skip to main content


Title: Impact of individual and treatment characteristics on wearable sensor-based digital biomarkers of opioid use
Abstract

Opioid use disorder is one of the most pressing public health problems of our time. Mobile health tools, including wearable sensors, have great potential in this space, but have been underutilized. Of specific interest are digital biomarkers, or end-user generated physiologic or behavioral measurements that correlate with health or pathology. The current manuscript describes a longitudinal, observational study of adult patients receiving opioid analgesics for acute painful conditions. Participants in the study are monitored with a wrist-worn E4 sensor, during which time physiologic parameters (heart rate/variability, electrodermal activity, skin temperature, and accelerometry) are collected continuously. Opioid use events are recorded via electronic medical record and self-report. Three-hundred thirty-nine discreet dose opioid events from 36 participant are analyzed among 2070 h of sensor data. Fifty-one features are extracted from the data and initially compared pre- and post-opioid administration, and subsequently are used to generate machine learning models. Model performance is compared based on individual and treatment characteristics. The best performing machine learning model to detect opioid administration is a Channel-Temporal Attention-Temporal Convolutional Network (CTA-TCN) model using raw data from the wearable sensor. History of intravenous drug use is associated with better model performance, while middle age, and co-administration of non-narcotic analgesia or sedative drugs are associated with worse model performance. These characteristics may be candidate input features for future opioid detection model iterations. Once mature, this technology could provide clinicians with actionable data on opioid use patterns in real-world settings, and predictive analytics for early identification of opioid use disorder risk.

 
more » « less
Award ID(s):
2124282
NSF-PAR ID:
10370298
Author(s) / Creator(s):
; ; ; ; ;
Publisher / Repository:
Nature Publishing Group
Date Published:
Journal Name:
npj Digital Medicine
Volume:
5
Issue:
1
ISSN:
2398-6352
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Opioid use disorder is a medical condition with major social and economic consequences. While ubiquitous physiological sensing technologies have been widely adopted and extensively used to monitor day-to-day activities and deliver targeted interventions to improve human health, the use of these technologies to detect drug use in natural environments has been largely underexplored. The long-term goal of our work is to develop a mobile technology system that can identify high-risk opioid-related events (i.e., development of tolerance in the setting of prescription opioid use, return-to-use events in the setting of opioid use disorder) and deploy just-in-time interventions to mitigate the risk of overdose morbidity and mortality. In the current paper, we take an initial step by asking a crucial question: Can opioid use be detected using physiological signals obtained from a wrist-mounted sensor? Thirty-six individuals who were admitted to the hospital for an acute painful condition and received opioid analgesics as part of their clinical care were enrolled. Subjects wore a noninvasive wrist sensor during this time (1-14 days) that continuously measured physiological signals (heart rate, skin temperature, accelerometry, electrodermal activity, and interbeat interval). We collected a total of 2070 hours (≈ 86 days) of physiological data and observed a total of 339 opioid administrations. Our results are encouraging and show that using a Channel-Temporal Attention TCN (CTA-TCN) model, we can detect an opioid administration in a time-window with an F1-score of 0.80, a specificity of 0.77, sensitivity of 0.80, and an AUC of 0.77. We also predict the exact moment of administration in this time-window with a normalized mean absolute error of 8.6% and R2 coefficient of 0.85. 
    more » « less
  2. Machine learning (ML) can be an appropriate approach to overcoming common problems associated with sensors for low-cost, point-of-care diagnostics, such as non-linearity, multidimensionality, sensor-to-sensor variations, presence of anomalies, and ambiguity in key features. This study proposes a novel approach based on ML algorithms (neural nets, Gaussian Process Regression, among others) to model the electrochemiluminescence (ECL) quenching mechanism of the [Ru(bpy)3]2+/TPrA system by phenolic compounds, thus allowing their detection and quantification. The relationships between the concentration of phenolic compounds and their effect on the ECL intensity and current data measured using a mobile phone-based ECL sensor is investigated. The ML regression tasks with a tri-layer neural net using minimally processed time series data showed better or comparable detection performance compared to the performance using extracted key features without extra preprocessing. Combined multimodal characteristics produced an 80% more enhanced performance with multilayer neural net algorithms than a single feature based-regression analysis. The results demonstrated that the ML could provide a robust analysis framework for sensor data with noises and variability. It demonstrates that ML strategies can play a crucial role in chemical or biosensor data analysis, providing a robust model by maximizing all the obtained information and integrating nonlinearity and sensor-to-sensor variations. 
    more » « less
  3. Estimating human joint moments using wearable sensors has utility for personalized health monitoring and generalized exoskeleton control. Data-driven models have potential to map wearable sensor data to human joint moments, even with a reduced sensor suite and without subject-specific calibration. In this study, we quantified the RMSE and R 2 of a temporal convolutional network (TCN), trained to estimate human hip moments in the sagittal plane using exoskeleton sensor data (i.e., a hip encoder and thigh- and pelvis-mounted inertial measurement units). We conducted three analyses in which we iteratively retrained the network while: 1) varying the input sequence length of the model, 2) incorporating noncausal data into the input sequence, thus delaying the network estimates, and 3) time shifting the labels to train the model to anticipate (i.e., predict) human hip moments. We found that 930 ms of causal input data maintained model performance while minimizing input sequence length (validation RMSE and R 2 of 0.141±0.014 Nm/kg and 0.883±0.025, respectively). Further, delaying the model estimate by up to 200 ms significantly improved model performance compared to the best causal estimators (p<0.05), improving estimator fidelity in use cases where delayed estimates are acceptable (e.g., in personalized health monitoring or diagnoses). Finally, we found that anticipating hip moments further in time linearly increased model RMSE and decreased R 2 (p<0.05); however, performance remained strong (R 2 >0.85) when predicting up to 200 ms ahead. 
    more » « less
  4. Stress detection and monitoring is an active area of research with important implications for an individual's personal, professional, and social health. Current approaches for stress classification use traditional machine learning algorithms trained on features computed from multiple sensor modalities. These methods are data and computation-intensive, rely on hand-crafted features, and lack reproducibility. These limitations impede the practical use of stress detection and classification systems in the real world. To overcome these shortcomings, we propose Stressalyzer, a novel stress classification and personalization framework from single-modality sensor data without feature computation and selection. Stressalyzer uses only Electrodermal activity (EDA) sensor data while providing competitive results compared to the state-of-the-art techniques that use multiple sensor modalities and are computationally expensive due to the calculation of large number of features. Using the dataset collected in a laboratory setting from $15$ subjects, our single-channel neural network-based model achieves a classification accuracy of 92.9% and an f1 score of 0.89 for binary stress classification. Our leave-one-subject-out analysis establishes the subjective nature of stress and shows that personalizing stress models using Stressalyzer significantly improves the model performance. Without model personalization, we found a performance decline in 40% of the subjects, suggesting the need for model personalization. 
    more » « less
  5. Background:

