Journal of Medical Internet Research

Application of Digital Health Technologies in Scoliosis Rehabilitation: Systematic Review Based on the Technology Classification Framework

2026-08-28T17:00:18-04:00

Background: Digital health technologies are increasingly being used in scoliosis rehabilitation, but current evidence remains fragmented, and the effectiveness of these technologies across different categories, and their implementation in the real world remains unclear. Objective: We systematically evaluated the efficacy, implementation factors, and potential mechanisms of action of digital health technologies across various categories for scoliosis rehabilitation. Methods: We searched PubMed, IEEE Xplore, Embase, and Web of Science from inception to November 7, 2025, with an updated search on May 12, 2026. We included English-language controlled trials, cohort studies, and feasibility studies of digital interventions for any type of scoliosis that reported at least one quantitative outcome; nonoriginal publications, purely surgical studies, and those without extractable data were excluded. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool (RoB 2) and the Risk of Bias in Nonrandomized Studies of Interventions (ROBINS-I), and intervention reporting completeness was assessed using the Template for Intervention Description and Replication (TIDieR). Interventions were categorized into 5 technology types and further stratified by evidence maturity into 3 tiers. Evidence was synthesized using vote counting based on the direction of effect, with prespecified subgroup and sensitivity analyses, and certainty of evidence assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE). Results: Thirteen studies (574 patients) from 5 countries were included. Among the randomized trials, 14.3% (n=1) were at high risk of bias, and among the nonrandomized studies, 33.3% (n=2) were at serious risk of bias. Across the 4 outcome domains, clinical outcomes showed a generally favorable direction for Cobb angle and flexibility but inconsistent evidence for the angle of trunk rotation (ATR); functional outcomes were consistently favorable for respiratory function and postural control; patient-reported quality-of-life outcomes were predominantly favorable, whereas evidence for pain and body image was inconsistent; implementation outcomes showed the least favorable overall direction of effect, with adherence and dropout more closely tied to supervision intensity than technology category. The main findings remained robust in sensitivity analyses; no serious adverse events were reported, and the overall certainty of evidence was low. Conclusions: Digital health interventions showed a generally favorable direction of effect for spinal alignment, function, and quality of life; however, evidence certainty was low, and outcomes should be interpreted with caution. To our knowledge, this is the first review to integrate diverse digital interventions into a single framework of 5 technology categories and 3 maturity tiers and to synthesize effectiveness across technologies, unlike prior single-technology or single-function reviews. Its main contribution is to shift the focus from technology effectiveness toward sustained use and to identify human supervision as a key factor influencing implementation. Clinically, technologies may be selected by maturity tier and paired with an appropriate supervision strategy to extend access to rehabilitation. Larger, long-term multicenter trials are needed for confirmation. Trial Registration: PROSPERO CRD420251250074; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251250074

Machine Learning, Large Language Models, and Multimodal AI for Diagnosing Pediatric Rare Diseases: Scoping Review

