Perceived and calculated diet quality improvements in a randomized mHealth weight loss trial.

The diet quality of US adults is poor and cross-sectional analyses suggest self-perception of healthful dietary intake may be overestimated. This analysis assessed the concordance between calculated and perceived diet quality and changes in diet quality among adults seeking weight loss and enrolled in a 12-month randomized behavioral trial. Healthy Eating Index-2015 diet quality (HEI) was calculated from self-administered 24-hour recalls. Perceived diet quality (PDQ) was measured on a 100-point scale. Higher scores indicate better diet quality. Concordance was assessed using the concordance correlation coefficient and Bland-Altman plots. The one hundred and five participants with complete dietary data were mostly female and white. There was good agreement between HEI and PDQ scores at 12 months for less than a third of participants. Most of the disagreement arose from PDQ scores being higher than HEI scores. Even fewer participants had good agreement between HEI changes and PDQ changes. Participants perceived greater improvement in diet quality than indicated by HEI score changes. Concordance was low at 12 months and for change in diet quality. Despite the diet quality of adults seeking weight loss being suboptimal and not improving, many perceived their diet quality and diet quality improvements as better than calculated. Future studies might explore the effect of misperceptions on weight loss outcomes.

Effect of an mHealth weight loss intervention on Healthy Eating Index diet quality: the SMARTER randomised controlled trial.

In the few weight loss studies assessing diet quality, improvements have been minimal and recommended calculation methods have not been used. This secondary analysis of a parallel group randomised trial (regsitered: https://clinicaltrials.gov/ct2/show/NCT03367936) assessed whether self-monitoring with feedback (SM + FB) v. self-monitoring alone (SM) improved diet quality. Adults with overweight/obesity (randomised: SM n 251, SM + FB n 251; analysed SM n 170, SM + FB n 186) self-monitored diet, physical activity and weight. Real-time, personalised feedback, delivered via a study-specific app up to three times daily, was based on reported energy, fat and added sugar intake. Healthy Eating Index 2015 (HEI-2015) scores were calculated from 24-hour recalls. Higher scores represent better diet quality. Data were collected August 2018 to March 2021 and analysed spring 2022. The sample was mostly female (78·9 %) and white (85·4 %). At baseline, HEI-2015 total scores and bootstrapped 95 % CI were similar by treatment group (SM + FB: 63·11 (60·41, 65·24); SM: 61·02 (58·72, 62·81)) with similar minimal improvement observed at 6 months (SM + FB: 65·42 (63·30, 67·20); SM: 63·19 (61·22, 64·97)) and 12 months (SM + FB: 63·94 (61·40, 66·29); SM: 63·56 (60·81, 65·42)). Among those who lost ≥ 5 % of baseline weight, HEI-2015 scores improved (baseline: 62·00 (58·94, 64·12); 6 months: 68·02 (65·41, 71·23); 12 months: 65·93 (63·40, 68·61)). There was no effect of the intervention on diet quality change. Clinically meaningful weight loss was related to diet quality improvement. Feedback may need to incorporate more targeted nutritional content.

The Effect of Tailored, Daily, Smartphone Feedback to Lifestyle Self-Monitoring on Weight Loss at 12 Months: the SMARTER Randomized Clinical Trial.

