Quantitative identification of ventral/dorsal nerves through intraoperative neurophysiological monitoring by supervised machine learning

• DOI: 10.3389/fped.2023.1118924
• Publication Date: 2023-05-18T07:30:33Z
• Authors: Wenbin Jiang; Qijia Zhan; Junlu Wang; Min Wei; Sen Li; Rong Mei; Bo Xiao
• Keywords: Neurophysiology; Intraoperative neurophysiological monitoring; Identification

Objective This study aimed to investigate the electro-neurophysiological characteristics of ventral and dorsal nerves at L2 segment in a quantitative manner. Methods Medical records consecutive patients who underwent single-level approach selective rhizotomy (SDR) from June 2019 January 2022 were retrospectively reviewed. Intraoperative data analyzed. Results A total 74 males 27 females included current with mean age 6.2 years old. Quadriceps adductors two main muscle groups innervated by nerve roots both roots. Dorsal have higher threshold than that ones, muscles first reached 200 µV longer latency smaller compound action potential (CMAP) those ones. Supervised machine learning can efficiently distinguish ventral/dorsal using + or CMAP as predictors. Conclusion Electro-neurophysiological parameters could be used differentiate fibers during SDR.