Exploratory Outlier Detection for Acceleromyographic Neuromuscular Monitoring: Machine Learning Approach.

Michaël Verdonck,Hugo Carvalho,Johan Berghmans,Jan Poelaert,Patrice Forget

doi:10.2196/25913

Abstract

BackgroundPerioperative quantitative monitoring of neuromuscular function in patients receiving neuromuscular blockers has become internationally recognized as an absolute and core necessity in modern anesthesia care. Because of their kinetic nature, artifactual recordings of acceleromyography-based neuromuscular monitoring devices are not unusual. These generate a great deal of cynicism among anesthesiologists, constituting an obstacle toward their widespread adoption. Through outlier analysis techniques, monitoring devices can learn to detect and flag signal abnormalities. Outlier analysis (or anomaly detection) refers to the problem of finding patterns in data that do not conform to expected behavior.ObjectiveThis study was motivated by the development of a smartphone app intended for neuromuscular monitoring based on combined accelerometric and angular hand movement data. During the paired comparison stage of this app against existing acceleromyography monitoring devices, it was noted that the results from both devices did not always concur. This study aims to engineer a set of features that enable the detection of outliers in the form of erroneous train-of-four (TOF) measurements from an acceleromyographic-based device. These features are tested for their potential in the detection of erroneous TOF measurements by developing an outlier detection algorithm.MethodsA data set encompassing 533 high-sensitivity TOF measurements from 35 patients was created based on a multicentric open label trial of a purpose-built accelero- and gyroscopic-based neuromuscular monitoring app. A basic set of features was extracted based on raw data while a second set of features was purpose engineered based on TOF pattern characteristics. Two cost-sensitive logistic regression (CSLR) models were deployed to evaluate the performance of these features. The final output of the developed models was a binary classification, indicating if a TOF measurement was an outlier or not.ResultsA total of 7 basic features were extracted based on raw data, while another 8 features were engineered based on TOF pattern characteristics. The model training and testing were based on separate data sets: one with 319 measurements (18 outliers) and a second with 214 measurements (12 outliers). The F1 score (95% CI) was 0.86 (0.48-0.97) for the CSLR model with engineered features, significantly larger than the CSLR model with the basic features (0.29 [0.17-0.53]; P<.001).ConclusionsThe set of engineered features and their corresponding incorporation in an outlier detection algorithm have the potential to increase overall neuromuscular monitoring data consistency. Integrating outlier flagging algorithms within neuromuscular monitors could potentially reduce overall acceleromyography-based reliability issues.Trial RegistrationClinicalTrials.gov NCT03605225; https://clinicaltrials.gov/ct2/show/NCT03605225

Highlights

Postoperative residual curarization remains a frequent and often concealed event within modern anesthesia care [1]
A data set encompassing 533 high-sensitivity TOF measurements from 35 patients was created based on a multicentric open label trial of a purpose-built accelero- and gyroscopic-based neuromuscular monitoring app
Besides a reduction of the incidence of severe respiratory complications [8,9,10], quantitative monitoring potentially leads to considerable financial health care savings, with complications stemming from suboptimal neuromuscular monitoring being estimated to be as high as US $25.000 per patient per event [7]

Summary

Introduction

Postoperative residual curarization remains a frequent and often concealed event within modern anesthesia care [1]. Kopman et al [13] have scrutinized some algorithmic simplifications used by common AMG monitors (T4/T2 ratio as a substitute for T4/T1) and how their validity is dependent on the degree of recovery from nondepolarizing neuromuscular block Such interpretative considerations, associated frequent artifactual confounders, and known overestimation tendencies when compared with electromyography (EMG) or mechanomyography [14] contribute to the perpetuation of anesthesiologist’s cynicism toward objective neuromuscular monitoring methods, further hindering their widespread adoption [15]. Perioperative quantitative monitoring of neuromuscular function in patients receiving neuromuscular blockers has become internationally recognized as an absolute and core necessity in modern anesthesia care Because of their kinetic nature, artifactual recordings of acceleromyography-based neuromuscular monitoring devices are not unusual. Outlier analysis (or anomaly detection) refers to the problem of finding patterns in data that do not conform to expected behavior

Methods

Results

Conclusion