2026-08-28T16:45:11-04:00

Background: Pediatric rare diseases often cause a prolonged diagnostic odyssey. AI, including machine learning, deep learning, large language models (LLMs), and multimodal systems, may support diagnosis, but these applications in children have not been systematically mapped. Objective: The aim of the study is to map diagnostic applications, data modalities, validation strategies, and evidence maturity of AI methods for pediatric rare diseases. Methods: We conducted a scoping review following Joanna Briggs Institute methodology and reported it according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). On June 26, 2026, we searched PubMed, Scopus, Web of Science Core Collection, Embase, China National Knowledge Infrastructure (CNKI), Wanfang Data, and the Cochrane Library for records published from January 1, 2015, through June 1, 2026. We additionally searched medRxiv and arXiv and hand-searched the reference lists of included studies and relevant reviews. Eligibility was defined using the population-concept-context framework: pediatric rare diseases, diagnostic AI, and any clinical or research setting. JZ and JL independently screened titles and abstracts and assessed potentially eligible full-text reports. JZ charted the data, and JL verified every field. Findings were synthesized descriptively according to disease focus, AI technology, input modality, diagnostic task, validation strategy, and evidence maturity. Results: Database searches identified 2557 records; 2063 remained after deduplication. Of 106 full-text reports assessed, 77 database studies and 4 studies from hand searching and preprint servers were included, yielding 81 studies. Studies were published from 2016 through 2026, with 55 of 81 (67.9%) published from 2024 through 2026. Using a mutually exclusive primary technology classification, classical machine learning accounted for 38 (46.9%) studies, facial AI for 18 (22.2%), deep learning for 15 (18.5%), LLMs for 6 (7.4%), and multimodal AI for 4 (4.9%). Electronic health records, claims, clinical text, or structured clinical vignettes were used in 25 (30.9%) studies, facial images in 17 (21%), and other medical imaging in 13 (16%). Evidence remained mainly retrospective and internally validated: 62 (76.5%) studies included a retrospective component and 76 (93.8%) reported internal validation, whereas 21 (25.9%) included external validation and 12 (14.8%) included a prospective component. Conclusions: Research on AI-assisted diagnosis of pediatric rare diseases has expanded rapidly, but evidence maturity has not kept pace. Most studies established technical feasibility rather than generalizable clinical benefit, and performance should be interpreted by task, inputs, reference standard, and validation design rather than used to rank technologies. Evidence for LLMs and multimodal AI remains limited. Future research should prioritize multicenter validation, reproducible task-specific benchmarks, prospective evaluation, and assessment of incremental clinical value. Trial Registration: PROSPERO CRD420261326146; https://www.crd.york.ac.uk/PROSPERO/view/CRD420261326146

National Survey on Implementation Patterns and Functional Needs for Hospital-Wide Glycemic Management Systems in the Chinese Mainland: Mixed Methods Study

2026-08-28T16:00:38-04:00

Background: Diabetes mellitus is a major global public health burden, and achieving optimal glycemic control remains challenging. Multidisciplinary, hospital-wide glycemic management systems (HGMS) have emerged as a promising strategy. In the Chinese mainland, the adoption of HGMS has accelerated; yet, substantial variations remain in system functionality, regional coverage, and management effectiveness across hospitals, underscoring the need for a national survey. Objective: This study aimed to survey the implementation patterns and functional needs for HGMS in the Chinese mainland. Methods: A mixed methods design was used, integrating a cross-sectional quantitative survey and qualitative interviews. From April 2024 to December 2024, health care professionals across 31 provincial-level administrative regions in the Chinese mainland were recruited. Data were collected using standardized questionnaires and purposive semistructured interviews to evaluate the adoption, functional demands, and satisfaction with HGMS. Quantitative data were analyzed using SPSS (version 27.0; IBM Corp). Differences in mean ratings across HGMS functions were evaluated using the Friedman test, with post hoc pairwise comparisons performed via the Wilcoxon signed-rank test and Bonferroni correction. Qualitative data underwent thematic analysis following Braun and Clarke’s framework, supported by NVivo (version 12; QSR International). Results: A total of 988 health care professionals from 265 hospitals participated. HGMS implementation across hospitals in the Chinese mainland exhibited pronounced heterogeneity: 12.45% remained at level 0 (handwritten glucose records), 47.92% at level 1 (manual data entry), and 21.51% at level 2 (department-level data sharing). Levels 3, 4, and 5 accounted for 13.21%, 3.40%, and 1.51%, respectively, revealing significant regional disparities. Functional demand analysis indicated consensus on the importance of basic HGMS functions, including real-time hypo- or hyperglycemia alerts (68.42%), Computerized Physician Order Entry–Electronic Medical Record interoperability (68.02%), and automatic glucose data transmission (67.51%). Among advanced functions, standardized education and training (64.47%), clinical decision support (61.54%), and tele-endocrinology consultation (61.13%) were also valued. Satisfaction was highest for automatic glucose data transmission (81.63%), while clinical decision support received the lowest satisfaction (45.51%). The Friedman test indicated significant differences in perceived importance and satisfaction across the 7 functions (χ²₆=89.895, 286.680, respectively; P<.001). Post hoc analyses with Bonferroni adjustment revealed that participants ranked the relative importance and satisfaction of functions in distinct orders, with automated glucose data transmission consistently rated highest for both. Qualitative analysis revealed persistent barriers in nonspecialist wards, including insufficient information integration and collaboration, absence of risk stratification and misaligned management, and insufficient knowledge and education need. Conclusions: HGMS implementation in the Chinese mainland is generally transitioning from digitized glucose recording toward intelligent glycemic management across hospitals. Future efforts should focus on establishing virtual wards for integrated care, developing dynamic glucose monitoring-guided management pathways, creating patient-centered collaborative platforms, and incorporating AI-enhanced clinical decision support to overcome existing barriers. Trial Registration:

