medRxiv : the preprint server for health sciences最新文献

{"title":"Generalizability of proteomic risk prediction across biobanks reveals dependence on phenotype definitions.","authors":"Sophie J Parsa, Yon Ho Jee, M Austin Argentieri, Jeremy Guez, Wenhan Lu, Evin M Padhi, Elizabeth Kiernan, Reza Jabal, Mitja I Kurki, Benjamin M Neale, Lee Lichtenstein, Michael E Talkowski, Stephen B Montgomery, Niall J Lennon, Alicia R Martin, Mark J Daly, Konrad J Karczewski","doi":"10.64898/2026.10.01.26364038","DOIUrl":"https://doi.org/10.64898/2026.10.01.26364038","url":null,"abstract":"<p><p>Advances in high-throughput proteomics technologies have enabled the assessment of dynamic health states across biobank-scale cohorts. Disease prediction models built on these data have higher accuracy than baseline clinical models for a broad range of diseases, and provide avenues to understand disease pathogenesis. However, the generalizability of these prediction models across cohorts has yet to be established at scale. Here, we train models for 15 diseases in the UK Biobank (UKB; n=53,026) based on Olink proteomics data, and find high accuracy for disease prediction (mean AUC=0.74; range 0.56-0.89). These models have improved accuracy over models using only clinical factors (mean ΔAUC = 0.03), and do not depend on model architecture, with simpler models (e.g. L2) performing as well as complex models (e.g. transformer). We assess generalizability of the UKB-trained models in two external cohorts: FinnGen (FG; n=5,865) and <i>All of Us</i> (AoU; n=7,405), spanning two Olink platforms. We find that proteomics-based prevalent (classification) and incident (prediction over next 5 years) disease models largely generalize across cohorts, but performance varies across diseases. Specifically, 10 of 14 prevalent and 13 of 15 incident disease models show no significant decrease in performance across any biobank. Adjusting for demographic, ancestry, and technical covariates, we demonstrate that differences in phenotyping quality are likely the major drivers of variability across cohorts. These results indicate that proteomic risk models can be powerful and generalizable predictors of disease across multiple cohorts.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648823/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149167440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interaction Between Exercise Intensity and Sedentary Behavior on VO ₂ max Change in Patients with De Novo Parkinson Disease: The Study in Parkinson Disease of Exercise (SPARX) Phase 2 Randomized Clinical Trial.","authors":"Byung Joon Pae, Tom S Novak, Garett J Griffith, Charity G Patterson, Madeleine E Hackney, Cory L Christiansen, Daniel M Corcos, Joe R Nocera","doi":"10.64898/2026.09.30.26364437","DOIUrl":"https://doi.org/10.64898/2026.09.30.26364437","url":null,"abstract":"<p><strong>Background: </strong>Parkinson disease (PD) is associated with increased sedentary time and reduced cardiorespiratory fitness, making exercise an important component of disease management. Although exercise improves cardiorespiratory fitness, individuals with PD vary substantially in their response to exercise. Sedentary time may contribute to this variability, yet whether sedentary time modifies the response to intervention and varying exercise intensities within an intervention is unknown. We examined whether sedentary time interacted with 6 months of moderate-intensity or high- intensity aerobic exercise on maximal oxygen consumption (VO₂max) in individuals with PD.</p><p><strong>Methods: </strong>This secondary analysis used data from the SPARX phase 2 randomized clinical trial including 128 de novo PD patients assigned to high- or moderate-intensity treadmill exercise, or usual care for 6 months. Interactions between randomized group and post-randomization 6-months average sedentary time were assessed for associations with 6-month change in VO ₂ max by using linear regression models corresponding to Analysis of Covariance (ANCOVA) models with interaction.</p><p><strong>Results: </strong>The interaction between exercise intensity and sedentary behavior on 6-month change in VO ₂ max in the moderate-intensity group compared with usual care was not significant (β: -1.78, 95% CI: -3.76 to 0.20, p-value: 0.08), although the sensitivity analysis restricted to complete cases showed a significant interaction where each 10% increase in 6-months average sedentary time was associated with a 2.51 mL/kg/min decrease in VO ₂ max change in the moderate-intensity group compared with usual care (β: -2.51, 95% CI: -4.53 to -0.48, p-value: 0.02). There was no significant interaction between exercise intensity and sedentary behavior on 6-month change in VO ₂ max in the high-intensity group compared with usual care (β: -1.46, 95% CI: -4.74 to 1.83, p-value: 0.38). The sensitivity analysis showed similar results.