• Paper

    Hiroko Yamano, Assistant Professor

A research paper from Asst. Prof. Hiroko Yamano’s group published at “PLoS ONE”

The human musculoskeletal network is a complex system consisting of diverse components and their interactions. Although anatomical knowledge of individual elements and their connections is extensive, the interdependence between these elements as a system remains largely unexplored. The links between bones and muscles are not limited to the one-to-one (pairwise) connections used as typical of network analysis but also include many-to-many connections, where one bone can connect to several muscles and vice versa.

This study uses hypergraphs, which can represent multiple connections, to model the musculoskeletal system. Both hypergraph and pairwise models were used to learn the embedding representations of muscle and bone connections. A comparison of these models showed that the hypergraph-based approach better distinguished the roles of muscles connecting different parts of the body, outperforming the pairwise method.

Research title:


Hypergraph modeling of complex interactions: Applications from human musculoskeletal structures to complex system dynamics

Joint Researchers:


Yamano H, Liu S, Toriumi F

Journal:


PLoS ONE 19(11): e0310189.
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0310189

Contact:


Hiroko Yamano
Assistant Professor, Institute for Future Initiatives