Objective: Children and adolescents with intellectual and developmental disabilities (IDD), particularly those with autism spectrum disorder, are at increased risk of challenging behaviors such as self-injury, aggression, elopement, and property destruction. To mitigate these challenges, it is crucial to focus on early signs of distress that may lead to these behaviors. These early signs might not be visible to the human eye but could be detected by predictive machine learning (ML) models that utilizes real-time sensing. Current behavioral assessment practices lack such proactive predictive models. This study developed and pilot-tested real-time early agitation capture technology (REACT), a real-time multimodal ML model to detect early signs of distress, termed “agitations.” Integrating multimodal sensing, ML, and human expertise could make behavioral assessments for people with IDD safer and more efficient. Methods: We leveraged wearable technology to collect behavioral and physiological data from three children with IDD aged 6 to 9 years. The effectiveness of the REACT system was measured using F1 score, assessing its usefulness at the time of agitation to 20s prior. Results: The REACT system was able to detect agitations with an average F1 score of 78.69% at the time of agitation and 68.20% 20s prior. Conclusion: The findings support the use of the REACT model for real-time, proactive detection of agitations in children with IDD. This approach not only improves the accuracy of detecting distress signals that are imperceptible to the human eye but also increases the window for timely intervention before behavioral escalation, thereby enhancing safety, well-being, and inclusion for this vulnerable population. We believe that such technological support system will enhance user autonomy, self-advocacy, and self-determination.

Background: Expedient access to early intervention (EI) systems has been identified as a priority for children with developmental delays, identified disabilities, and other special health care needs. Despite the mandated availability of EI, it remains challenging for families to navigate referral processes and establish appropriate services. Such challenges disproportionately affect families from traditionally underserved communities. Mobile health apps can improve clinical outcomes, increase accessibility to health services, and promote adherence to health-related interventions. Though promising, the implementation of apps within routine care is in its infancy, with limited research examining the components of what makes an effective app or how to reach families most impacted by inequities in health care delivery. Objective: In study 1, we conducted focus groups to access a broad range of perspectives on the process of navigating the EI system, with the dual goals of identifying ways in which a patient-facing app might facilitate this process and identifying barriers to use with traditionally underrepresented and underserved groups. In study 2, focus group findings informed the development of a patient-facing app, which was subsequently tested with a pilot sample of 5 families. Methods: In study 1, the focus groups included 29 participants from 4 shareholder groups. Targeted sampling was used to recruit participants from traditionally underrepresented groups. Focus group questions sought information about barriers families experience as they navigate the EI system, ideal features of a patient-facing app designed to track family engagement with the EI system, and potential barriers. Focus group procedures were informed by the Consolidated Framework for Implementation Research framework. In study 2, a pilot app was developed. The app was tested with a sample of 5 families of young children involved in the EI system. Families provided information on app functionality and usability. Results: Qualitative analysis revealed a desire for increased communication and information about the process of accessing EI services, potential utility of an app for communication purposes, and clear recommendations for app features. Insights from focus groups were used to inform the development of the Family on Track app and related implementation supports. App features included survey customization, timing and delivery of prompts, and questions related to barriers and service satisfaction. Implementation supports include a visual guide for app installation, resources related to common family questions, and availability of study personnel to guide families through installation and provide ongoing support. Field testing provided preliminary information about app usability, including identifying future directions. Conclusions: The results of this study could support the development of a new way for the EI system to communicate and connect with families, provide families with a means to communicate satisfaction and frustration, and access the supports they need to be active participants in their child’s care.

Background: Apathy is common among older adults residing in long-term care (LTC) and impairs quality of life for both older adults and care providers. Few pharmacological remedies exist, and nonpharmacologic approaches that engage those with apathy require extensive personnel time. Thus, technological approaches have been encouraged, including virtual reality (VR) and socially assistive robots (SAR). Despite a growing interest in their use, input from older adults and staff is often absent in their design. Involving older adults in the development of interactive health technologies is necessary to enhance the functionality, usability, and likelihood of promoting the intended health outcomes. Objective: We aimed to design and evaluate SAR and nonimmersive VR (SAR-VR) activities for pairs of older adults that would encourage human-to-human interaction, an essential activity to mitigate apathy. Methods: We implemented a multistep, user-centered design. A humanoid and dog SAR were used in combination with nonimmersive VR activities for pairs of older adults. An interdisciplinary team of engineers, nurses, and physicians collaborated with older adults and staff to create 4 activity prototypes, 3 with the humanoid robot and 1 with the dog robot. A total of 14 older adults at 2 sites participated in the design and evaluation of the different components of the system throughout all stages. Site 1 participants were instrumental in the development, and Site 2 participants validated the prototype activities. Data were collected at each session via observations, interviews, and a 6-item questionnaire that rated their degree of comfort and confidence in (1) using the wands, (2) interacting with the robot, and (3) interacting with the nonimmersive VR environment using a 5-point Likert response. Additionally, 5 staff from Site 2 were recruited to evaluate the ease of setting up and running the system at 2 different sessions. After each session, the system setup and interface were refined based on their feedback. Results: A total of 4 of 6 older adults (mean age 85, SD 9.3 years; 2 male) at Site 1 completed field testing development, and 8 residents (mean age 80, SD 4.7 years; 2 male) at Site 2 completed field testing validation. Participant comfort and confidence increased significantly over successive iterations of the system across most categories (Site 1: Wilcoxon signed rank test P=.03; Site 2: Wilcoxon signed rank test P<.001). Additionally, 5 LTC staff members successfully set up the system with minimal cueing from the researchers, demonstrating the usability of the system for caregivers. Iterative design changes incorporated hardware, software, and activity domains. Conclusions: These initial results demonstrate that LTC older adults and staff are capable and critical to the development and implementation of SAR-VR activities. Future studies are needed to evaluate the feasibility of implementation and effectiveness in reducing apathy. Trial Registration: ClinicalTrials.gov NCT05178992; https://clinicaltrials.gov/study/NCT05178992

Physiological measurements involves observing variables that attribute to the normative functioning of human systems and subsystems directly or indirectly. The measurements can be used to detect affective states of a person with aims such as improving human-computer interactions. There are several methods of collecting physiological data, but wearable sensors are a common, non-invasive tool for accurate readings. However, valuable information is hard to extract from the raw physiological data, especially for affective state detection. Machine Learning techniques are used to detect the affective state of a person through labeled physiological data. A clear problem with using labeled data is creating accurate labels. An expert is needed to analyze a form of recording of participants and mark sections with different states such as stress and calm. While expensive, this method delivers a complete dataset with labeled data that can be used in any number of supervised algorithms. An interesting question arises from the expensive labeling: how can we reduce the cost while maintaining high accuracy? Semi-Supervised learning (SSL) is a potential solution to this problem. These algorithms allow for machine learning models to be trained with only a small subset of labeled data (unlike unsupervised which use no labels). They provide a way of avoiding expensive labeling. This paper compares a fully supervised algorithm to a SSL on the public WESAD (Wearable Stress and Affect Detection) Dataset for stress detection. This paper shows that Semi-Supervised algorithms are a viable method for inexpensive affective state detection systems with accurate results.