BACKGROUND: Self-monitoring (SM) is the centerpiece of behavioral weight loss treatment, but the efficacy of smartphone-delivered SM feedback (FB) has not been tested in large, long-term, randomized trials. OBJECTIVE: The aim of this study was to establish the efficacy of providing remote FB to diet, physical activity (PA), and weight SM on improving weight loss outcomes when comparing the SM plus FB (SM+FB) condition to the SM-only condition in a 12-month randomized controlled trial. The study was a single-site, population-based trial that took place in southwestern Pennsylvania, USA, conducted between 2018 and 2021. Participants were smartphone users age ≥18 years, able to engage in moderate PA, with a mean BMI between 27 and 43 kg/m2. METHODS: All participants received a 90-minute, one-to-one, in-person behavioral weight loss counseling session addressing behavioral strategies, establishing participants' dietary and PA goals, and instructing on use of the PA tracker (Fitbit Charge 2), smart scale, and diet SM app. Only SM+FB participants had access to an investigator-developed smartphone app that read SM data, in which an algorithm selected tailored messages sent to the smartphone up to 3 times daily. The SM-only participants did not receive any tailored FB based on SM data. The primary outcome was percent weight change from baseline to 12 months. Secondary outcomes included engagement with digital tools (eg, monthly percentage of FB messages opened and monthly percentage of days adherent to the calorie goal). RESULTS: Participants (N=502) were on average 45.0 (SD 14.4) years old with a mean BMI of 33.7 (SD 4.0) kg/m2. The sample was 79.5% female (n=399/502) and 82.5% White (n=414/502). At 12 months, retention was 78.5% (n=394/502) and similar by group (SM+FB: 202/251, 80.5%; SM: 192/251, 76.5%; P=.28). There was significant percent weight loss from baseline in both groups (SM+FB: -2.12%, 95% CI -3.04% to -1.21%, P<.001; SM: -2.39%, 95% CI -3.32% to -1.47%; P<.001), but no difference between the groups (-0.27%; 95% CI -1.57% to 1.03%; t =-0.41; P=.68). Similarly, 26.3% (66/251) of the SM+FB group and 29.1% (73/251) of the SM group achieved ≥5% weight loss (chi-square value=0.49; P=.49). A 1% increase in FB messages opened was associated with a 0.10 greater percent weight loss at 12 months (b=-0.10; 95% CI -0.13 to -0.07; t =-5.90; P<.001). A 1% increase in FB messages opened was associated with 0.12 greater percentage of days adherent to the calorie goal per month (b=0.12; 95% CI 0.07-0.17; F=22.19; P<.001). CONCLUSIONS: There were no significant between-group differences in weight loss; however, the findings suggested that the use of commercially available digital SM tools with or without FB resulted in a clinically significant weight loss in over 25% of participants. Future studies need to test additional strategies that will promote greater engagement with digital tools. TRIAL REGISTRATION: Clinicaltrials.gov NCT03367936; https://clinicaltrials.gov/ct2/show/NCT03367936.

Development and Validation of a Comprehensive Well-Being Scale for People in the University Environment (Pitt Wellness Scale) Using a Crowdsourcing Approach: Cross-Sectional Study.

BACKGROUND: Well-being has multiple domains, and these domains are unique to the population being examined. Therefore, to precisely assess the well-being of a population, a scale specifically designed for that population is needed. OBJECTIVE: The goal of this study was to design and validate a comprehensive well-being scale for people in a university environment, including students, faculty, and staff. METHODS: A crowdsourcing approach was used to determine relevant domains for the comprehensive well-being scale in this population and identify specific questions to include in each domain. A web-based questionnaire (Q1) was used to collect opinions from a group of university students, faculty, and staff about the domains and subdomains of the scale. A draft of a new well-being scale (Q2) was created in response to the information collected via Q1, and a second group of study participants was invited to evaluate the relevance and clarity of each statement. A newly created well-being scale (Q3) was then used by a third group of university students, faculty, and staff. A psychometric analysis was performed on the data collected via Q3 to determine the validity and reliability of the well-being scale. RESULTS: In the first step, a group of 518 university community members (students, faculty, and staff) indicated the domains and subdomains that they desired to have in a comprehensive well-being scale. In the second step, a second group of 167 students, faculty, and staff evaluated the relevance and clarity of the proposed statements in each domain. In the third step, a third group of 546 students, faculty, and staff provided their responses to the new well-being scale (Pitt Wellness Scale). The psychometric analysis indicated that the reliability of the well-being scale was high. CONCLUSIONS: Using a crowdsourcing approach, we successfully created a comprehensive and highly reliable well-being scale for people in the university environment. Our new Pitt Wellness Scale may be used to measure the well-being of people in the university environment.

Systematic Review of Mobile Health Applications in Rehabilitation.

