SHIP: A Shapelet-based Approach for Interpretable Patient-Ventilator Asynchrony Detection

2025-03-09

AI Summary

This study addresses patient–ventilator asynchrony (PVA), a prevalent clinical challenge affecting up to 85% of mechanically ventilated patients. To overcome key limitations of existing deep learning and statistical models—including severe class imbalance, lack of interpretability, and insufficient clinical validation—we propose the first shapelet-based, interpretable PVA detection framework. Our method identifies discriminative time-series shapelets via unsupervised shapelet mining to construct a shapelet dictionary; leverages shapelet-driven data augmentation and integrates shapelet-derived statistical features; and employs a lightweight classifier (e.g., XGBoost) for efficient, real-time classification. Evaluated on a real-world clinical dataset, the approach achieves significant improvements in F1-score and AUC over state-of-the-art baselines. Crucially, it provides human-interpretable, visualizable shapelet matching evidence—enabling clinicians to trace model decisions directly to physiological waveform patterns. This work establishes a new paradigm for clinically deployable, real-time PVA monitoring and intervention support.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Multi-modal Vision
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Patient-ventilator asynchrony (PVA) is a common and critical issue during mechanical ventilation, affecting up to 85% of patients. PVA can result in clinical complications such as discomfort, sleep disruption, and potentially more severe conditions like ventilator-induced lung injury and diaphragm dysfunction. Traditional PVA management, which relies on manual adjustments by healthcare providers, is often inadequate due to delays and errors. While various computational methods, including rule-based, statistical, and deep learning approaches, have been developed to detect PVA events, they face challenges related to dataset imbalances and lack of interpretability. In this work, we propose a shapelet-based approach SHIP for PVA detection, utilizing shapelets - discriminative subsequences in time-series data - to enhance detection accuracy and interpretability. Our method addresses dataset imbalances through shapelet-based data augmentation and constructs a shapelet pool to transform the dataset for more effective classification. The combined shapelet and statistical features are then used in a classifier to identify PVA events. Experimental results on medical datasets show that SHIP significantly improves PVA detection while providing interpretable insights into model decisions.

Problem

• Detects patient-ventilator asynchrony accurately and interpretably.
• Addresses dataset imbalances using shapelet-based data augmentation.
• Improves clinical outcomes by reducing manual adjustment errors.

Innovation

• Shapelet-based PVA detection enhances accuracy
• Data augmentation addresses dataset imbalances
• Combined shapelet and statistical features improve classification