Domestic researchers have discovered a new possibility for the early diagnosis and treatment of Parkinson's disease using artificial intelligence (AI) and optogenetics technology. A joint research team led by Professor Kim Dae-soo of KAIST and Director Lee Chang-jun of the Institute for Basic Science (IBS) announced these research findings on the 22nd.

Parkinson's disease occurs as nerve cells in the brain that secrete 'dopamine', which controls body movements, gradually die. Symptoms include trembling hands and feet, body stiffness, and abnormal gait. However, until now, diagnosis was only possible after the disease had progressed to some extent, and drug treatments to regulate brain signals were also limited in effectiveness.

The research team analyzed behavior using AI-based three-dimensional (3D) posture estimation technology on a Parkinson's disease mouse model induced by alpha-synuclein protein aggregation, known as a cause of Parkinson's disease. They created a 'Parkinson's Behavior Index' by analyzing over 340 behavioral signals such as gait, hand and foot movements, and tremors using AI. This index was used to more sensitively determine the degree of the disease from the early onset of Parkinson's disease compared to existing motor ability tests.

As a result, the Parkinson's Behavior Index showed a significant difference from the control group two weeks after disease induction. In particular, changes in stride length, asymmetry in hand and foot movements, and chest tremors were identified as key factors in diagnosing Parkinson's disease.

Additionally, the research team applied optogenetics technology for treatment. This involved using light to precisely control the activity of brain nerve cells. Experiments exposing mice to light under certain conditions confirmed that the gait of mice with Parkinson's disease became smoother and tremor symptoms were reduced. The 'alternate-day cycle' method, where light was applied every other day, was found to be the most effective, and it was also shown to protect dopamine-producing nerve cells.

KAIST Distinguished Professor Heo Won-do stated, “By combining AI-based behavior analysis and optogenetics, we achieved results that connect early diagnosis to treatment of Parkinson's disease,” adding, “This will aid in the development of treatments tailored to individual patients.”

These research results were published in the online edition of the world-renowned scientific journal 'Nature Communications' on the 21st of last month.

Popular News