OBJECTIVE: To conduct systematic review to better define how medical mobile applications (apps) have been used in environments relevant to physical medicine and rehabilitation. DATA SOURCES: PUBMED, IEEE, ACM Digital Library, SCOPUS, INSPEC, and EMBASE. STUDY SELECTION: A 10-year date limit was used, spanning publication dates from June 1, 2006, to June 30, 2016. Terms related to physical medicine and rehabilitation as well as mobile apps were used in 10 individual search strategies. DATA EXTRACTION: Two investigators screened abstracts and applied inclusion and exclusion criteria. Full-length articles were retrieved. Duplicate articles were removed. If a study met all criteria, the article was reviewed in full. DATA SYNTHESIS: Specific variables of interest were extracted and added to summary tables. Summary tables were used to categorize studies according themes, and a list of app features was generated. CONCLUSIONS: The search yielded abstracts from 8116 studies, and 102 studies were included in the systematic review. Approximately one-third of the studies evaluated apps as interventions, and the remaining two-thirds of the studies assessed functioning of the app or participant interaction with the app. Some apps may have positive benefits when used to deliver exercise or gait training interventions, as self-management systems, or as measurement tools. REGISTRATION: The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) network (no. CRD42016046672).

Reaching People With Disabilities in Underserved Areas Through Digital Interventions: Systematic Review.

BACKGROUND: People with disabilities need rehabilitation interventions to improve their physical functioning, mental status, and quality of life. Many rehabilitation interventions can be delivered electronically ("digitally") via telehealth systems. For people with disabilities in underserved areas, electronically delivered rehabilitation interventions may be the only feasible service available for them. OBJECTIVE: The objective of this study was to evaluate the current status of digital interventions for people with disabilities in remote and underserved areas. METHODS: A systematic review was conducted on this topic. Keyword searches in multiple databases (PubMed, CINAHL, and Inspec) were performed to collect articles published in this field. The obtained articles were selected based on our selection criteria. Of the 198 identified articles, 16 duplicates were removed. After a review of the titles and abstracts of the remaining articles, 165 were determined to be irrelevant to this study and were therefore removed. The full texts of the remaining 17 articles were reviewed, and 6 of these articles were removed as being irrelevant to this study. The 11 articles remaining were discussed and summarized by 2 reviewers. RESULTS: These 11 studies cover a few types of disabilities, such as developmental disabilities and mobility impairments as well as several types of disability-causing disorders such as stroke, multiple sclerosis, traumatic brain injury, and facio-scapulo-humeral muscular dystrophy. Most of these studies were small-scale case studies and relatively larger-scale cohort studies; the project evaluation methods were mainly pre-post comparison, questionnaires, and interviews. A few studies also performed objective assessment of functional improvement. The intervention technology was mainly videoconferencing. Moreover, 10 of these studies were for people with disabilities in rural areas and 1 was for people in urban communities. CONCLUSIONS: A small number of small-scale studies have been conducted on digital interventions for people with disabilities in underserved areas. Although the results reported in these studies were mostly positive, they are not sufficient to prove the effectiveness of telehealth-based digital intervention in improving the situation among people with disabilities because of the small sample sizes and lack of randomized controlled trials.

The Effect of the Interactive Mobile Health and Rehabilitation System on Health and Psychosocial Outcomes in Spinal Cord Injury: Randomized Controlled Trial.

BACKGROUND: Individuals with spinal cord injury (SCI) are at risk for secondary medical complications, such as urinary tract infections (UTIs) and pressure injuries, that could potentially be mitigated through improved self-management techniques. The Interactive Mobile Health and Rehabilitation (iMHere) mobile health (mHealth) system was developed to support self-management for individuals with disabilities. OBJECTIVE: The main objective of this study was to determine if the use of iMHere would be associated with improved health outcomes over a 9-month period. A secondary objective was to determine if the use of iMHere would be associated with improved psychosocial outcomes. Phone usage, app usage, and training time data were also collected to analyze trends in iMHere use. METHODS: Overall, 38 participants with SCI were randomized into either the intervention group who used the iMHere system and received standard care or the control group who received standard care without any technology intervention. Health outcomes were recorded for the year before entry into the study and during the 9 months of the study. Participants completed surveys at baseline and every 3 months to measure psychosocial outcomes. RESULTS: The intervention group had a statistically significant reduction in UTIs (0.47 events per person; P=.03; number needed to treat=2.11). Although no psychosocial outcomes changed significantly, there was a nonsignificant trend toward a reduction in mood symptoms in the intervention group compared with the control group meeting the threshold for clinical significance. Approximately 34 min per participant per month were needed on average to manage the system and provide technical support through this mHealth system. CONCLUSIONS: The use of the iMHere mHealth system may be a valuable tool in the prevention of UTIs or reductions in depressive symptoms. Given these findings, iMHere has potential scalability for larger populations. TRIAL REGISTRATION: ClinicalTrials.gov NCT02592291; https://clinicaltrials.gov/ct2/show/NCT02592291.

