Fig.4 Classification accuracy of different classifiers in the three conditions
Conclusion : The active rehabilitation system based on BCI is an important tool in the future process. In this study, we proposed an effective method to detect the motion intention of low limbs using EEG signals. Firstly, a movement experiment was designed with the EEG signal synchronously recorded and the brain was divided into eight regions. Then, a series of entropy features and complexity parameters were discussed respectively for the grand averaged EEGs from each brain region. Finally, through the statistical analysis, relevant features with significant differences among the three motion conditions are screened out, combined with SVM, LDA and LR to realize the classification of standing, sitting and quiet. The highest classification accuracy is up to 87%, which proves the feasibility of detecting and recognizing the lower limb motion intention based on the complexity parameters of EEGs. This study also provides a new insight for EEG based motion intention detection, and has a certain reference value for the development of BCI based lower limb walking aid and rehabilitation robot.
Acknowledgments: This work was supported by the Science and Technology Project of Lanzhou City in China (No. 2021-1-150), and the Science and Technology Project of Gansu Province in China (No. 20JR10RA215), and the open project of Neuracle Company (No. BRKOT-LZJTU-20220329C). We also would like to thanks the support from Tianyou Youth Talent Lift Program of Lanzhou Jiaotong University, Lanzhou, China.
Received: 22 Augest 2022 Accepted: xx March 2022
doi: 10.****
References
  1. Hamacher D., et al.: ‘Brain activity during walking: A systematic review’, Neuroscience and Biobehavioral Reviews ., 2015, 57: 310-327.
  2. Lennon O., et al.: ‘A systematic review establishing the current state-of-the-art, the limitations, and the DESIRED checklist in studies of direct neural interfacing with robotic gait devices in stroke rehabilitation’, Frontiers in Neuroscience , 2020, 14: 578-580.
  3. Villa-Parra AC., et al.: ‘Towards a robtic knee exoskeleton control based on human motion intention through EEG and sEMG signals’, Procedia Manufacturing, 2015, 3: 1379-1386.
  4. Bulea TC., et al.: ‘Sitting and Standing intention can be decoded from scalp EEG recorded prior to movement execution’, Frontiers in Neuroscience , 2014, 8: 376(1-19).
  5. Gwin JT., et al.: ‘Electrocortical activity is coupled to gait cycle phase during treadmill walking’, NeruoImage , 2011,54:1289-1296.
  6. Bulea TC., et al.: ‘Classification of stand-to-Sit and Sit-to-Stand Movement from Low Frequency EEG with Locality Preserving Dimensionality Reduction’, The 35th Annual International Conference of the IEEE EMBS , Osaka, Japan, 3-7. Jul. 2014.
  7. Zhong Z., et al.: ‘Event-related EEG changes during attempted standing up task’, The 6th International Symposium Noninvasive Functional Source Imaging of the Brain and Heart & International Conference Functional Biomedical Imaging (NFSI-ICFBI ), 2007, 66–69.
  8. Li J., et al.: ‘Brain Functional Connectivity in Unconstrained Walking With and Without an Exoskeleton’, IEEE Trans Neural Syst Rehabil Eng , 2020, 28(3): 730-739.
  9. Handiru VS., et al.: ‘Graph-theoretical analysis of EEG functional connectivity during balance perturbation in traumatic brain injury: A pilot study’, Human Brain Mapping , 2021, 1-21.
  10. Lau TM., et al.: ‘Walking reduces sensorimotor network connectivity compared to standing’, Journal of Neuro Engineering and Rehabilitation , 2014, 11:14.
  11. Rodríguez-Sotelo L., et al.: ‘Automatic Sleep Stages Classification Using EEG Entropy Features and Unsupervised Pattern Analysis Techniques’, Entropy , 2014, 16: 6573-6589.
  12. Xi X., et a.: ‘Weighted kernel FDA fall recognition of EMG fuzzy entropy’, Acta Electronica Sinica , 2016, 44(06): 1376-1382. (in Chinese)