Category: Statistics

Brief Bioinform. 2026 May 4;27(3):bbag275. doi: 10.1093/bib/bbag275.

ABSTRACT

Nonlinear genetic architectures, including epistasis and threshold effects, are increasingly recognized as contributors to complex disease risk, yet most existing SNP-set association tests rely on linear modeling assumptions, resulting in reduced power and unstable inference when genetic effects are nonlinear or heterogeneously distributed across variants. To address this limitation, we propose a nonlinear high-dimensional inference framework for set-based genetic association analysis that integrates scalable kernel representations with valid statistical inference. The framework combines distance correlation-based sure independence screening to reduce ultra-high dimensional predictors, kernel principal component analysis with Nyström approximation for nonlinear feature extraction, and de-sparsified LASSO to enable asymptotically valid hypothesis testing in high dimensions, together with a two-stage omnibus testing strategy that adaptively aggregates evidence across complementary signal models. Extensive simulation studies demonstrate that the proposed method maintains well-calibrated Type I error and consistently achieves higher power than established set-based approaches, including Sequence Kernel Association Test and adaptive Sum of Powered Score test, particularly under nonlinear and heterogeneous genetic effect scenarios, while remaining competitive in linear settings. Application to Alzheimer’s Disease Neuroimaging Initiative data identifies gene-level associations with brain regional volumes that converge on neuronal excitability, calcium signaling, and cytoskeletal regulation, biological processes centrally implicated in neurodegeneration. Together, this work provides a robust and scalable framework for nonlinear set-based inference in genome-wide studies, expanding the analytical toolbox for dissecting complex genetic contributions to disease.

PMID:42202283 | DOI:10.1093/bib/bbag275

Brief Bioinform. 2026 May 4;27(3):bbag258. doi: 10.1093/bib/bbag258.

ABSTRACT

Expression quantitative trait locus (eQTL) mapping is used to identify the functional link between a genomic variant and a gene’s expression. A significant eQTL association does not mean a causal relationship or mechanism, and further investigation is needed to understand how a single-nucleotide polymorphism (SNP) impacts gene expression. One of the most plausible explanations for eQTL is that a genomic variant affects transcription factor (TF) binding and thus impacts its regulation on target genes (TGs). However, the current eQTL model does not provide information on the TF and how its regulation is mediated by the SNP’s genotypes. Here, we propose a new method called differential co-expression QTL (ceQTL) among different alleles using Chow statistics to specifically detect eQTLs that are bound by a particular TF. We start with building a trio of TF, its TG, and related SNP, and then test the significant coefficient difference among different genotypes of the SNP. We applied this ceQTL model to simulated data and the lung tissue datasets from the genotype-tissue expression project. The simulated data results showed that the model was robust to detect true ceQTLs at variable sample sizes and different minor allele frequencies as measured by area under the curve. Our tool also performed a TF binding affinity analysis to add another layer of evidence for functional interpretation. In summary, ceQTL analysis provides a more interpretable and biological insight into the mechanism of eQTL and transcriptomic regulation, which would help us better understand how genomic variants affect phenotypes and diseases.

PMID:42202282 | DOI:10.1093/bib/bbag258

Brief Bioinform. 2026 May 4;27(3):bbag267. doi: 10.1093/bib/bbag267.

ABSTRACT

Transcriptome assembly and quantification are crucial steps in the differential expression analysis of RNA-seq data. As transcriptome assembly precedes quantification, its results inevitably influence the outcomes of quantification. This study investigates the impact of transcriptome assembly algorithms on quantification outcomes in next-generation RNA-seq data analysis. From the perspective of quantification results, we evaluate the performance of transcriptome assembly algorithms. We assess the assembly quality and stability of three commonly used transcriptome assemblers-StringTie2, Scallop, and Cufflinks-on both simulated and real datasets. Our evaluation provides references for downstream analyses and identifies the most effective and stable pipeline, which is specifically the pipeline combining HISAT2 (for transcriptome alignment) and StringTie2 (for assembly). Furthermore, we compare simulated data generated by RNA-seq data simulation tools with real RNA-seq data and reveal that simulated data fails to fully capture the complexity of real data. Through this analysis, we identify transcript features associated with poor assembly and quantification performance, specifically highlighting two extreme cases: long, low-expression transcripts that are often overlooked and short transcripts that are prone to quantification errors. These findings offer valuable insights into future software development directions.

