Brunnstrom stage automatic evaluation for stroke patients using extreme learning machine

Brunnstrom stage is widely used to evaluate the movement function of stroke patients during rehabilitation by physicians. In this paper, a new method, which is based on extreme learning machine (ELM) and the Internet technology, is proposed to realize intelligent Brunnstrom stages evaluation for upper limb movement function of stroke patients. Preliminary experiment has been conducted with movement data collected from 23 stroke patients and 4 healthy people. The experiment results show that, compared with the experienced physicians evaluation results, the accuracy of the established ELM model can reach 92.1%, which means the proposed method is helpful for physicians to remotely evaluate those stroke patients who finish rehabilitation exercises at home or community, and is helpful to solving the problem of the lack of medical resource and the high cost of inpatient rehabilitation.