Clinical Feasibility of a Just-in-Time Adaptive Intervention App (iREST) as a Behavioral Sleep Treatment in a Military Population: Feasibility Comparative Effectiveness Study.

BACKGROUND: Although evidence-based cognitive behavioral sleep treatments have been shown to be safe and effective, these treatments have limited scalability. Mobile health tools can address this scalability challenge. iREST, or interactive Resilience Enhancing Sleep Tactics, is a mobile health platform designed to provide a just-in-time adaptive intervention (JITAI) in the assessment, monitoring, and delivery of evidence-based sleep recommendations in a scalable and personalized manner. The platform includes a mobile phone-based patient app linked to a clinician portal. OBJECTIVE: The first aim of the pilot study was to evaluate the effectiveness of JITAI using the iREST platform for delivering evidence-based sleep interventions in a sample of military service members and veterans. The second aim was to explore the potential effectiveness of this treatment delivery form relative to habitual in-person delivery. METHODS: In this pilot study, military service members and veterans between the ages of 18 and 60 years who reported clinically significant service-related sleep disturbances were enrolled as participants. Participants were asked to use iREST for a period of 4 to 6 weeks during which time they completed a daily sleep/wake diary. Through the clinician portal, trained clinicians offered recommendations consistent with evidence-based behavioral sleep treatments on weeks 2 through 4. To explore potential effectiveness, self-report measures were used, including the Insomnia Severity Index (ISI), the Pittsburgh Sleep Quality Index (PSQI), and the PSQI Addendum for Posttraumatic Stress Disorder. RESULTS: A total of 27 participants completed the posttreatment assessments. Between pre- and postintervention, clinically and statistically significant improvements in primary and secondary outcomes were detected (eg, a mean reduction on the ISI of 9.96, t26=9.99, P<.001). At posttreatment, 70% (19/27) of participants met the criteria for treatment response and 59% (16/27) achieved remission. Comparing these response and remission rates with previously published results for in-person trials showed no significant differences. CONCLUSION: Participants who received evidence-based recommendations from their assigned clinicians through the iREST platform showed clinically significant improvements in insomnia severity, overall sleep quality, and disruptive nocturnal disturbances. These findings are promising, and a larger noninferiority clinical trial is warranted.

Pilot feasibility of an mHealth system for conducting ecological momentary assessment of mood-related symptoms following traumatic brain injury.

OBJECTIVE: This study assessed pilot feasibility and validity of a mobile health (mHealth) system for tracking mood-related symptoms after traumatic brain injury (TBI). DESIGN: A prospective, repeated measures design was used to assess compliance with daily ecological momentary assessments (EMA) conducted via a smartphone application over an 8-week period. METHODS: An mHealth system was developed specifically for individuals with TBI and utilized previously validated tools for depressive and anxiety symptoms (Patient Health Questionnaire-9, Generalized Anxiety Disorder-7). Feasibility was assessed in 20 community-dwelling adults with TBI via an assessment of compliance, satisfaction and usability of the smartphone applications. The authors also developed and implemented a clinical patient safety management mechanism for those endorsing suicidality. RESULTS: Participants correctly completed 73.4% of all scheduled assessments, demonstrating good compliance. Daily assessments took <2 minutes to complete. Participants reported high satisfaction with smartphone applications (6.3 of 7) and found them easy to use (6.2 of 7). Comparison of assessments obtained via telephone-based interview and EMA demonstrated high correlations (r = 0.81-0.97), supporting the validity of conducting these assessments via smartphone application in this population. CONCLUSIONS: EMA conducted via smartphone demonstrates initial feasibility among adults with TBI and presents numerous opportunities for long-term monitoring of mood-related symptoms in real-world settings.

Development of mHealth system for supporting self-management and remote consultation of skincare.

BACKGROUND: Individuals with spina bifida (SB) are vulnerable to chronic skin complications such as wounds on the buttocks and lower extremities. Most of these complications can be prevented with adherence to self-care routines. We have developed a mobile health (mHealth) system for supporting self-care and management of skin problems called SkinCare as part of an mHealth suite called iMHere (interactive Mobile Health and Rehabilitation). The objective of this research is to develop an innovative mHealth system to support self-skincare tasks, skin condition monitoring, adherence to self-care regimens, skincare consultation, and secure two-way communications between patients and clinicians. METHODS: In order to support self-skincare tasks, the SkinCare app requires three main functions: (1) self-care task schedule and reminders, (2) skin condition monitoring and communications that include imaging, information about the skin problem, and consultation with clinician, and (3) secure two-way messaging between the patient and clinician (wellness coordinator). The SkinCare system we have developed consists of the SkinCare app, a clinician portal, and a two-way communication protocol connecting the two. The SkinCare system is one component of a more comprehensive system to support a wellness program for individuals with SB. RESULTS: The SkinCare app has several features that include reminders to perform daily skin checks as well as the ability to report skin breakdown and injury, which uses a combination of skin images and descriptions. The SkinCare app provides reminders to visually inspect one's skin as a preventative measure, often termed a "skin check." The data is sent to the portal where clinicians can monitor patients' conditions. Using the two-way communication, clinicians can receive pictures of the skin conditions, track progress in healing over time, and provide instructions for how to best care for the wound. CONCLUSIONS: The system was capable of supporting self-care and adherence to regimen, monitoring adherence, and supporting clinician engagement with patients, as well as testing its feasibility in a long-term implementation. The study shows the feasibility of a long-term implementation of skincare mHealth systems to support self-care and two-way interactions between patients and clinicians.

Usability and reliability of a remotely administered adult autism assessment, the autism diagnostic observation schedule (ADOS) module 4.

INTRODUCTION: The Autism Diagnostic Observation Schedule (ADOS) Module 4 is an autism assessment designed for verbally fluent adolescents and adults. Because of a shortage of available clinical expertise, it can be difficult for adults to receive a proper autism spectrum disorder (ASD) diagnostic assessment. A potential option to address this shortage is remote assessment. The objective of this study was to examine the feasibility, usability, and reliability of administering the ADOS Module 4 remotely using the Versatile and Integrated System for Telerehabilitation (VISYTER). MATERIALS AND METHODS: VISYTER consists of computer stations at the client site and clinician site for video communication and a Web portal for managing and coordinating the assessment process. Twenty-three adults with an ASD diagnosis participated in a within-subject crossover design study in which both a remote ADOS and a face-to-face ADOS were administered. After completing the remote ADOS, participants completed a satisfaction survey. RESULTS: Participant satisfaction with the remote ADOS delivery system was high. The kappa value was greater than 0.61 on 21 of 31 ADOS items. There was substantial agreement on ADOS classification (i.e., diagnosis) between assessments delivered face-to-face versus assessments delivered remotely (interclass coefficient=0.92). Non-agreement may have been due to outside factors or practice effect despite a washout period. CONCLUSIONS: The results of this study demonstrate that an autism assessment designed to be delivered face to face can be administered remotely using an integrated Web-based system with high levels of usability and reliability.

The SmartCAT: an m-health platform for ecological momentary intervention in child anxiety treatment.

INTRODUCTION: Cognitive behavioral therapy (CBT) for child anxiety, although efficacious, typically requires 16-20 weekly sessions with a therapist. Brief CBT (BCBT; eight sessions) for child anxiety is promising but may have less favorable outcomes owing to reduced session time. Mobile health (m-health) has the potential to improve BCBT efficacy by delivering ecological momentary intervention to engage youth in learning and practicing CBT skills in their everyday lives (in vivo). MATERIALS AND METHODS: We developed an m-health platform entitled SmartCAT (Smartphone-enhanced Child Anxiety Treatment). SmartCAT consists of (1) a smartphone application (app) that cues youth to use the CBT skills taught in sessions, (2) an online portal that allows therapists to monitor skill use, to send cues and treatment-related materials, to engage youth in real-time via secure messages, and to manage rewards, and (3) a communication protocol that allows real-time bidirectional exchange between the app and the portal. A pilot study with nine youth (9-14 years old) examined the platform's feasibility as an adjunct to BCBT. RESULTS: SmartCAT was found to be capable of supporting BCBT for child anxiety and received positive feedback from both therapists and youth patients. Patients rated the app as highly usable (mean=1.7 on a 1-7 scale, with 1="easy"). Patients completed 5.36 skills coach entries per session (standard deviation=1.95) and took an average of 3.14 min (standard deviation=0.98 min) to complete the entries. CONCLUSIONS: A smartphone app is feasible within CBT for child anxiety. Users found SmartCAT both acceptable and easy to use. Integrating an m-health platform within BCBT for anxious children may facilitate involvement in treatment and dissemination of effective procedures.

