Publication: Peripheral nerve function estimation by linear model of multi-CMAP responses for surgical intervention in acoustic neuroma surgery
Issued Date
2017-12-01
Resource Type
ISSN
2051817X
Other identifier(s)
2-s2.0-85037654841
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
Physiological Reports. Vol.5, No.23 (2017)
Suggested Citation
Dilok Puanhvuan, Sorayouth Chumnanvej, Yodchanan Wongsawat Peripheral nerve function estimation by linear model of multi-CMAP responses for surgical intervention in acoustic neuroma surgery. Physiological Reports. Vol.5, No.23 (2017). doi:10.14814/phy2.13495 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/41677
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Peripheral nerve function estimation by linear model of multi-CMAP responses for surgical intervention in acoustic neuroma surgery
Other Contributor(s)
Abstract
© 2017 The Authors. Nerve function assessments are crucial for surgical intervention during acoustic neuroma surgery. Cranial nerves such as acoustic and facial nerves, can be possibly damaged during tumor dissection. Proper surgical intervention should prevent neurological deficit and achieve total tumor removal. Conventionally, nerve function is qualitatively evaluated by surgeon and neurologist. Facial nerves can be preserved by monitoring the compound muscle action potential (CMAP) response. The differences in the amplitude and latency of CMAP are used as indicators during surgical interventions. However, baseline CMAPs cannot be recorded in the presence of large acoustic tumors. This paper presents a new way of estimating nerve function. Instead of a single CMAP examination, multi-CMAP responses are obtained from a train of varied stimulus intensities and these are applied a mathematical model. Shifts in the mathematical model parameters reflect changes in facial nerve function. In this study, experiments conducted in frog revealed that shifts in the linear model parameters were related to the level of induced nerve injury. Significant differences in the slope parameter of the linear model were found between each nerve condition. The identification of healthy and severed nerves via a support vector machine (SVM) corresponded to 94% accuracy. This classification criterion could be used with surgical intervention to prevent severed facial nerve palsy in acoustic neuroma surgery. The proposed method could be used to estimate nerve outcomes without prior information of a CMAP baseline.