The new study was performed by a research team at the University of Miami using clinical data from Mayo Clinic and the University of Debrecen to develop and validate an artificial intelligence (AI)-based convolutional neural network (CNN) that correctly identifies valid compound muscle action potentials (CMAPs) from the TetraGraph quantitative neuromuscular monitoring system.

The study used Senzime’s TetraGraph system to demonstrate the feasibility of using AI to separate valid cMAPs from artifact. The CNN algorithm showed an accuracy exceeding 99.5 percent in distinguishing the TetraGraph’s valid CMAPs from artifact.

“TetraGraph is well-suited for the application of AI to improve the process of achieving and maintaining deep, stable levels of neuromuscular block while minimizing the risk of underdosing and overdosing neuromuscular blocking drugs,” comments Professor Richard H. Epstein, from the University of Miami, USA, the first author of the clinical study.

“The study is the first of its kind to validate TetraGraph waveforms using an AI-based neural network, demonstrating impressive accuracy. This confirms the leading quality of our innovative algorithm-powered patient monitoring technology,” comments Philip Siberg, CEO of Senzime.

The study Validation of a convolutional neural network that reliably identifies electromyographic compound motor action potentials following train-of-four stimulation: an algorithm development experimental study has been published in British Journal of Anesthesia Open (BJA Open). A related abstract was selected as one of the 12 Best Basic Science abstracts presented at the 2023 Annual Meeting of the American Society of Anesthesiologists in San Francisco, USA.