Development and Validation of a Mobile Phone Application for Measuring Knee Range of Motion.

Knee range of motion (ROM) is an important indicator of knee function. Outside the clinical setting, patients may not be able to accurately assess knee ROM, which may impair recovery following trauma or surgery. This study aims to validate a smartphone mobile application developed to measure knee ROM compared to visual and goniometer ROM measurements. A knee ROM Android mobile application was developed to measure knee ROM. Patients ≥ 18 years old presenting to an orthopaedic clinic with native knee complaints were approached to participate. Knee ROM was measured bilaterally by an arthroplasty-trained surgeon using (1) vision, (2) goniometer, and (3) the mobile application. Measurements were compared in flexion and extension using a one-way analysis of variance with post hoc Tukey test (alpha = 0.05). Eighty-four knee ROM measurements (40 left, 44 right) were obtained in 47 patients. Median Kellgren-Lawrence grade from available radiographs was grade 3. In flexion, mobile application (117.6 ± 14.7 degrees) measurements were not significantly different from visual (116.1 ± 13.6 degrees) or goniometer (116.2 ± 13.6 degrees) measurements. In extension, mobile application (4.8 ± 7.3 degrees) measurements were significantly different from visual (1.9 ± 4.1 degrees) measurements on post hoc analysis (p < 0.01), while no differences were present compared to goniometer (3.1 ± 5.8 degrees) measurements. Our study found that a mobile application for evaluating knee ROM was noninferior to goniometer-based measurements performed by an arthroplasty-trained surgeon. Future studies will investigate this application's utility in (1) remote patient care, (2) accelerating recovery during rehabilitation, (3) detecting early postoperative complications including arthrofibrosis, and (4) adding additional functionalities to the application to provide more detail-oriented descriptive analyses of patient knee function.

A Reliable and Accessible Caregiving Language Model (CaLM) to Support Tools for Caregivers: Development and Evaluation Study.

BACKGROUND: In the United States, 1 in 5 adults currently serves as a family caregiver for an individual with a serious illness or disability. Unlike professional caregivers, family caregivers often assume this role without formal preparation or training. Thus, there is an urgent need to enhance the capacity of family caregivers to provide quality care. Leveraging technology as an educational tool or an adjunct to care is a promising approach that has the potential to enhance the learning and caregiving capabilities of family caregivers. Large language models (LLMs) can potentially be used as a foundation technology for supporting caregivers. An LLM can be categorized as a foundation model (FM), which is a large-scale model trained on a broad data set that can be adapted to a range of different domain tasks. Despite their potential, FMs have the critical weakness of "hallucination," where the models generate information that can be misleading or inaccurate. Information reliability is essential when language models are deployed as front-line help tools for caregivers. OBJECTIVE: This study aimed to (1) develop a reliable caregiving language model (CaLM) by using FMs and a caregiving knowledge base, (2) develop an accessible CaLM using a small FM that requires fewer computing resources, and (3) evaluate the model's performance compared with a large FM. METHODS: We developed a CaLM using the retrieval augmented generation (RAG) framework combined with FM fine-tuning for improving the quality of FM answers by grounding the model on a caregiving knowledge base. The key components of the CaLM are the caregiving knowledge base, a fine-tuned FM, and a retriever module. We used 2 small FMs as candidates for the foundation of the CaLM (LLaMA [large language model Meta AI] 2 and Falcon with 7 billion parameters) and adopted a large FM (GPT-3.5 with an estimated 175 billion parameters) as a benchmark. We developed the caregiving knowledge base by gathering various types of documents from the internet. We focused on caregivers of individuals with Alzheimer disease and related dementias. We evaluated the models' performances using the benchmark metrics commonly used in evaluating language models and their reliability for providing accurate references with their answers. RESULTS: The RAG framework improved the performance of all FMs used in this study across all measures. As expected, the large FM performed better than the small FMs across all metrics. Interestingly, the small fine-tuned FMs with RAG performed significantly better than GPT 3.5 across all metrics. The fine-tuned LLaMA 2 with a small FM performed better than GPT 3.5 (even with RAG) in returning references with the answers. CONCLUSIONS: The study shows that a reliable and accessible CaLM can be developed using small FMs with a knowledge base specific to the caregiving domain.

