Ensemble averaging for categorical variables: Validation study of imputing lost data in 24-h recorded postures of inpatients

Abstract

Acceleration sensors are widely used in consumer wearable devices and smartphones. Postures estimated from recorded accelerations are commonly used as features indicating the activities of patients in medical studies. However, recording for over 24 h is more likely to result in data losses than recording for a few hours, especially when consumer-grade wearable devices are used. Here, to impute postures over a period of 24 h, we propose an imputation method that uses ensemble averaging. This method outputs a time series of postures over 24 h with less lost data by calculating the ratios of postures taken at the same time of day during several measurement-session days. Whereas conventional imputation methods are based on approaches with groups of subjects having multiple variables, the proposed method imputes the lost data variables individually and does not require other variables except posture. We validated the method on 306 measurement data from 99 stroke inpatients in a hospital rehabilitation ward. First, to classify postures from acceleration data measured by a wearable sensor placed on the patient’s trunk, we preliminary estimated possible thresholds for classifying postures as ‘reclining’ and ‘sitting or standing’ by investigating the valleys in the histogram of occurrences of trunk angles during a long-term recording. Next, the imputations of the proposed method were validated. The proposed method significantly reduced the missing data rate from 5.76% to 0.21%, outperforming a conventional method.

Prediction of stroke patients’ bedroom-stay duration: machine-learning approach using wearable sensor data

“…Specifically, this allowed the system to differentiate between walking and non-walking states when the initial posture was identified as sitting/standing. This posture-classification method has been validated through studies involving both healthy individuals and clinical patients ( Rauen et al, 2018 ; Ogasawara et al, 2023 ). In addition to posture classifications, moving standard averaging of trunk acceleration (MSDA) is used as an indicator of the physical-activity intensity of stroke patients ( Mukaino et al, 2022 ).…”

“…This loss can occur due to various factors including equipment failure, limitations in data-collection capabilities, and human error ( Cismondi et al, 2013 ; Lai et al, 2020 ). To mitigate the impact of missing data, we implemented a data-imputation technique that is based on averaging ( Ogasawara et al, 2021 ; Ogasawara et al, 2023 ). This technique generates a time series of features of stroke-patient activity over a 24-h period with reduced data loss by calculating the averages and ratios of these features at the same time of day during several measurement-session days.…”

Clinical impact of non-lying time on hospital-associated functional decline in older patients undergoing transcatheter aortic valve implantation

“…Using the angle and a designated threshold, postures are initially classified as ‘lying’ or ‘non-lying’. The reliability of this posture recognition approach has been reported in both healthy subjects [ 13 ] and clinical patients [ 15 ].…”