Feasibility and Methodological Reflections From the Co-Creation of a Randomized Controlled Trial (The Kid’s Trial): Decentralized, Child-Led Citizen-Science Study

2026-08-28T16:00:05-04:00

Background: Limited public understanding of randomized controlled trials (RCTs) hinders recruitment, retention, and confidence in research. Early exposure to trial concepts may strengthen health literacy and research engagement. The Kid’s Trial was a global, decentralized, child-led study that cocreated and conducted an RCT to help children understand trials and their importance and to improve critical thinking. Objective: This paper evaluated the feasibility and methodological implications of engaging children in the cocreation and conduct of a fully online RCT. Methods: The Kid’s Trial ed a dedicated website guiding children through each step of designing and conducting an RCT. Materials were codeveloped with 2 patient and public involvement groups of children and parents. Any child aged 7-12 years could take part in as many steps as desired. Recruitment combined online and offline strategies, and engagement and self-reported learning were descriptively analyzed. The cocreated Randomized Evaluation of Sleeping With a Toy or Comfort Item (REST) trial was a 2-arm, pragmatic RCT comparing one week of sleeping with a comfort item versus without a comfort item. The primary outcome was sleep-related impairment, and the secondary outcome was overall sleep quality. Analyses followed an intention-to-treat (ITT) approach using mixed effects models adjusted for baseline measures. Results: Overall, 224 children participated in at least one step of The Kid’s Trial. Participation varied: 37% (n=82) completed one step, and 21% (n=48) completed 6 surveys. The REST trial randomized 139 children, with 73% (n=101) completing outcome surveys. Adjusted mean differences (intervention – control) were −0.53 (95% CI −3.40 to 2.34) for sleep-related impairment (=.71) and 0.28 (95% CI 0.01-0.55) for sleep quality (=.04). The difference was small and was not supported by sensitivity analyses. Poststudy responses (n=20) suggested improved self-reported trial understanding among respondents but were limited by low response rate and potential selection bias. Conclusions: The Kid’s Trial demonstrates the feasibility of a decentralized, child-led RCT cocreated through participatory citizen-science methods. Children can meaningfully contribute to trial design and conduct, and experiential participation may support engagement with trial concepts. Future studies should enhance engagement through community partnerships, shorter intervals between steps, and embedded learning assessments to improve inclusivity and retention. Trial Registration: ISRCTN Registry ISRCTN13756306; https://www.isrctn.com/ISRCTN13756306

Comparative Effectiveness of Noninvasive Brain-Computer Interface–Based Interventions for Upper Limb Rehabilitation in Poststroke Hemiplegia: Systematic Review and Network Meta-Analysis of Randomized Controlled Trials