    Homelessness is an important social determinant of health (SDOH), impacting health outcomes for many medical conditions. Although homelessness is common among people with opioid use disorder (OUD), few studies systematically evaluate homelessness and other SDOH among people enrolled in standard of care treatment for OUD, medication for opioid use disorder (MOUD), or examine whether homelessness affects treatment engagement.

    Methods:

    Using data from the 2016 to 2018 U.S. Treatment Episode Dataset Discharges (TEDS-D), patient demographic, social, and clinical characteristics were compared between episodes of outpatient MOUD where homelessness was reported at treatment enrollment versus independent housing using pairwise tests adjusted for multiple testing. A logistic regression model examined the relationship between homelessness and treatment length and treatment completion while accounting for covariates.

    Results:

    There were 188 238 eligible treatment episodes. Homelessness was reported in 17 158 episodes (8.7%). In pairwise analysis, episodes involving homelessness were significantly different from those involving independent living on most demographic, social, and clinical characteristics, with significantly greater social vulnerability in most SDOH variables ( P’s < .05). Homelessness was significantly and negatively associated with treatment completion (coefficient = −0.0853, P < 0.001, 95% CI = [−0.114, −0.056], OR = 0.918) and remaining in treatment for greater than 180 days (coefficient = −0.3435, P < 0.001, 95% CI = [−0.371, −0.316], OR = 0.709) after accounting for covariates.

    Conclusions:

    Patients reporting homelessness at treatment entry in outpatient MOUD in the U.S. represent a clinically distinct and socially vulnerable population from those not reporting homelessness. Homelessness independently predicts poorer engagement in MOUD confirming that homelessness may be an independent predictor for MOUD treatment discontinuation nationally.

     
    more » « less