Wearable Biosensing to Predict Imminent Aggressive Behavior in Psychiatric Inpatient Youths With Autism

Abstract

Importance: Aggressive behavior is a prevalent and challenging issue in individuals with autism.

Objective: To investigate whether changes in peripheral physiology recorded by a wearable biosensor and machine learning can be used to predict imminent aggressive behavior before it occurs in inpatient youths with autism.

Design, setting, and participants: This noninterventional prognostic study used data collected from March 2019 to March 2020 from 4 primary care psychiatric inpatient hospitals. Enrolled participants were 86 psychiatric inpatients with confirmed diagnoses of autism exhibiting operationally defined self-injurious behavior, emotion dysregulation, or aggression toward others; 16 individuals were not included (18.6%) because they would not wear the biosensor (8 individuals) or were discharged before an observation could be made (8 individuals). Data were analyzed from March 2020 through October 2023.

Main outcomes and measures: Research staff performed live behavioral coding of aggressive behavior while inpatient study participants wore a commercially available biosensor that recorded peripheral physiological signals (cardiovascular activity, electrodermal activity, and motion). Logistic regression, support vector machines, neural networks, and domain adaptation were used to analyze time-series features extracted from biosensor data. Area under the receiver operating characteristic curve (AUROC) values were used to evaluate the performance of population- and person-dependent models.

Results: There were 70 study participants (mean [range; SD] age, 11.9 [5-19; 3.5] years; 62 males [88.6%]; 1 Asian [1.4%], 5 Black [7.1%], 1 Native Hawaiian or Other Pacific Islander [1.4%], and 63 White [90.0%]; 5 Hispanic [7.5%] and 62 non-Hispanic [92.5%] among 67 individuals with ethnicity data). Nearly half of the population (32 individuals [45.7%]) was minimally verbal, and 30 individuals (42.8%) had an intellectual disability. Participant length of inpatient hospital stay ranged from 8 to 201 days, and the mean (SD) length was 37.28 (33.95) days. A total of 429 naturalistic observational coding sessions were recorded, totaling 497 hours, wherein 6665 aggressive behaviors were documented, including self-injury (3983 behaviors [59.8%]), emotion dysregulation (2063 behaviors [31.0%]), and aggression toward others (619 behaviors [9.3%]). Logistic regression was the best-performing overall classifier across all experiments; for example, it predicted aggressive behavior 3 minutes before onset with a mean AUROC of 0.80 (95% CI, 0.79-0.81).

Conclusions and relevance: This study replicated and extended previous findings suggesting that machine learning analyses of preceding changes in peripheral physiology may be used to predict imminent aggressive behaviors before they occur in inpatient youths with autism. Further research will explore clinical implications and the potential for personalized interventions.

Figure 1.. Mean Area Under the Receiver Operating Characteristic Curves (AUROCs) Across Time Parameter Values

A, B, and C, Each marker represents the mean AUROC of 1 of 9 combinations of window lengths from the past (τ_p) and in the future (τ_f) (τ_p, τ_f ∈ {60,120,180} seconds). A, Mean AUROCs for experiments 1, 2, 3, and 4 for 3 scenarios (feature vectors [FVs], offset, and onset) are presented. B, Mean AUROC values of multiclass population models (PMs) with leave-individuals-out (LIO) (experiment 5) and multiclass person-dependent models (PDMs) with leave-sessions-out (LIO) (experiment 6) are presented. C, Mean AUROCs by aggressive behavior motion intensity cluster (experiment 7) are presented. D, Median AUROC changes after domain adaptation (DA) across data split settings are presented. Comparisons are made between PM performance on individual participant data before DA. AFV indicates augmented FV; ATO, aggression toward others; ED, emotion dysregulation; LR, logistic regression; NN, neural network; SIB, self-injurious behavior; SS, session split; SVM, support vector machine.

Figure 2.. Aggressive Behavior Properties and Area Under the Receiver Operating Characteristic Curves (AUROCs)

A, The AUROC by the number of participant observational sessions is presented. B, The AUROC by the number of aggressive behavior episodes is presented. C, The AUROC by the mean aggressive behavior episode duration is presented. D, The AUROC by the observation duration is presented. LR indicates logistic regression; NN, neural network; SMV, support vector machine.