Dataset Information

Predicting Multiple Sclerosis Outcomes During the COVID-19 Stay-at-home Period: Observational Study Using Passively Sensed Behaviors and Digital Phenotyping.

ABSTRACT:

Background

The COVID-19 pandemic has broad negative impact on the physical and mental health of people with chronic neurological disorders such as multiple sclerosis (MS).

Objective

We presented a machine learning approach leveraging passive sensor data from smartphones and fitness trackers of people with MS to predict their health outcomes in a natural experiment during a state-mandated stay-at-home period due to a global pandemic.

Methods

First, we extracted features that capture behavior changes due to the stay-at-home order. Then, we adapted and applied an existing algorithm to these behavior-change features to predict the presence of depression, high global MS symptom burden, severe fatigue, and poor sleep quality during the stay-at-home period.

Results

Using data collected between November 2019 and May 2020, the algorithm detected depression with an accuracy of 82.5% (65% improvement over baseline; F₁-score: 0.84), high global MS symptom burden with an accuracy of 90% (39% improvement over baseline; F₁-score: 0.93), severe fatigue with an accuracy of 75.5% (22% improvement over baseline; F₁-score: 0.80), and poor sleep quality with an accuracy of 84% (28% improvement over baseline; F₁-score: 0.84).

Conclusions

Our approach could help clinicians better triage patients with MS and potentially other chronic neurological disorders for interventions and aid patient self-monitoring in their own environment, particularly during extraordinarily stressful circumstances such as pandemics, which would cause drastic behavior changes.

SUBMITTER: Chikersal P

PROVIDER: S-EPMC9407162 | biostudies-literature | 2022 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Multiple Sclerosis Outcomes During the COVID-19 Stay-at-home Period: Observational Study Using Passively Sensed Behaviors and Digital Phenotyping.

Chikersal Prerna P Venkatesh Shruthi S Masown Karman K Walker Elizabeth E Quraishi Danyal D Dey Anind A Goel Mayank M Xia Zongqi Z

JMIR mental health 20220824 8

<h4>Background</h4>The COVID-19 pandemic has broad negative impact on the physical and mental health of people with chronic neurological disorders such as multiple sclerosis (MS).<h4>Objective</h4>We presented a machine learning approach leveraging passive sensor data from smartphones and fitness trackers of people with MS to predict their health outcomes in a natural experiment during a state-mandated stay-at-home period due to a global pandemic.<h4>Methods</h4>First, we extracted features that ...[more]

PMID: 35849686

Similar Datasets

Project description:BackgroundLongitudinal tracking of multiple sclerosis (MS) symptoms in an individual's own environment may improve self-monitoring and clinical management for people with MS (pwMS).ObjectiveWe present a machine learning approach that enables longitudinal monitoring of clinically relevant patient-reported symptoms for pwMS by harnessing passively collected data from sensors in smartphones and fitness trackers.MethodsWe divide the collected data into discrete periods for each patient. For each prediction period, we first extract patient-level behavioral features from the current period (action features) and the previous period (context features). Then, we apply a machine learning (ML) approach based on Support Vector Machine with Radial Bias Function Kernel and AdaBoost to predict the presence of depressive symptoms (every two weeks) and high global MS symptom burden, severe fatigue, and poor sleep quality (every four weeks).ResultsBetween November 16, 2019, and January 24, 2021, 104 pwMS (84.6% women, 93.3% non-Hispanic White, 44.0±11.8 years mean±SD age) from a clinic-based MS cohort completed 12-weeks of data collection, including a subset of 44 pwMS (88.6% women, 95.5% non-Hispanic White, 45.7±11.2 years) who completed 24-weeks of data collection. In total, we collected approximately 12,500 days of passive sensor and behavioral health data from the participants. Among the best-performing models with the least sensor data requirement, ML algorithm predicts depressive symptoms with an accuracy of 80.6% (35.5% improvement over baseline; F1-score: 0.76), high global MS symptom burden with an accuracy of 77.3% (51.3% improvement over baseline; F1-score: 0.77), severe fatigue with an accuracy of 73.8% (45.0% improvement over baseline; F1-score: 0.74), and poor sleep quality with an accuracy of 72.0% (28.1% improvement over baseline; F1-score: 0.70). Further, sensor data were largely sufficient for predicting symptom severity, while the prediction of depressive symptoms benefited from minimal active patient input in the form of response to two brief questions on the day before the prediction point.ConclusionsOur digital phenotyping approach using passive sensors on smartphones and fitness trackers may help patients with real-world, continuous, self-monitoring of common symptoms in their own environment and assist clinicians with better triage of patient needs for timely interventions in MS (and potentially other chronic neurological disorders).

