|
|
|
| Establishment of a Rehabilitation Training Prescription Recommended Model for Stroke Based on Brain Function and Movement Assessment |
| Zhang Tengyu1#&, Zhang Jingsha1#&, Xu Gongcheng2, Wang Zheng1, Zhang Xuemin1, Li Zengyong1,2#* |
1(Key Laboratory of Neuro-Functional Information and Rehabilitation Engineering of the Ministry of Civil Affairs, Beijing Key Laboratory of Rehabilitation Technical Aids for Old-Age Disability, National Research Center for Rehabilitation Technical Aids, Beijing 100176, China)
2(School of Biological Science and Medical Engineering,Beihang University, Beijing 100191, China) |
|
|
|
|
Abstract To establish intelligent recommendation model for rehabilitation training prescriptions based on the evaluation indexes of movement and brain function for stroke patients,120 stroke patients were evaluated by Brunnstrom, Holden walking ability scale, Berg balance scale and improved Ashworth scale, and the cerebral blood oxygen data were collected at rest state by near-infrared brain functional imaging system. The scale evaluation results and the brain function evaluation indexes (activate degrees, cornering, and function connection between different brain regions) were extracted, the support vector machine (SVM), the convolutional neural network (CNN) and the improved CNN-SVM algorithm were used to build the prescription recommended model, and the training modes of eight kinds of training contents were recommended. The model using the scale evaluation results and the functional connectivity index between different brain regions as features had the highest recognition accuracy, and the average recognition accuracy of the three algorithms was above 93%. Among them, the recognition accuracy of the improved CNN-SVM algorithm was highest, which reached 96.43%. This method realized the intelligent recommendation of rehabilitation training prescription based on the movement and brain function evaluation data of stroke patients, which would be beneficial to personalized and accurate rehabilitation.
|
|
Received: 20 January 2021
|
|
|
|
|
|
[1] 王陇德,刘建民,杨弋,等. 我国脑卒中防治仍面临巨大挑战——《中国脑卒中防治报告2018》概要[J]. 中国循环杂志, 2019,34(2):105-119.
[2] 张星星,段宏为,周晨,等. 虚拟现实技术应用于脑卒中康复的有效性及安全性的系统评价再评价[J]. 中国全科医学, 2020, 23(5):67-73.
[3] 王萌亚,王仲朋,陈龙,等. 卒中后运动神经反馈康复训练研究进展与前景[J]. 中国生物医学工程学报, 2019,38(6):742-752.
[4] 卢清朗,刘善铭,陈海清,等. 个性化康复训练处方对脑卒中后肩手综合征患者的影响[J]. 临床心身疾病杂志, 2015,21(Z1):157-158.
[5] Krebs HI, Volpe BT, Aisen ML, et al. Increasing productivity and quality of care: Robot-aided neuro-rehabilitation [J]. Journal of Rehabilitation Research & Development, 2000, 37(6):639-652.
[6] 张俊杰,孙光民,李煜,等. 基于加速度传感器的上肢运动信息采集与姿态识别[J]. 北京工业大学学报,2017,43(7):978-986.
[7] Yu L, Xiong D, Guo L, et al. A remote quantitative Fugl-Meyer assessment framework for stroke patients based on wearable sensor networks [J]. Comput Methods Programs Biomed, 2016, (128):100-110.
[8] 廖梦佳,秦亚杰,汪源源,等. 基于多传感器融合的用于脑卒中患者的可穿戴式上肢动作识别系统[J]. 中国康复医学杂志,2015,30(5):443-446.
[9] 陈少发,马强,赵君豪,等. 基于上肢运动评分的动作检测与动作识别的方法研究[J]. 中国康复医学杂志, 2019(6):707-710.
[10] 王群,谢斌,黄真,等. 脑卒中偏瘫患者上肢运动功能障碍的生物力学机制研究[J]. 中华物理医学与康复杂志, 2017(39):727-731.
[11] Bar-On L, Aertbeliën E, Wambacq H, et al. A clinical measurement to quantify spasticity in children with cerebral palsy by integration of multidimensional signals [J]. Gait & Posture, 2013, 38(1):141-147.
[12] 朱晓斐. 基于表面肌电和运动传感器的肌痉挛量化评估[D]. 合肥:中国科学技术大学. 2018.
[13] 徐功铖. 基于近红外光谱技术的驾驶员注意力分散态脑功能连接特性分析[D]. 济南:山东大学,2018.
[14] 徐功铖, 霍聪聪, 李文昊,等. 基于功能性近红外光谱的脑卒中后偏侧与双侧运动训练对脑功能的影响分析[J]. 医用生物力学, 2019(S1).182-182.
[15] Xu G, Zhang M, Wang Y, et al. Functional connectivity analysis of distracted drivers based on the wavelet phase coherence of functional near-infrared spectroscopy signals [J]. PLoS ONE, 2017, 12(11):e0188329.
[16] Niu XX, Suen CY. A novel hybrid CNN-SVM classifier for recognizing handwritten digits [J]. Pattern Recognition, 2012, 45(4):1318-1325.
[17] 王英杰, 沈辉. 脑静息功能连接自发波动的可变性分析[J]. 中国生物医学工程学报, 2017, 36(1):20-27.
[18] Park CH, Chang WH, Ohn SH, et al. Longitudinal changes of resting-state functional connectivity during motor recovery after stroke [J]. Stroke, 2011, 42(5):1357-1362.
[19] 闫铮,俞谢益,吴畏. 基于信息流增益算法的脑运动功能康复效果评价研究 [J]. 中国生物医学工程学报, 2015, 34(3):365-369.
[20] Dubovik S, Pignat JM, Ptak R, et al. The behavioral significance of coherent resting-state oscillations after stroke [J]. Neuroimage, 2012, 61(1): 249-57.
[21] Golestani AM, Tymchuk S, Demchuk A, et al. Longitudinal evaluation of resting-state FMRI after acute stroke with hemiparesis [J]. Neurorehabil Neural Repair, 2013, 27(2): 153-163.
[22] Wu J, Burke QE, Lucy D, et al. Connectivity measures are robust biomarkers of cortical function and plasticity after stroke [J]. Brain, 2015, 138:1359-1359.
[23] Fabrizio DVF, Pichiorri F, Morone G, et al. Multiscale topological properties of functional brain networks during motor imagery after stroke [J]. Neuroimage, 2013, 83(Complete): 438-449.
[24] Nicolo P, Rizk S, Magnin C, et al. Coherent neural oscillations predict future motor and language improvement after stroke[J]. Brain, 2015, 138: 3048-3060.
[25] 王志斌. 仿真图像应用于深度学习的可行性方法研究[D]. 西安:西安电子科技大学,2020. |
|
|
|
