Parkinson's disease classification and prediction via adaptive sparse learning from multiple modalities
Abstract: Highlights•Manifold learning and sparse regression are utilized to learn low-dimensional structures for feature selection.•Similarity matrices among samples and features can be dynamically updated to improve the performance of the model.•Experiments on the Parkinson's Progression Marker Initiative database demonstrate that our method outperforms others.
External IDs:dblp:journals/bspc/HuangLYWCYSZG25
Loading