Sensor-Aware Data Imputation for Time-Series Machine Learning on Low-Power Wearable Devices
Open Access
- 8 November 2024
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 30 (1) , 1-27
Abstract
Wearable devices that have low-power sensors, processors, and communication capabilities are gaining wide adoption in several health applications. The machine learning algorithms on these devices assume that data from all sensors are available during runtime. However, data from one or more sensors may be unavailable due to energy or communication challenges. This loss of sensor data can result in accuracy degradation of the application. Prior approaches to handle missing data, such as generative models or training multiple classifiers for each combination of missing sensors are not suitable for low-energy wearable devices due to their high overhead at runtime. In contrast to prior approaches, we present an energy-efficient approach, referred to as Sensor-Aware iMputation (SAM), to accurately impute missing data at runtime and recover application accuracy. SAM first uses unsupervised clustering to obtain clusters of similar sensor data patterns. Next, it learns inter-relationship between clusters to obtain imputation patterns for each combination of clusters using a principled sensor-aware search algorithm. Using sensor data for clustering before choosing imputation patterns ensures that the imputation is aware of sensor data observations. Experiments on seven diverse wearable sensor-based time-series datasets demonstrate that SAM is able to maintain accuracy within 5% of the baseline with no missing data when one sensor is missing. We also compare SAM against generative adversarial imputation networks (GAIN), transformers, and k-nearest neighbor methods. Results show that SAM outperforms all three approaches on average by more than 25% when two sensors are missing with negligible overhead compared to the baseline.Keywords
Funding Information
- NSF CAREER (Award CNS-2238257)
- AgAID AI Institute supported by the United States Department of Agriculture - National Institute of Food and Agriculture (award #2021-67021-35344)
This publication has 37 references indexed in Scilit:
- Stress Monitoring in Free-Living EnvironmentsIEEE Journal of Biomedical and Health Informatics, 2023
- Analyzing Deep Learning for Time-Series Data Through Adversarial Lens in Mobile and IoT ApplicationsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020
- Validity of the Empatica E4 Wristband to Measure Heart Rate Variability (HRV) Parameters: a Comparison to Electrocardiography (ECG)Journal of Medical Systems, 2020
- HERMIT: A Benchmark Suite for the Internet of Medical ThingsIEEE Internet of Things Journal, 2018
- Wearable Medical Sensor-Based System Design: A SurveyIEEE Transactions on Multi-Scale Computing Systems, 2017
- A clinical view on the development of technology-based tools in managing Parkinson's diseaseMovement Disorders, 2016
- Technology in Parkinson's disease: Challenges and opportunitiesMovement Disorders, 2016
- The Teardown: Apple WatchEngineering & Technology, 2015
- A Survey on Human Activity Recognition using Wearable SensorsIEEE Communications Surveys & Tutorials, 2012
- Analysis of impairments influencing gait velocity and asymmetry of hemiplegic patients after mild to moderate stroke11No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors(s) or upon any organization with which the author(s) is/are associated.Archives of Physical Medicine and Rehabilitation, 2003