![](https://contents-cdn.viewus.co.kr/image/2024/07/CP-2023-0193/image-4132b42b-d56b-42f1-a5d5-a9b725d43085.jpeg)
Researchers have published findings on a machine-learning model that can classify subgroups of Parkinson’s disease based on the speed of disease progression. The study appeared in npj Digital Medicine.
Parkinson’s disease, known as the second most common degenerative brain disorder after dementia, is caused by a deficiency of the neurotransmitter dopamine. It primarily manifests through symptoms such as tremors, muscle rigidity, and slowed movements. As motor dysfunction worsens, patients may struggle to walk normally.
Parkinson’s disease progresses slowly, and its symptoms often closely mimic those of normal aging, which makes early diagnosis difficult. Moreover, it can lead to several complications, including sleep disorders, depression, and urinary issues.
A research team from Cornell University analyzed data from 406 participants in an international observational study. They developed a deep learning model capable of comprehensively modeling the participants’ multidimensional, longitudinal progression data.
The researchers also emphasized that patients with Parkinson’s disease report a wide range of progression and symptoms, necessitating personalized treatment approaches.
Using machine learning, the researchers identified three subgroups of Parkinson’s disease based on the speed of disease progression. They suggested that this classification could lead to more targeted clinical approaches and effective treatments.
Most Commented