Answers are grounded in this paper and cite their sources.
Background Hand movement involves complex joint patterns that vary across tasks and individuals, and identifying the components that underlie functional hand use is essential. Objective We aimed to construct a Commonality Model that decomposes real-world finger motions into dynamic forms and quantify probabilistic transitions to develop rehabilitation systems for post-stroke patients. Methods Fifteen healthy young adults performed upper limb tasks based on a Motor Activity Log (MAL), with hand motion captured using a camera and analyzed using MediaPipe. Joint angle data were extracted, k-means clustering was used to classify the movement patterns, and silhouette analysis was performed to determine the optimal number of clusters. Euclidean distance was used to identify overlapping forms across tasks grouped into Common Forms (CFs). A first-order Markov model was constructed to calculate the transition probabilities from CFs to unique task-specific forms. Results Seven operationally defined CFs were identified as recurrent image-derived hand-joint configurations observed across multiple activities of daily living (ADL) related tasks. Some tasks were largely composed of CF-associated transitions, whereas others showed higher proportions of task-dependent Individual Forms. Conclusion The proposed Commonality Model provides an exploratory quantitative framework for identifying recurring and task-dependent hand-joint configurations during ADL-related movements in healthy young adults. Further studies are required to validate its reproducibility, generalizability, and clinical relevance in larger and clinical populations.
No takes yet. Share an insight, caveat, or question.
Suwabe et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: