potential of early and exact analysis and remedy in an animal mannequin of Parkinson’s illness. The examine, with Dr. Bobae Hyeon, a postdoctoral researcher on the KAIST Institute for Organic Science, as the primary writer, is printed within the journal Nature Communications.
The analysis group created a Parkinson’s illness mouse mannequin with two levels of severity. These had been male mice with alpha-synuclein protein abnormalities, an ordinary mannequin used to simulate human Parkinson’s illness for diagnostic and therapeutic analysis.
In collaboration with Professor Kim’s group at KAIST, they launched AI-based 3D pose estimation for behavioral evaluation. The group analyzed greater than 340 behavioral options—equivalent to gait, limb actions, and tremors—from the Parkinson’s mice and condensed them right into a single metric: the AI-predicted Parkinson’s illness rating (APS).
The evaluation confirmed that the APS exhibited a major distinction from the management group as early as two weeks after the illness was induced. It additionally proved extra delicate in assessing the illness’s severity than conventional motor operate checks. The examine recognized key diagnostic options, together with adjustments in stride, asymmetrical limb actions, and chest tremors. The highest 20 behavioral options included hand/foot asymmetry, adjustments in stride and posture, and a rise in high-frequency chest motion.
To verify that these behavioral indicators weren’t simply basic motor decline however particular to Parkinson’s, the group utilized the identical evaluation to a mouse mannequin of Amyotrophic Lateral Sclerosis (ALS), also referred to as Lou Gehrig’s illness, in collaboration with Director Lee’s group at IBS. Since each Parkinson’s and ALS trigger motor operate issues, if the APS merely mirrored poor motor abilities, a excessive rating ought to have appeared in each ailments.
Nevertheless, the evaluation of the ALS animal mannequin confirmed that regardless of a decline in motor operate, the mice didn’t exhibit the excessive APS seen within the Parkinson’s mannequin. As a substitute, their scores remained low, and their behavioral adjustments had been distinctly completely different. This demonstrates that APS is immediately associated to particular, attribute adjustments that solely seem in Parkinson’s illness.
For remedy, the analysis group used optoRET, an optogenetics expertise that exactly controls neurotrophic alerts with gentle. This method proved efficient within the animal mannequin, resulting in smoother gait and limb actions and a discount in tremors.
Particularly, a routine of shining gentle on alternate days was discovered to be the simplest, and it additionally confirmed a bent to guard dopamine-producing neurons within the mind.
Professor Gained Do Heo of KAIST acknowledged, “This is the first time in the world that a preclinical framework has been implemented that connects early diagnosis, treatment evaluation, and mechanism verification of Parkinson’s disease by combining AI-based behavioral analysis with optogenetics. This lays a crucial foundation for future personalized medicine and customized treatments for patients.”
Extra data:
Bobae Hyeon et al, Integrating synthetic intelligence and optogenetics for Parkinson’s illness analysis and therapeutics in male mice, Nature Communications (2025). DOI: 10.1038/s41467-025-63025-w
Offered by
The Korea Superior Institute of Science and Expertise (KAIST)
Quotation:
AI and optogenetics allow exact Parkinson’s analysis and remedy in mice (2025, September 26)
retrieved 27 September 2025
from https://medicalxpress.com/information/2025-09-ai-optogenetics-enable-precise-parkinson.html
This doc is topic to copyright. Aside from any truthful dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is supplied for data functions solely.