2026-08-28T14:30:13-04:00

Background: Noninvasive brain-computer interface (BCI)–based interventions show promise for poststroke motor recovery. However, the intrinsic complexity of BCI-based interventions limits the determination of their comparative efficacy. Objective: Guided by the International Classification of Functioning, Disability and Health framework, this review evaluated the effectiveness of BCI-based interventions in poststroke upper limb rehabilitation and identify the optimal intervention. Methods: We searched PubMed, Cochrane Library, EBSCOhost, Web of Science, Embase, Wiley Online Library,CNKI, Wanfang, VIP, and SinoMed through July 2026. Randomized controlled trials (RCTs) assessing BCI-based interventions for poststroke upper limb rehabilitation were included. Outcomes were body functions and structures (Fugl-Meyer Assessment of Upper Extremity [FMA-UE]) and activities and participation (Action Research Arm Test [ARAT], Wolf Motor Function Test [WMFT], and Modified Barthel Index [MBI]). Risk of bias was assessed using Cochrane RoB 2, and evidence quality was graded using the Grading of Recommendations, Assessment, Development, and Evaluation framework. We used pairwise meta-analyses to evaluate the overall effectiveness of BCI-based interventions vs controls and network meta-analysis to compare the interventions. Results: Seventy-two RCTs involving 2906 patients with stroke were included, evaluating 12 BCI-based interventions. Pairwise meta-analyses demonstrated that, compared with control groups, BCI-based interventions improved FMA-UE (mean difference [MD] 5.33, 95% CI 4.28 to 6.38; 95% prediction interval [PI] −1.76 to 12.43), ARAT (MD 5.26, 95% CI 3.90 to 6.62; 95% PI 0.41 to 10.11), WMFT (MD 7.25, 95% CI 5.06 to 9.44; 95% PI 0.71 to 13.79), and MBI (MD 8.18, 95% CI 6.04 to 10.32; 95% PI −1.87 to 18.23). Network meta-analysis revealed that BCI-motor imagery-transcutaneous electrical acupoint stimulation (BCI-MI-TEAS) achieved the highest surface under the cumulative ranking curve (SUCRA; 95.5%) in improving FMA-UE. For ARAT, BCI-MI–end-effector robots and transcranial direct current stimulation (tDCS; 86.3%) alongside BCI-MI-TEAS (86.3%) yielded the highest SUCRA. BCI-MI–exoskeleton robot showed the highest SUCRA for WMFT (92.7%), whereas BCI-MI-TEAS (85.3%) and BCI-MI–exoskeleton robot (81.7%) ranked highest for MBI. The evidence quality ranged from very low to high across these interventions. Conclusions: This study represents the first network meta-analysis comparing the efficacy of different BCI-based interventions. Unlike previous reviews, interventions were categorized by experimental paradigms, external feedback devices, and adjunctive noninvasive brain stimulation, to enable clinically meaningful comparisons. Overall, BCI-based interventions significantly improved poststroke upper limb rehabilitation. Among evaluated interventions, BCI-MI-TEAS demonstrated the most performance across body functions, structures, and activities and participation, whereas BCI-MI–end-effector robot + tDCS showed advantages for fine motor dexterity and BCI-MI–exoskeleton robot improved activities of daily living.Given low to moderate evidence certainty and substantial heterogeneity, these findings remain exploratory. High-quality trials are needed to establish the clinical utility of these interventions. Trial Registration: PROSPERO CRD420251155441; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251155441

Digital Phenotyping in Health Research: Scoping Review of Methods, Gaps, and Opportunities

2026-08-27T17:00:37-04:00

Background: Wearable and connected digital devices continuously generate large volumes of real-world behavioral, physiological, and environmental data, offering new opportunities for health monitoring and personalized care. Digital phenotyping has emerged as a promising paradigm; yet, there is limited consensus regarding how such data should be analyzed. Inconsistent analytical practices and insufficient methodological reporting may compromise reproducibility, comparability, and the validity of findings. Objective: This scoping review examined current analytical practices in digital phenotyping in health research, identified methodological gaps, and highlighted the need for transparency and standardized approaches. Methods: Literature searches were conducted in PubMed, Scopus, Embase, and Web of Science, with Google Scholar used as a complementary source. Eligible studies were published up to December 31, 2024, involved human populations, and used wearable digital devices in longitudinal health research. Studies were screened independently by multiple reviewers using predefined eligibility criteria. Data extraction focused on 6 methodological domains: sample size determination, variable selection and definition, data cleaning and preprocessing, digital phenotyping methods, predictive modeling, and statistical significance handling in large-scale data contexts. Results: A total of 162 studies were included. Most studies were published from 2018 onward (n=144, 89%) and were conducted primarily in North America (n=92, 57%). Activity trackers (n=81, 50%), smartphones (n=35, 22%), accelerometers (n=26, 16%), and smartwatches (n=24, 15%) were the most frequently used devices. The most common outcomes were activity level (n=67, 41%), sleep (n=65, 40%), step counts (n=63, 39%), and heart rate (n=46, 28%). Heterogeneity and limited reporting were observed across all methodological domains. Preprocessing was the most frequently reported component (n=88, 54%), although specific aspects such as missing-data handling remained inconsistently described (n=35, 22%). Sample size determination methods were reported in only 30% (n=48) of studies, and variable selection methods were described in 27% (n=43). Digital phenotyping approaches were identified in 30% (n=48) of studies and predominantly used regression-based models. Predictive modeling approaches were reported in 17% (n=28) of studies, with substantial diversity in algorithms and limited reporting of validation procedures. Only 4.3% (n=7) of studies explicitly discussed statistical challenges related to large or high-dimensional datasets. Practices varied across health fields, with no domain consistently demonstrating comprehensive reporting across all methodological components. Conclusions: Digital phenotyping research is expanding rapidly, but methodological practices remain heterogeneous and insufficiently standardized. By providing a cross-domain overview of how wearable-derived longitudinal data are currently processed and analyzed in health research, this review highlights recurring gaps in the reporting and justification of analytical choices, particularly regarding sample size determination, preprocessing, variable definition, predictive modeling, and statistical inference. These findings emphasize the need for clearer analytical frameworks and more consistent reporting practices to improve transparency, reproducibility, and methodological rigor in wearable-based digital health research. Trial Registration: PROSPERO CRD420251234000; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251234000

