Odum Library
dc.contributor.author | Bettis, Alan M. | |
dc.date.accessioned | 2020-04-19T20:15:50Z | |
dc.date.available | 2020-04-19T20:15:50Z | |
dc.date.issued | 2020 | |
dc.identifier.uri | https://hdl.handle.net/10428/4169 | |
dc.description.abstract | The human hand is complex, with many possible configurations (poses) spanning 27 degrees of freedom. This complexity is further increased by diseases such as rheumatoid arthritis (RA), which can cause deformities that lead to disease-specific poses outside of the normal configuration space. In this project, we use machine learning to map the manifold of natural human hand poses. We use the Leap Motion, a human-computer interaction (HCI) device, which uses infrared light to detect hand pose. Using dimensionality reduction techniques such as t-distributed stochastic neighbor embedding (t-SNE), the healthy human hand pose is visualized as a point in a low-dimensional space. This low-dimensional representation of hand poses can then be used as input to off-the-shelf machine learning algorithms, to identify poses that could predict RA or other musculoskeletal diseases of the hand. | en_US |
dc.language.iso | en_US | en_US |
dc.title | On Human Hand Configurations | en_US |
dc.type | Presentation | en_US |