PMID:42202281 | DOI:10.1093/bib/bbag267

J Med Internet Res. 2026 May 27;28:e87049. doi: 10.2196/87049.

ABSTRACT

BACKGROUND: Health care workers (HCWs) face sustained psychological demands that place them at heightened risk for burnout and posttraumatic stress disorder (PTSD). However, assessing psychological distress in this population remains challenging because of stigma, underreporting, and the limitations of self-report tools. Although nonverbal behaviors such as facial expressions and gaze hold diagnostic promise, most approaches overlook the fine-grained, temporal fluctuations in these signals. In this study, we focused on microbehavior intervals-brief, involuntary changes in multimodal nonverbal signals-that emerge during emotion-eliciting interviews.

OBJECTIVE: This study aimed to determine whether microbehavior intervals improve the discrimination of psychological distress profiles among HCWs with symptoms of burnout and PTSD.

METHODS: HCWs participated in a semistructured interview that included 5 work-related, emotionally charged questions and that was recorded via Webex (online video platform). Participants also completed validated questionnaires for burnout (Maslach Burnout Inventory General Survey 9-item) and PTSD (PTSD checklist for Diagnostic and Statistical Manual, 5th edition). Recordings were analyzed with computer vision models to generate time-series data of facial expressions, head movement, gaze, body posture, and hand gestures. An unsupervised anomaly detection model (MOMENT [a Family of Open Time-Series Foundation Models]) isolated microbehavior intervals without requiring manual labels. Features derived from these intervals were used to train a deep learning classifier that predicted 4 symptom classes of psychological distress: “moderate-severe burnout,” “subthreshold-provisional PTSD,” “burnout+PTSD,” and “resilient.” We conducted an ablation study by systematically removing one behavioral data stream at a time. Finally, we conducted an explainability analysis to characterize the features driving model predictions.

RESULTS: We analyzed 258 interview recordings from 151 HCWs. Per interview, an average of 19.65 (SD 6.01) microbehavior intervals were detected, each lasting an average of 1.31 (SD 1.10) seconds. The classifier demonstrated robust performance across classes, achieving a macro- F₁-score of 0.75 and a macro area under the receiver operating characteristic curve of 0.80 on held-out data. Ablation analysis showed that excluding gaze or arousal-valence signals caused the largest performance declines, particularly in recall and F₁-score. The explainability analysis revealed distinct temporal patterns across symptom classes, with irregularity and variability in microbehaviors emerging as key predictors.

CONCLUSIONS: Focusing on microbehavior intervals yields a scalable, interpretable, and annotation-free framework for detecting psychological distress from nonverbal signals. By moving from whole-video features to fine-grained multimodal temporal modeling, we successfully captured subtle, involuntary fluctuations in nonverbal responses to emotion-eliciting questions. This multimodal approach enables an objective, robust, and explainable assessment of psychological distress and offers a promising complement to conventional psychometric assessments.

PMID:42202278 | DOI:10.2196/87049

Eur Stroke J. 2026 May 6;11(5):aakag050. doi: 10.1093/esj/aakag050.

ABSTRACT

INTRODUCTION: Although epidemiological studies often group the Baltic states together, they differ significantly in national stroke care legislation and infrastructure. Our study aimed to explore and compare the current state of stroke care in Lithuania, Latvia and Estonia.

PATIENTS AND METHODS: We analysed the Stroke Action Plan for Europe (SAP-E) Stroke Service Tracker data from 2022, including data from the respective National Health Insurance Funds and direct centre-level queries. Geographic Information System-based modelling assessed population access to stroke-ready hospitals within 1 h. Key metrics, including hospitalised stroke incidence, stroke unit admission, recanalisation therapy and in-hospital as well as 30-day mortality, were compared using Z-tests for proportions.

