Learning from EMG: semi-automated grading of facial nerve function

Abstract

The current grading of facial nerve function is based on subjective impression with the established assessment scale of House and Brackmann (HB). Especially for research a more objective method is needed to lower the interobserver variability to a minimum. We developed a semi-automated grading system based on (facial) surface EMG-data measuring the facial nerve function of 28 patients with vestibular schwannoma surgery. The sEMG was recorded preoperatively, postoperatively and after 3–12 months. In addition, the HB grade was determined. After manual selection and preprocessing, the data were subjected to machine learning classificators (Logistic regression, SVM and KNN). Lateralization indices were calculated and multivariant machine learning analysis was performed according to three scenarios [differentiation of normal (1) and slight (2) vs. impaired facial nerve function and classification of HB 1-3 (3)]. The calculated AUC for each scenario showed overall good differentiation capability with a median AUC of 0.72 for scenario 1, 0.91 for scenario 2 and multiclass AUC of 0.74 for scenario 3. This study approach using sEMG and machine learning shows feasibility regarding facial nerve grading in perioperative VS-surgery setting. sEMG may be a viable alternative to House Brackmann regarding objective evaluation of facial function especially for research purposes.

Facial Paralysis Diagnosis and Treatment Assessment Computational Model

“…Holze et al proposed an automatic grading assessment method for facial paralysis based on video analysis. (11) Existing research methods primarily focus on facial paralysis detection and the assessment of facial asymmetry. Improving the accuracy of these detection methods and asymmetry assessments is crucial, and achieving such improvement requires precise quantification and evaluation of facial features throughout the diagnosis and treatment process for facial paralysis patients.…”

“…However, current research methods often lack detailed quantification and comprehensive evaluation of facial asymmetry. (10,11) In this study, we aim to construct an asymmetry calculation model and algorithm specifically designed for facial paralysis diagnosis and treatment. The proposed model and algorithm are aimed at quantifying facial asymmetry and will assist physicians to enhance the accuracy and efficiency of diagnosing facial paralysis.…”

Correlation of 3D Morphometric Changes, Kinematics, and Muscle Activity During Smile

“…4 The utilization of sEMG combined with machine learning techniques has shown promise in assessing facial nerve grading and could serve as a viable alternative to the House-Brackmann scale for objectively evaluating facial function, particularly in research settings. 5 With the development of three-dimensional (3D) surface scanners, it has become possible to quantitatively evaluate changes in facial morphology at rest and even during facial expressions. 6,7 Furthermore, functional recovery could be demonstrated quantitatively.…”

“…However, the effectiveness of surface electromyography (sEMG), the non‐invasive equivalent of the method, in examining the activity of mimic muscles has been proven in studies 4 . The utilization of sEMG combined with machine learning techniques has shown promise in assessing facial nerve grading and could serve as a viable alternative to the House‐Brackmann scale for objectively evaluating facial function, particularly in research settings 5 …”

Opportunities and challenges of supervised machine learning for the classification of motor evoked potentials according to muscles

“…Recently, there has been an increased interest in applying ML to IOM data [ 16 ]. Among other examples, Holze et al applied supervised ML to facial surface electromyography (EMG) data to assess facial function [ 17 ], Jamaludin et al used algorithms to predict functional outcomes based on transcranial MEPs [ 18 ] and Zha et al used neural networks to investigate automated classification of free-running EMG waveforms [ 19 ]. Presently, Mirallave Pescador et al propose to use Bayesian Networks to assess evidence in IOM [ 20 ].…”