Project description:IntroductionStay-at-home orders during the COVID-19 pandemic decreased population mobility to reduce SARS-CoV-2 infection rates. We empirically tested the hypothesis that this public health measure was associated with a higher likelihood of opioid- and stimulant-involved deaths occurring in homes located in Cook County, Illinois.MethodsThe stay-at-home period was from March 21, 2020 to May 30, 2020. We analyzed overdose data from the Cook County Medical Examiner's Office using a death location description from case investigations categorized as home, medical, motel, scene, and other. Two groups of decedents were defined as either having an opioid or stimulant listed in the primary cause of death field. We modeled a weekly time series to detect changes in deaths (number) and trends during segmented time periods. Chi-square or Fisher's exact and adjusted logistic regression was used for testing the differences between the stay-at-home and a 13-week preceding period.ResultsThere were 4,169 and 2,012 opioid- and stimulant-involved deaths, respectively, from 2018 to 2020. Both groups were demographically similar: 75% male, 52% White, and aged 45 years (mean). In the 13 weeks before stay-at-home orders, 51% of opioid-involved deaths occurred in homes, which increased to 59% (p<0.0001) during the 10 weeks of the order and decreased back to 51% in the 18 weeks after the order expired. For stimulant-involved deaths, 51% were residential immediately before the orders, with a nonsignificant increase to 52% during the stay-at-home period. Before the pandemic, there were 20 deaths/week, increasing to 37 deaths/week (p<0.0001) during stay-at-home enactment. Deaths involving fentanyl among the opioid-involved group increased from 76% to 89%, whereas those involving heroin decreased from 55% to 37%. The adjusted OR for opioid-involved fatal overdoses occurring at home during this period compared with that occurring the 13 weeks before was 1.37 (95% CI=1.05, 1.79).ConclusionsThe likelihood of a death occurring at home, especially for people using opioids, increased during the stay-at-home order period. Findings have implications for mitigating overdose risks during social isolation.

