Exploring a Gradient-based Explainable AI Technique for Time-Series Data: A Case Study of Assessing Stroke Rehabilitation Exercises
This work addresses the problem of interpretability in AI for healthcare professionals, specifically therapists assessing stroke rehabilitation, but it is incremental as it adapts existing methods to a new domain.
The paper tackled the limited exploration of explainable AI for time-series data in healthcare by applying a gradient-based technique to identify salient frames in stroke rehabilitation exercises, achieving a recall of 0.96 and an F2-score of 0.91.
Explainable artificial intelligence (AI) techniques are increasingly being explored to provide insights into why AI and machine learning (ML) models provide a certain outcome in various applications. However, there has been limited exploration of explainable AI techniques on time-series data, especially in the healthcare context. In this paper, we describe a threshold-based method that utilizes a weakly supervised model and a gradient-based explainable AI technique (i.e. saliency map) and explore its feasibility to identify salient frames of time-series data. Using the dataset from 15 post-stroke survivors performing three upper-limb exercises and labels on whether a compensatory motion is observed or not, we implemented a feed-forward neural network model and utilized gradients of each input on model outcomes to identify salient frames that involve compensatory motions. According to the evaluation using frame-level annotations, our approach achieved a recall of 0.96 and an F2-score of 0.91. Our results demonstrated the potential of a gradient-based explainable AI technique (e.g. saliency map) for time-series data, such as highlighting the frames of a video that therapists should focus on reviewing and reducing the efforts on frame-level labeling for model training.