CPR-Coach: Recognizing Composite Error Actions based on Single-class Training

TL;DR

This paper introduces CPR-Coach, a vision-based system for recognizing error actions in CPR training, proposing a new framework to improve multi-error recognition with limited supervision, and providing a benchmark dataset for future research.

Contribution

It constructs the first comprehensive CPR error action dataset and proposes ImagineNet, a novel framework to enhance multi-error recognition under single-class training constraints.

Findings

ImagineNet improves multi-error recognition accuracy.

CPR-Coach dataset enables detailed error analysis.

Existing models are evaluated on the new dataset.

Abstract

The fine-grained medical action analysis task has received considerable attention from pattern recognition communities recently, but it faces the problems of data and algorithm shortage. Cardiopulmonary Resuscitation (CPR) is an essential skill in emergency treatment. Currently, the assessment of CPR skills mainly depends on dummies and trainers, leading to high training costs and low efficiency. For the first time, this paper constructs a vision-based system to complete error action recognition and skill assessment in CPR. Specifically, we define 13 types of single-error actions and 74 types of composite error actions during external cardiac compression and then develop a video dataset named CPR-Coach. By taking the CPR-Coach as a benchmark, this paper thoroughly investigates and compares the performance of existing action recognition models based on different data modalities. To solve the unavoidable Single-class Training & Multi-class Testing problem, we propose a humancognition-inspired framework named ImagineNet to improve the model's multierror recognition performance under restricted supervision. Extensive experiments verify the effectiveness of the framework. We hope this work could advance research toward fine-grained medical action analysis and skill assessment. The CPR-Coach dataset and the code of ImagineNet are publicly available on Github.