Project description:BackgroundDental disease (including dental caries and periodontal disease) is largely preventable and closely linked to inadequate oral health behaviors. Digital health technologies have great potential for unobtrusively monitoring brushing behaviors in home settings and promoting optimal oral self-care routines at scale.ObjectiveThe aim of this study is to leverage the ubiquity of electronic toothbrushes and smartphones with the development of a Remote Oral Behaviors Assessment System (ROBAS) and evaluate its feasibility for passively tracking brushing behaviors in real-world settings.MethodsWe developed ROBAS by linking inertial sensors contained within consumer electronic toothbrushes to a scalable software platform comprised of a smartphone app linked to a cloud platform. First, the criterion validity of ROBAS for accurately capturing brushing details was established in a laboratory setting. Next, real-world performance and usability were evaluated in a stratified community sample of 32 participants who used ROBAS daily for 1 month and maintained a diary of their brushing episodes. Semistructured interviews at baseline and exit captured the user experience. We used regression models and Bland-Altman analyses to assess the criterion validity, functionality, accuracy, and consistency of ROBAS.ResultsUsing a stopwatch as the criterion reference, ROBAS showed a mean absolute percent error (MAPE) of 1.8%, an estimated bias of 0.64 seconds that was not statistically distinguishable from zero (95% CI -0.93 to 2.22 seconds, SE 0.79), and a connection failure rate of 6.7% (95% CI 0.8%-22.1%, SE 4.6%). In real-world testing, ROBAS showed close agreement with the daily diary recordings of brushing episodes; estimated average discrepancies between the diary and ROBAS were 0.13 sessions per day (95% CI 0.01-0.26, SE 0.06), 8.0 seconds per brushing session (95% CI 1.4-14.7, SE 3.3), and 30 seconds of brushing per day (95% CI 9.3-50.1, SE 10.0). Retrospective self-reports produced substantially higher estimates of brushing frequency and duration compared to ROBAS measurements. Participants reported ROBAS was easy to use and expressed an interest in receiving ROBAS-delivered feedback on their brushing behaviors. Most participants were bothered by the use of an additional study phone, and some reported connectivity-related issues.ConclusionsROBAS has a high criterion validity for measuring oral health behaviors. It can accurately and reliably monitor brushing patterns in home settings for extended periods. Unobtrusive data collection through ROBAS sets the stage for automated coaching and optimization of oral self-care practices at the individual and population level.

Project description:BackgroundMental health disorders affect multiple aspects of patients' lives, including mood, cognition, and behavior. eHealth and mobile health (mHealth) technologies enable rich sets of information to be collected noninvasively, representing a promising opportunity to construct behavioral markers of mental health. Combining such data with self-reported information about psychological symptoms may provide a more comprehensive and contextualized view of a patient's mental state than questionnaire data alone. However, mobile sensed data are usually noisy and incomplete, with significant amounts of missing observations. Therefore, recognizing the clinical potential of mHealth tools depends critically on developing methods to cope with such data issues.ObjectiveThis study aims to present a machine learning-based approach for emotional state prediction that uses passively collected data from mobile phones and wearable devices and self-reported emotions. The proposed methods must cope with high-dimensional and heterogeneous time-series data with a large percentage of missing observations.MethodsPassively sensed behavior and self-reported emotional state data from a cohort of 943 individuals (outpatients recruited from community clinics) were available for analysis. All patients had at least 30 days' worth of naturally occurring behavior observations, including information about physical activity, geolocation, sleep, and smartphone app use. These regularly sampled but frequently missing and heterogeneous time series were analyzed with the following probabilistic latent variable models for data averaging and feature extraction: mixture model (MM) and hidden Markov model (HMM). The extracted features were then combined with a classifier to predict emotional state. A variety of classical machine learning methods and recurrent neural networks were compared. Finally, a personalized Bayesian model was proposed to improve performance by considering the individual differences in the data and applying a different classifier bias term for each patient.ResultsProbabilistic generative models proved to be good preprocessing and feature extractor tools for data with large percentages of missing observations. Models that took into account the posterior probabilities of the MM and HMM latent states outperformed those that did not by more than 20%, suggesting that the underlying behavioral patterns identified were meaningful for individuals' overall emotional state. The best performing generalized models achieved a 0.81 area under the curve of the receiver operating characteristic and 0.71 area under the precision-recall curve when predicting self-reported emotional valence from behavior in held-out test data. Moreover, the proposed personalized models demonstrated that accounting for individual differences through a simple hierarchical model can substantially improve emotional state prediction performance without relying on previous days' data.ConclusionsThese findings demonstrate the feasibility of designing machine learning models for predicting emotional states from mobile sensing data capable of dealing with heterogeneous data with large numbers of missing observations. Such models may represent valuable tools for clinicians to monitor patients' mood states.

