Decoding accelerometry for classification and prediction of critically ill patients with severe brain injury

Shubhayu Bhattacharyay; John Rattray; Matthew Wang; Peter H Dziedzic; Eusebia Calvillo; Han B Kim; Eshan Joshi; Pawel Kudela; Ralph Etienne-Cummings; Robert D Stevens

doi:10.1038/s41598-021-02974-w

Decoding accelerometry for classification and prediction of critically ill patients with severe brain injury

Sci Rep. 2021 Dec 8;11(1):23654. doi: 10.1038/s41598-021-02974-w.

Authors

Shubhayu Bhattacharyay^{1

2

3

4}, John Rattray⁵, Matthew Wang⁶, Peter H Dziedzic^{7

8}, Eusebia Calvillo⁸, Han B Kim^{7

9}, Eshan Joshi⁶, Pawel Kudela¹⁰, Ralph Etienne-Cummings⁵, Robert D Stevens^{7

8

9

10}

Affiliations

¹ Laboratory of Computational Intensive Care Medicine, Johns Hopkins University, Baltimore, MD, USA. sb2406@cam.ac.uk.
² Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK. sb2406@cam.ac.uk.
³ Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA. sb2406@cam.ac.uk.
⁴ Department of Applied Mathematics and Statistics, Johns Hopkins University, Baltimore, MD, USA. sb2406@cam.ac.uk.
⁵ Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA.
⁶ Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA.
⁷ Laboratory of Computational Intensive Care Medicine, Johns Hopkins University, Baltimore, MD, USA.
⁸ Department of Neurology, Johns Hopkins University, Baltimore, MD, USA.
⁹ Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, Baltimore, MD, USA.
¹⁰ Department of Neurosurgery, Johns Hopkins University, Baltimore, MD, USA.

Abstract

Our goal is to explore quantitative motor features in critically ill patients with severe brain injury (SBI). We hypothesized that computational decoding of these features would yield information on underlying neurological states and outcomes. Using wearable microsensors placed on all extremities, we recorded a median 24.1 (IQR: 22.8-25.1) hours of high-frequency accelerometry data per patient from a prospective cohort (n = 69) admitted to the ICU with SBI. Models were trained using time-, frequency-, and wavelet-domain features and levels of responsiveness and outcome as labels. The two primary tasks were detection of levels of responsiveness, assessed by motor sub-score of the Glasgow Coma Scale (GCSm), and prediction of functional outcome at discharge, measured with the Glasgow Outcome Scale-Extended (GOSE). Detection models achieved significant (AUC: 0.70 [95% CI: 0.53-0.85]) and consistent (observation windows: 12 min-9 h) discrimination of SBI patients capable of purposeful movement (GCSm > 4). Prediction models accurately discriminated patients of upper moderate disability or better (GOSE > 5) with 2-6 h of observation (AUC: 0.82 [95% CI: 0.75-0.90]). Results suggest that time series analysis of motor activity yields clinically relevant insights on underlying functional states and short-term outcomes in patients with SBI.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Accelerometry
Aged
Brain Injuries / classification*
Brain Injuries / pathology
Critical Illness*
Female
Glasgow Outcome Scale
Humans
Male
Middle Aged
Pilot Projects
Severity of Illness Index