RESULTS: The hospitalised stroke incidence per 100,000 inhabitants was similar in Lithuania (353) and Latvia (354), but lower in Estonia (246), despite similar population structures. Lithuania had the highest proportion of its population (94.0%) with access to a stroke-ready hospital within 1 h, followed by Latvia (87.1%) and Estonia (84.7%, P < .001). Estonia had the highest proportion of stroke unit admission rates and the lowest mortality rates-9.6% (in-hospital) and 15.0% (30-day) for ischaemic stroke. Endovascular treatment was most frequent in Lithuania (8.6% of all strokes, P < .001), while Estonia had the highest rate of intravenous thrombolysis (29.0%, P < .001).

CONCLUSIONS: Despite broadly comparable populations and formal SAP-E alignment, the Baltic states exhibit marked differences in stroke access, treatment and outcomes. High stroke unit admissions and high recanalisation rates in Estonia may be associated with lower ischaemic stroke mortality, underscoring the importance of system design beyond geographic coverage alone.

PMID:42202277 | DOI:10.1093/esj/aakag050

JMIR Med Educ. 2026 May 27;12:e87483. doi: 10.2196/87483.

ABSTRACT

BACKGROUND: The Objective Structured Clinical Examination (OSCE) is a prevalent method for evaluating clinical competence in medical education. As OSCEs become increasingly standardized and resource intensive, alternative evaluation methods are being explored, particularly because of the limited availability of certified examiners. However, few studies have investigated whether wearable technologies can support OSCE assessment. Wearable devices may provide a means of recording clinical skills from the examiner’s perspective.

OBJECTIVE: This pilot study, conducted in 2024, aimed to investigate the feasibility of using an examiner-worn neck-mounted camera for recording OSCE scenarios and to evaluate the evaluability of clinical performance from the recorded footage.

METHODS: In total, 9 experienced medical educators participated in a simulated OSCE scenario involving electrocardiogram lead placement. All participants completed the initial live assessment and the postuse questionnaire, while 8 of 9 (89%) participants completed the subsequent video-based reassessment. Video recordings from both a fixed camera and a neck-mounted camera (THINKLET) were used to assess the evaluability of each OSCE item. Following a washout period, evaluators reassessed the neck-mounted camera recordings by using the original checklist, while fixed-camera recordings were used to judge the evaluability of each item. Agreement between live and video-based scoring was analyzed using percent agreement and the Cohen κ coefficient. A postevaluation questionnaire captured evaluators’ experiences with the wearable device.

RESULTS: Cohen κ ranged from 0.258 to 0.913 (mean 0.67, SD 0.20). Across checklist observations, more items were judged to be evaluable in the neck-mounted camera recordings than in the fixed-camera recordings, particularly for tasks requiring observation of fine motor skills. Evaluators reported generally positive experiences with the device, although some noted issues related to audio quality, comfort, posture restriction, and limited visibility at low angles.

CONCLUSIONS: Although further investigation is needed, this pilot study suggests that an examiner-worn neck-mounted camera may be a valuable supplementary assessment tool for selected OSCE tasks. Further work is needed to refine the device, standardize recording protocols, and clarify how it can best support review and verification alongside live evaluation.

PMID:42202275 | DOI:10.2196/87483

JMIR Res Protoc. 2026 May 27;15:e92264. doi: 10.2196/92264.

ABSTRACT

BACKGROUND: Mis-triage represents a global concern, with reported rates ranging from 15% to 33%. Understanding its causes and contributing factors is essential for ensuring patient safety. Currently, available studies have mainly focused on evaluating triage systems rather than investigating the human factors affecting triage performance. A major limitation in triage evaluation studies is the lack of standardized criteria to assess patient acuity and the absence of a clear consensus on how to measure triage accuracy. Most studies rely on retrospective data, which often fail to capture real-life clinical complexity. Therefore, the underlying causes and consequences of mis-triage remain partially understood.