Data Collection and Management of mHealth, Wearables, and Internet of Things in Digital Behavioral Health Interventions With the Awesome Data Acquisition Method (ADAM): Development of a Novel Informatics Architecture.

Unlabelled: The integration of health and activity data from various wearable devices into research studies presents technical and operational challenges. The Awesome Data Acquisition Method (ADAM) is a versatile, web-based system that was designed for integrating data from various sources and managing a large-scale multiphase research study. As a data collecting system, ADAM allows real-time data collection from wearable devices through the device's application programmable interface and the mobile app's adaptive real-time questionnaires. As a clinical trial management system, ADAM integrates clinical trial management processes and efficiently supports recruitment, screening, randomization, data tracking, data reporting, and data analysis during the entire research study process. We used a behavioral weight-loss intervention study (SMARTER trial) as a test case to evaluate the ADAM system. SMARTER was a randomized controlled trial that screened 1741 participants and enrolled 502 adults. As a result, the ADAM system was efficiently and successfully deployed to organize and manage the SMARTER trial. Moreover, with its versatile integration capability, the ADAM system made the necessary switch to fully remote assessments and tracking that are performed seamlessly and promptly when the COVID-19 pandemic ceased in-person contact. The remote-native features afforded by the ADAM system minimized the effects of the COVID-19 lockdown on the SMARTER trial. The success of SMARTER proved the comprehensiveness and efficiency of the ADAM system. Moreover, ADAM was designed to be generalizable and scalable to fit other studies with minimal editing, redevelopment, and customization. The ADAM system can benefit various behavioral interventions and different populations.

Generative AI in orthopedics: an explainable deep few-shot image augmentation pipeline for plain knee radiographs and Kellgren-Lawrence grading.

OBJECTIVES: Recently, deep learning medical image analysis in orthopedics has become highly active. However, progress has been restricted by the absence of large-scale and standardized ground-truth images. To the best of our knowledge, this study is the first to propose an innovative solution, namely a deep few-shot image augmentation pipeline, that addresses this challenge by synthetically generating knee radiographs for training downstream tasks, with a specific focus on knee osteoarthritis Kellgren-Lawrence (KL) grading. MATERIALS AND METHODS: This study leverages a deep few-shot image augmentation pipeline to generate synthetic knee radiographs. Despite the limited availability of training samples, we demonstrate the capability of our proposed computational strategy to produce high-fidelity plain knee radiographs and use them to successfully train a KL grade classifier. RESULTS: Our experimental results showcase the effectiveness of the proposed computational pipeline. The generated synthetic radiographs exhibit remarkable fidelity, evidenced by the achieved average Frechet Inception Distance (FID) score of 26.33 for KL grading and 22.538 for bilateral knee radiographs. For KL grading classification, the classifier achieved a test Cohen's Kappa and accuracy of 0.451 and 0.727, respectively. Our computational strategy also resulted in a publicly and freely available imaging dataset of 86 000 synthetic knee radiographs. CONCLUSIONS: Our approach demonstrates the capability to produce top-notch synthetic knee radiographs and use them for KL grading classification, even when working with a constrained training dataset. The results obtained emphasize the effectiveness of the pipeline in augmenting datasets for knee osteoarthritis research, opening doors for broader applications in orthopedics, medical image analysis, and AI-powered diagnosis.

An integrated telehealth system for remote administration of an adult autism assessment.

