Artificial Intelligence in Medicine
EEG signal classification using PCA, ICA, LDA and support vector machines
Expert Systems with Applications: An International Journal
| Hi-index | 12.05 |
Post-stroke rehabilitation has been considered vitally important for improving the life quality of stroke patients. In order to perform rehabilitation effectively, it is crucial to have a standardized system to examine patients' impairment severity in prior to any treatments and to track their training results. Brunnstrom stages classification is one of the most common measures of stroke patients' rehabilitation progress and usually can only be performed by experienced physicians. A fuzzy inference system based upper-extremity motion evaluation system is presented in this paper to provide a reliable computerized solution for objective motion quality assessment and automatic Brunnstrom stage classification. An Adaptive Neuro-Fuzzy Inference System (ANFIS) is implemented for system modeling with Principle Component Analysis (PCA) as a feature extraction measure in order to reduce the complexity and improve the performance of the system. Experiments have been conducted and the results have demonstrated that the system is able to produce quantified outcomes reflecting patient's motion performance according to the Brunnstrom approach. An 87.5% of cross-validation correct rate was achieved for Brunnstrom stages classification.