Imputing missing multi-sensor data in the healthcare domain: A systematic review

2025-10-01

Image and Vision Computing

AI Summary

This study addresses the critical challenge posed by prevalent missing values in multi-sensor health monitoring data, particularly under prolonged missingness scenarios that severely hinder timely prediction of chronic events such as hypoglycemia. To this end, the authors propose a systematic imputation framework tailored for multi-sensor time series, which adaptively selects or integrates multiple advanced imputation strategies—including both machine learning and deep learning approaches—based on feature types, temporal dynamics, and the duration of missing segments. Experimental results demonstrate that the proposed method significantly improves imputation accuracy and robustness, thereby enhancing the reliability of hypoglycemia event prediction and offering an effective paradigm for preprocessing complex health data.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Problem

• missing data
• multisensor data
• hypoglycemia prediction
• time-series imputation
• healthcare monitoring

Innovation

• missing data imputation
• multisensor time-series
• feature-specific imputation
• temporal dynamics
• healthcare monitoring