OBJECTIVE: This study aims to improve triage by defining the optimal triage evaluation process and identifying clinician-, patient-, and system-level factors that compromise its accuracy and safety.

METHODS: Reducing Mis-Triage in Emergency Departments (RemEDy) will be a 4-phase, mixed methods project conducted across 7 Swiss emergency departments. The first phase will focus on developing a standardized triage evaluation instrument, combining evidence from a scoping review of triage evaluation processes, workshops with triage clinicians using design thinking methodology, and a modified Research and Development-University of California Delphi involving international experts and patient representatives. The second phase will prospectively implement this instrument in real time within a multicenter observational cohort study to evaluate triage performance; quantify mis-triage; and identify predictors at the patient level (eg, demographics), clinician level (eg, training), and system level (eg, crowding and length of stay). The third phase will focus on designing and validating an artificial intelligence-based decision support tool, applying multimodal models that integrate real-time triage data to enhance acuity prediction and minimize human error. The fourth phase will develop and evaluate a targeted training program, guided by the Capability, Opportunity, Motivation, and Behavior model, to strengthen triage accuracy and mitigate cognitive biases.

RESULTS: The project was funded by the Swiss National Science Foundation in March 2025 (grant 10004535). At submission, the scoping review is ongoing and expected to be completed in early 2026. Development and piloting of the triage evaluation instrument will take place in 2026. A multicenter cohort study is planned between October 2026 and June 2027. The intervention study is scheduled between October 2027 and December 2028. Final results are expected in 2029.

CONCLUSIONS: The RemEDy project addresses key limitations of current triage research, including the lack of standardized evaluation methods. By combining expert and clinician consensus; real-time assessment; and multilevel analysis of patient-, clinician-, and emergency department-level factors, RemEDy is expected to provide a more comprehensive understanding of mis-triage and its causes. RemEDy will establish a novel framework for real-time triage evaluation and inform the development of targeted training programs with the potential to improve triage accuracy, safety, and equity.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/92264.

PMID:42202274 | DOI:10.2196/92264

JMIR Perioper Med. 2026 May 27;9:e89277. doi: 10.2196/89277.

ABSTRACT

BACKGROUND: Effective postoperative communication is vital for patient recovery, yet traditional text-based discharge instructions often lead to poor comprehension and adherence, particularly among patients with limited health literacy. Although educational videos improve understanding and retention, their widespread use has been hampered by high production costs. Generative artificial intelligence (AI) offers a scalable solution for creating engaging video content.

OBJECTIVE: The primary objective of this pilot study was to assess the feasibility of creating and deploying AI-generated, avatar-led videos for postoperative instruction delivery. Secondary objectives included comparing knowledge retention, engagement, perceived clarity, and user experience between AI-generated video and traditional text-based handout formats among health care workers.

METHODS: In this randomized pilot study, 38 health care worker volunteers were recruited as a convenience sample to pilot-test the intervention before patient implementation. Participants were assigned to either a text handout group (n=19, 50%) or an AI-generated video group (n=19, 50%). Both groups received information on 10 common postoperative topics. The primary outcome was objective knowledge, assessed via a 10-item quiz. Secondary outcomes, measured through surveys with 5-point Likert scales, included engagement time, subjective engagement, perceived clarity, usefulness, confidence in understanding, and information retention. Qualitative feedback was also collected.

RESULTS: Objective knowledge quiz scores did not differ significantly between groups (mean 8.89, SD 1.20 for the AI-generated video group vs mean 8.21, SD 1.78 for the text handout group; P=.17; Cohen d=0.45). Participants in the AI-generated video group demonstrated significantly higher engagement time (mean 15.11, SD 7.78 minutes vs mean 8.84, SD 4.03 minutes; P=.004; Cohen d=1.04). They also rated instructions as significantly clearer (mean 4.63, SD 0.50 vs mean 4.00, SD 0.82; P=.007; Cohen d=0.93), more engaging (mean 4.05, SD 0.78 vs mean 3.32, SD 1.00; P=.02; Cohen d=0.81), and more effective for retention (mean 4.42, SD 0.84 vs mean 3.37, SD 0.68; P<.001; Cohen d=1.38). Qualitative feedback highlighted the engaging nature of AI-generated videos but noted areas for avatar refinement.