Stigmatizing Language in Gender-Expansive Patient Records: Corpus Development, Disparity Analysis, and Natural Language Processing–Based Detection Study

2026-08-27T17:00:20-04:00

Background: Stigmatizing language (SL) in electronic health records (EHRs) can influence clinical decision-making, propagate bias across care encounters, and undermine patient trust. Gender-expansive patients (GEPs) may be particularly vulnerable to documentation-based stigma; however, large-scale quantitative evidence and fairness-aware evaluation of automated SL detection methods remain limited. Objective: This study aims to construct a gender-expansive-inclusive EHR corpus, quantify demographic disparities in SL using parallel outcome definitions with and without misgendering, and evaluate fairness-aware natural language processing (NLP) methods for automated detection of SL. Methods: We developed an annotated corpus of 754 clinical notes from the MIMIC-IV (Medical Information Mart for Intensive Care IV) database, including 366 GEP notes and 388 matched nongender-expansive patient (NGEP) notes, labeled for SL and its subtypes. Parallel outcome definitions were constructed with and without misgendering. Multivariable logistic regression was used to assess associations between gender-expansive status and SL while adjusting for race, age, and primary language. Multiple NLP models were evaluated for SL detection. Fairness-aware post hoc threshold optimization based on equalized odds principles was applied using training-set predictions to reduce subgroup error disparities. Results: SL was identified in 62.3% (228/366) of GEP notes compared with 25.5% (99/388) of NGEP notes. When misgendering was excluded, the prevalence remained higher among gender-expansive notes at 41.8% (153/366). In multivariable models, gender-expansive status was strongly associated with stigmatizing documentation when misgendering was included (adjusted odds ratio [OR] 4.87, 95% CI 3.54-6.70) and remained significant when misgendering was excluded (adjusted OR 2.12, 95% CI 1.54-2.91). Post hoc equalized odds threshold optimization for a state-of-the-art transformer-based detector reduced the difference in false-positive rate (ΔFPR) from 15.76 to 6.65 percentage points (pp) and the difference in true-positive rate (ΔTPR) from 7.22 pp to substantially lower levels while maintaining similar accuracy (82.78%-83.44%). When misgendering was excluded, fairness optimization reduced ΔFPR to 2.98 pp and ΔTPR to 0.65 pp, with an overall accuracy of 88.08%. Conclusions: SL is common in EHR documentation and disproportionately affects GEPs, and automated detection models show persistent subgroup performance gaps. Disparities remained significant even when misgendering was excluded, indicating that bias extends beyond identity-specific errors to broader evaluative language. This study introduces the first annotated corpus focused on SL in GEP documentation, quantifies demographic disparities, and demonstrates practical fairness-aware NLP strategies that can reduce error-rate inequities while preserving accuracy. These findings support equity-focused interventions to address SL through fairness-aware models as assistive auditing tools with human oversight.

Addendum: Exploring the Importance of Race and Gender Concordance Between Patients and Physical Therapists in Digital Rehabilitation for Musculoskeletal Conditions: Observational, Longitudinal Study

2026-08-27T16:30:31-04:00

Addendum: Impacts of Digital Care Programs for Musculoskeletal Conditions on Depression and Work Productivity: Longitudinal Cohort Study

2026-08-27T16:00:02-04:00

How AI Is Speeding Up the Diagnostic Odyssey for Rare Diseases

2026-08-27T10:00:19-04:00