Body Area Networks: Smart IoT and Big Data for Intelligent Health Management. 14th EAI International Conference, BODYNETS 2019, Florence, Italy, October 2-3, 2019, Proceedings

Research Article

Towards Body Sensor Network Based Gait Abnormality Evaluation for Stroke Survivors

  • @INPROCEEDINGS{10.1007/978-3-030-34833-5_9,
        author={Sen Qiu and Xiangyang Guo and Hongyu Zhao and Zhelong Wang and Qimeng Li and Raffaele Gravina},
        title={Towards Body Sensor Network Based Gait Abnormality Evaluation for Stroke Survivors},
        proceedings={Body Area Networks:  Smart IoT and Big Data for Intelligent Health Management. 14th EAI International Conference, BODYNETS 2019, Florence, Italy, October 2-3, 2019, Proceedings},
        proceedings_a={BODYNETS},
        year={2019},
        month={11},
        keywords={Body sensor network Human gait analysis Information fusion Rehabilitation Micro-electro-mechanical sensor},
        doi={10.1007/978-3-030-34833-5_9}
    }
    
  • Sen Qiu
    Xiangyang Guo
    Hongyu Zhao
    Zhelong Wang
    Qimeng Li
    Raffaele Gravina
    Year: 2019
    Towards Body Sensor Network Based Gait Abnormality Evaluation for Stroke Survivors
    BODYNETS
    Springer
    DOI: 10.1007/978-3-030-34833-5_9
Sen Qiu,*, Xiangyang Guo1, Hongyu Zhao,*, Zhelong Wang,*, Qimeng Li2, Raffaele Gravina2
  • 1: Dalian University of Technology
  • 2: University of Calabria
*Contact email: qiu@dlut.edu.cn, zhaohy@dlut.edu.cn, wangzl@dlut.edu.cn

Abstract

Due to the technological advances of micro-electro-mechanical sensor and wireless sensor network, gait analysis has been widely adopted as an significant indicator of mobility impairment for stroke survivors. This paper aims to propose an wearable computing based gait impairment evaluation method with distribute inertial sensor unit (IMU) mounted on human lower limbs. Temporal-spacial gait metrics were evaluated on more than twenty post stroke patients and ten healthy control subjects in the 10-meters-walk-test. Experimental results shown that significant differences exist between stroke patients and healthy subject in terms of various gait metrics. The extracted gait metrics are consistent with clinical observations, and the position estimation accuracy has been validated by optical device. The proposed method has the potential to serve as an objective and cost-efficient tool for rehabilitation-assisting therapy for post stroke survivors in clinical practice.