CONCLUSIONS: In this pilot study with health care workers, AI-generated avatar videos did not improve objective knowledge scores but significantly enhanced engagement, perceived retention and perceived clarity (Cohen d=0.81-1.38). Future studies in actual patient populations with diverse health literacy levels are needed to determine whether these engagement advantages translate into improved knowledge outcomes.

PMID:42202261 | DOI:10.2196/89277

J Clin Oncol. 2026 May 27:JCO2501441. doi: 10.1200/JCO-25-01441. Online ahead of print.

ABSTRACT

NRG Oncology RTOG 0539 was a prospective phase II trial of risk-adapted radiotherapy for patients with WHO grade 1-3 meningioma. Low-risk (group 1, n = 60) was defined as a grade 1 tumor after gross total resection or subtotal resection (GTR/STR) and prospectively monitored. Intermediate-risk (group 2, n = 52) was defined as recurrent grade 1 or newly diagnosed grade 2 tumor after GTR and treated with radiotherapy (54 Gy). High-risk (group 3, n = 53) included a newly diagnosed grade 2 tumor after STR, newly diagnosed grade 3 tumor, or recurrent grade 2 or 3 tumor and treated with radiotherapy (60 Gy). Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan-Meier method. The median follow-up times for the low-, intermediate-, and high-risk cohorts were 12.1, 12.0, and 11.1 years, respectively. The 10-year PFS and OS rates for the low-, intermediate-, and high-risk cohorts were 85.2% and 94.1%, 72.2% and 84.7%, and 42.5% and 51.1%, respectively. Five patients (9.6%) and eight patients (15.1%) had a grade 3+ toxicity attributed to radiotherapy in the intermediate- and high-risk cohorts, respectively. The long-term outcomes using this risk-adapted approach support observation for low-risk patients, inform radiotherapy patient selection and practice standards for intermediate- and high-risk patients, and provide comparative benchmarks for future trials.

PMID:42202246 | DOI:10.1200/JCO-25-01441

IEEE J Biomed Health Inform. 2026 May 27;PP. doi: 10.1109/JBHI.2026.3694093. Online ahead of print.

ABSTRACT

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition with increasing global prevalence and no standardized biological test for early detection. Current diagnosis methods rely heavily on behavioral assessments, which are subjective, time-consuming, and prone to variability. This study proposes a hybrid feature fusion framework for non-invasive ASD diagnosis using electroencephalogram (EEG) signals, specifically event-related potentials (ERPs) such as P300 components obtained from the BCIAUT-P300 dataset. EEG recordings were captured using a g.Nautilus wireless system with eight scalp electrodes, and preprocessed using 0.5-30 Hz bandpass filtering and baseline subtraction to enhance signal quality. Twenty-two EEG features were extracted across time, frequency, and time-frequency domains using methods such as Wavelet Transform, power spectral density, higher-order statistics, and principal component analysis. Five optimal methods, PCA, HOS, PSD, FDA, and CWT, were selected based on their classification potential and fused using both feature-level and decision-level strategies. Ensemble classifiers including SVM, XGBoost, LDA, and Random Forest were trained and evaluated on the fused feature set. The proposed hybrid fusion framework achieved a classification accuracy of 97.7%, sensitivity of 96.8%, and specificity of 98.5%, outperforming traditional single feature or single classifier approaches. The integration of multi-domain feature descriptors with ensemble learning contributes to increased robustness, generalizability, and diagnostic precision. Our work demonstrates the feasibility of combining EEG-based biomarkers with machine learning to support early ASD diagnosis. The framework offers a scalable approach that is aligned with biomedical informatics objectives, with potential for clinical deployment and integration into portable EEG-based screening systems.

PMID:42202207 | DOI:10.1109/JBHI.2026.3694093