We developed a telehealth system to administer an autism assessment remotely. The remote assessment system integrates videoconferencing, stimuli presentation, recording, image and video presentation, and electronic assessment scoring into an intuitive software platform. This is an advancement over existing technologies used in telemental health, which currently require several devices. The number of children, adolescents, and adults with autism spectrum disorders (ASDs) has increased dramatically over the past 20 years and is expected to continue to increase in coming years. In general, there are not many clinicians trained in either the diagnosis or treatment of adults with ASD. Given the number of adults with autism in need, a remote assessment system can potentially provide a solution to the lack of trained clinicians. The goal is to make the remote assessment system as close to face-to-face assessment as possible, yet versatile enough to support deployment in underserved areas. The primary challenge to achieving this goal is that the assessment requires social interaction that appears natural and fluid, so the remote system needs to be able to support fluid natural interaction. For this study we developed components to support this type of interaction and integrated these components into a system capable of supporting the entire autistic assessment protocol. We then implemented the system and evaluated the system on real patients. The results suggest that we have achieved our goal in developing a system with high-quality interaction that is easy to use.

Pilot RCT examining feasibility and disability outcomes of a mobile health platform for strategy training in inpatient stroke rehabilitation (iADAPT).

BACKGROUND: Strategy training is an intervention that may reduce disability when delivered in inpatient rehabilitation following stroke. However, shorter lengths of stay and challenges with continuity of care following discharge results in difficulties in achieving adequate intervention dosage and carryover of training. OBJECTIVE: We examined whether strategy training using a mobile health platform (iADAPT) is feasible during inpatient stroke rehabilitation and following discharge. METHODS: In this RCT, participants were randomized to receive strategy training using either the iADAPT application (n = 16) or a workbook (n = 15). Participants in both groups received 7 in-person sessions during inpatient rehabilitation and 3 remote sessions following discharge. We calculated descriptive statistics to examine acceptance, attendance, and adherence, and within-group effect sizes on satisfaction and disability. RESULTS: Participants in the iADAPT group attended fewer total intervention sessions (n = 5.5, workbook n = 9.0) but attempted a similar number of goals (n = 7.6, workbook n = 8.2). Both groups reported similar satisfaction with in-person intervention (Treatment Expectancy: iADAPT d = 0.60, workbook d = 0.47; Patient Provider Connection: iADAPT d = 0.18, workbook d = 0.31), but the mobile health group reported greater satisfaction with remote intervention (Treatment Expectancy: iADAPT d = -0.91, workbook d = -0.97; Patient Provider Connection: iADAPT d = 0.85, workbook d = -1.80). . CONCLUSIONS: Considering these promising feasibility metrics and the benefits of mobile health, it is worth continuing to explore the efficacy of strategy training using a mobile health platform.

Formative Development of ClockWork for the Postpartum Period: A Theory-Based Intervention to Harness the Circadian Timing System to Address Cardiometabolic Health-Related Behaviors.

Individuals with body mass index (BMI) ≥ 25 kg/m2 before pregnancy have greater difficulty losing the weight gained during pregnancy, and this postpartum weight retention predicts higher risk for cardiometabolic disease. The postpartum period involves substantial disruptions in circadian rhythms, including rhythms related to eating, physical activity, sleep, and light/dark exposure, each of which are linked to obesity and cardiometabolic disease in non-pregnant adult humans and animals. We posit that a multi-component, circadian timing system-based behavioral intervention that uses digital tools-ClockWork-will be feasible and acceptable to postpartum individuals and help promote weight- and cardiometabolic health-related behaviors. We provide data from stakeholder interviews with postpartum individuals (pre-pregnancy BMI ≥ 25; n = 7), which were conducted to obtain feedback on and improve the relevance and utility of digital self-monitoring tools for health behaviors and weight during the postpartum period. Participants perceived the ClockWork intervention and digital monitoring app to be helpful for management of postpartum weight-related health behaviors. They provided specific recommendations for increasing the feasibility intervention goals and improving app features for monitoring behaviors. Personalized, easily accessible interventions are needed to promote gestational weight loss after delivery; addressing circadian behaviors is an essential component of such interventions. Future studies will evaluate the efficacy of the ClockWork intervention and associated digital tools for improving cardiometabolic health-related behaviors linked to the circadian timing system during the postpartum period.