Project description:BackgroundDigital phenotyping provides insights into an individual's digital behaviors and has potential clinical utility.ObjectiveIn this observational study, we explored digital biomarkers collected from wrist-wearable devices and smartphones and their associations with clinical symptoms and functioning in patients with schizophrenia.MethodsWe recruited 100 outpatients with schizophrenia spectrum disorder, and we collected various digital data from commercially available wrist wearables and smartphones over a 6-month period. In this report, we analyzed the first week of digital data on heart rate, sleep, and physical activity from the wrist wearables and travel distance, sociability, touchscreen tapping speed, and screen time from the smartphones. We analyzed the relationships between these digital measures and patient baseline measurements of clinical symptoms assessed with the Positive and Negative Syndrome Scale, Brief Negative Symptoms Scale, and Calgary Depression Scale for Schizophrenia, as well as functioning as assessed with the Social and Occupational Functioning Assessment Scale. Linear regression was performed for each digital and clinical measure independently, with the digital measures being treated as predictors.ResultsDigital data were successfully collected from both the wearables and smartphones throughout the study, with 91% of the total possible data successfully collected from the wearables and 82% from the smartphones during the first week of the trial-the period under analysis in this report. Among the clinical outcomes, negative symptoms were associated with the greatest number of digital measures (10 of the 12 studied here), followed by overall measures of psychopathology symptoms, functioning, and positive symptoms, which were each associated with at least 3 digital measures. Cognition and cognitive/disorganization symptoms were each associated with 1 or 2 digital measures.ConclusionsWe found significant associations between nearly all digital measures and a wide range of symptoms and functioning in a community sample of individuals with schizophrenia. These findings provide insights into the digital behaviors of individuals with schizophrenia and highlight the potential of using commercially available wrist wearables and smartphones for passive monitoring in schizophrenia.

Project description:While studies show links between smartphone data and affective symptoms, we lack clarity on the temporal scale, specificity (e.g., to depression vs. anxiety), and person-specific (vs. group-level) nature of these associations. We conducted a large-scale (n = 1013) smartphone-based passive sensing study to identify within- and between-person digital markers of depression and anxiety symptoms over time. Participants (74.6% female; M age = 40.9) downloaded the LifeSense app, which facilitated continuous passive data collection (e.g., GPS, app and device use, communication) across 16 weeks. Hierarchical linear regression models tested the within- and between-person associations of 2-week windows of passively sensed data with depression (PHQ-8) or generalized anxiety (GAD-7). We used a shifting window to understand the time scale at which sensed features relate to mental health symptoms, predicting symptoms 2 weeks in the future (distal prediction), 1 week in the future (medial prediction), and 0 weeks in the future (proximal prediction). Spending more time at home relative to one's average was an early signal of PHQ-8 severity (distal β = 0.219, p = 0.012) and continued to relate to PHQ-8 at medial (β = 0.198, p = 0.022) and proximal (β = 0.183, p = 0.045) windows. In contrast, circadian movement was proximally related to (β = -0.131, p = 0.035) but did not predict (distal β = 0.034, p = 0.577; medial β = -0.089, p = 0.138) PHQ-8. Distinct communication features (i.e., call/text or app-based messaging) related to PHQ-8 and GAD-7. Findings have implications for identifying novel treatment targets, personalizing digital mental health interventions, and enhancing traditional patient-provider interactions. Certain features (e.g., circadian movement) may represent correlates but not true prospective indicators of affective symptoms. Conversely, other features like home duration may be such early signals of intra-individual symptom change, indicating the potential utility of prophylactic intervention (e.g., behavioral activation) in response to person-specific increases in these signals.

Dataset Information

Predicting Multiple Sclerosis Outcomes During the COVID-19 Stay-at-home Period: Observational Study Using Passively Sensed Behaviors and Digital Phenotyping.

Background

Objective

Methods

Results

Conclusions

Publications

Predicting Multiple Sclerosis Outcomes During the COVID-19 Stay-at-home Period: Observational Study Using Passively Sensed Behaviors and Digital Phenotyping.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets