Abstract
This study investigated to what extent multimodal data can be used to detect mistakes during Cardiopulmonary Resuscitation (CPR) training. We complemented the Laerdal QCPR ResusciAnne manikin with the Multimodal Tutor for CPR, a multi-sensor system consisting of a Microsoft Kinect for tracking body position and a Myo armband for collecting electromyogram information. We collected multimodal data from 11 medical students, each of them performing two sessions of two-minute chest compressions (CCs). We gathered in total 5254 CCs that were all labelled according to five performance indicators, corresponding to common CPR training mistakes. Three out of five indicators, CC rate, CC depth and CC release, were assessed automatically by the ReusciAnne manikin. The remaining two, related to arms and body position, were annotated manually by the research team. We trained five neural networks for classifying each of the five indicators. The results of the experiment show that multimodal data can provide accurate mistake detection as compared to the ResusciAnne manikin baseline. We also show that the Multimodal Tutor for CPR can detect additional CPR training mistakes such as the correct use of arms and body weight. Thus far, these mistakes were identified only by human instructors. Finally, to investigate user feedback in the future implementations of the Multimodal Tutor for CPR, we conducted a questionnaire to collect valuable feedback aspects of CPR training.
Highlights
Mastering practical skills is a requirement in several professions and working domains.For example, working in construction requires to learn how to use a circular saw; nursing requires learning how to draw blood samples from patients; and repairing clothes requires being able to sew.Practical skill training, known as psychomotor learning, entails the acquisition of an apprenticeship learning model [1]
We proposed the use of a neural network to detect training mistakes in terms of chest compressions (CCs) rate, CC depth, and CC release, as well as to detect additional training mistakes, not currently tracked by commercial manikins, such as the correct locking of the arms during and the correct use of body posture and body weight during CC
We aimed at validating the Multimodal Tutor for Cardiopulmonary Resuscitation (CPR) on performance indicators currently implemented in the ResusciAnne using the manikin data as baseline measurements of the CPR performance (RQ1—Validation)
Summary
Known as psychomotor learning, entails the acquisition of an apprenticeship learning model [1]. In this model, a human expert demonstrates to the learner how to perform a specific task. The learner mimics the expert movements to develop a mental model of the psychomotor skill and, after some practice, this model is automated. Human instructors are not always available to follow each learner step by step, and their time is costly. The lack of instructors leads to the shortage of on-task, real-time and actionable feedback affecting negatively the quality of the training and resulting in longer and less efficient training sessions for the aspiring professionals. CPR has clear and well-defined criteria to measure the quality of the performance (we use the performance indicators defined by the European CPR Guidelines [16])
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
