Neural-Network-based Adaptive Filtering of Cervical Spinal Evoked Potentials for the Localization Diagnosis of Cervical Spondylotic Myelopathy
Cervical Spondylotic Myelopathy (CSM) causes compression and damage to the spinal cord, which the most common spinal cord disorder affecting the elderly. Somatosensory evoked potentias(SEP), which indicate the functional status of the spinal cord, is a feasible methodology for diagnosis of CSM, but the poor signal to noise ratio limited the application of SEP diagnosis. This study proposes the use of an electrode array evoked potential measurement combined with advanced signal analysis techniques to enhance the SNR of surface SEP. Array SEPs were recorded from surface of cervical spine of 3 CSM patients. Neural-network-based adaptive filtering technique was employed to extract the SEP signals from the reference channel at adjacent level recording. The results of this study suggested the usefulness of neural-network-based adaptive filter to enable surface recorded SEP along cervical providing the localization diagnosis of CSM.
Yong Hu Keith DK Luk WW Lu
Department of Orthopaedic Surgery, The University of Hong Kong Hong Kong SAR, P.R. China
国际会议
8th International Conference on Neural Information Processing(ICONIP 2001)(第八届国际神经信息处理大会)
上海
英文
1627-1631
2001-11-14（万方平台首次上网日期，不代表论文的发表时间）