Breakthrough in Parkinson's Disease Diagnosis and Therapy Using AI and Optogenetics in Mice

Recent advancements in neuroscience have led to a groundbreaking approach for diagnosing and treating Parkinson's disease, employing the combined power of artificial intelligence (AI) and optogenetics. Researchers at the Korea Advanced Institute of Science and Technology (KAIST) have demonstrated that this integrated method offers unprecedented precision in early diagnosis and personalized treatment, at least in mouse models.

Parkinson's disease, a neurodegenerative disorder affecting millions worldwide, is characterized by motor symptoms such as tremors, rigidity, slowed movements, and postural instability. Traditional diagnostic methods often struggle to detect the disease early or accurately assess its severity, limiting effective intervention. Existing treatments mainly alleviate symptoms rather than addressing the underlying causes.

In a collaborative effort, teams led by Professor Won Do Heo, Professor Daesoo Kim, and Director Chang-Jun Lee developed a preclinical model mimicking human Parkinson's disease. They induced two severity stages in male mice expressing alpha-synuclein protein abnormalities, a hallmark of Parkinson's pathology. Using AI-based 3D pose estimation, the team analyzed over 340 behavioral features including gait, limb movements, and tremors, consolidating these into a single metric called the AI-predicted Parkinson's disease score (APS).

Remarkably, the APS distinguished diseased mice from healthy controls within just two weeks of disease induction. It proved to be more sensitive than traditional motor tests, capturing subtle behavioral shifts like stride alterations, limb asymmetry, and tremors. To ensure the specificity of these behavioral markers to Parkinson's, the team also applied the same analysis to mice with Amyotrophic Lateral Sclerosis (ALS). The results showed that despite motor decline, ALS models did not yield high APS scores, confirming that the metric is specific to Parkinson's disease features.

Beyond diagnosis, the researchers explored therapeutic avenues using optoRET, an optogenetics technique that precisely modulates nerve signals with light. Applying this method to the Parkinson's mouse model, they observed improvements in gait, limb control, and a reduction in tremors. The most effective treatment involved shining light on alternate days, which also showed potential in protecting dopamine-producing neurons.

Professor Won Do Heo emphasized the significance of this research, stating, "This is the first preclinical framework that links early diagnosis, treatment evaluation, and disease mechanism analysis for Parkinson's, combining AI behavioral analysis with optogenetics. It paves the way for future personalized medicine approaches."

This innovative research bridges the gap between early detection and targeted therapy, offering hope for more precise and effective interventions for Parkinson's disease in humans in the future.

Source: https://medicalxpress.com/news/2025-09-ai-optogenetics-enable-precise-parkinson.html