</p><p><strong>Conclusions: </strong>These hypothesis-generating, preliminary findings suggest that sedentary time may modify the response to exercise among individuals participating in moderate-intensity aerobic exercise. Future adequately powered randomized controlled trials are needed to confirm these findings and determine whether sedentary behavior should be considered when prescribing exercise for individuals with PD.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648803/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149167395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reliability and accuracy of medication-based performance measures for assessment of outpatient cardiology clinicians.","authors":"Christine Chow, Elise C Gunzburger, Chris Wilson, Taufiq S Salahuddin, Paul L Hess, Utibe R Essien, Fatima Rodriguez, Kai-Chun Yang, Boback Ziaeian, Jacob A Doll","doi":"10.64898/2026.09.29.26364366","DOIUrl":"10.64898/2026.09.29.26364366","url":null,"abstract":"<p><strong>Background: </strong>Multiple performance measures are used to assess the management of common cardiovascular conditions, such as coronary artery disease (CAD), heart failure with reduced ejection fraction (HFrEF), and atrial fibrillation (AF). These measures are sometimes used to assess individual clinicians, despite lack of evidence of reliability and accuracy. We used national Veteran Affairs (VA) data to test medication-based performance measures for assessment of clinician performance in outpatient cardiology care.</p><p><strong>Methods: </strong>This retrospective study examined VA cardiology clinician/Veteran dyads from 2017-2023 to evaluate 10 medication-based performance measures for CAD, HFrEF, or AF with elevated stroke risk. Achievement of each performance measure was defined as an active prescription of an indicated medication for each eligible patient encounter, summarized at the clinician level. Reliability was calculated using the beta-binomial method. A Bayesian causal model was used to assess the impact of case complexity on performance measure achievement. Finally, we performed peer review of a subset of cases (n = 952) to assess the accuracy of three representative performance measures.</p><p><strong>Results: </strong>Of 456,071 unique cardiology encounters included in this study, 312,331 included CAD, 224,599 included HFrEF, and 178,678 included AF. These patients were treated by 1,043 cardiology clinicians. Median performance measure achievement ranged from 0.23 (SGLT2i for HFrEF) to 0.99 (beta blocker for HFrEF). Measure reliability varied from 0.63 to 0.90. Case complexity did not significantly affect performance measure achievement. Using peer review as the gold standard, accuracy was poor to moderate (balanced accuracy 0.51-0.73), and measures underestimated performance relative to peer review.</p><p><strong>Conclusions: </strong>Ten medication-based performance measures for CAD, HFrEF, and AF can be efficiently measured using electronic health record data, and most have sufficient reliability to support clinician benchmarking and the identification of potential care gaps without need for risk adjustment. However, suboptimal accuracy compared with a peer review gold standard indicates that these measures may be insufficient as a standalone assessment of clinician quality.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648801/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149171936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-omics approach to map a clinically diverse large cohort with homozygous founder RAG1 p.C176F variant among Old Order Mennonites.","authors":"Evan Potts, Marta Toth, Karlla W Brigatti, Erik G Puffenberger, Paola R Suhet, Mustafa Talib, Alanna E Koehler, Zsuzsanna Gaál, Kellie A Larsen, Rahim Z Miller, Daryl J Gibson, Philip Mendez, Zixiao An, Jeffrey J Bednarski, Peter Blazso, Marita Bosticardo, Sharat Chandra, Ivan K Chinn, Joelle W Clark, Antonio Condino-Neto, Ottavia M Delmonte, Dimana Dimitrova, Emily S J Edwards, Ganesh V Halade, Jennifer R Heimall, Ayal Hendel, Sarah E Henrickson, Katherine E Herman, Avni Joshi, Heather Kenney, Maleewan Kitcharoensakkul, Gloria Magro, Eliza C Martin, Edgar B Oliveira-Junior, Mei-Sing Ong, Oscar E Ospina, Francesca Pala, Kady Palmer, Jason P Raasch, Mathilde Salamon, Christopher D Scharer, Christine M Seroogy, Svetlana O Sharapova, Raz Somech, Michael Teng, Timothy J Thauland, Boglarka Ujhazi, James W Verbsky, Brant R Ward, Shengfeng Xu, Melis Yilmaz, Yuhang Zhang, Wei Zhao, Krisztian Csomos, Yu Nee Lee, David G Schatz, Manish Butte, Nicholas L Rider, Laura E Poskitt, Luigi D Notarangelo, Kevin A Strauss, Jolan E Walter","doi":"10.64898/2026.09.29.26363255","DOIUrl":"https://doi.org/10.64898/2026.09.29.26363255","url":null,"abstract":"<p><p>Partially preserved function of recombination-activating genes (RAG) 1 and 2 can be linked to divergent and hard-to-predict clinical and immunological phenotypes, even among close relatives. It is unclear how intrinsic RAG hypomorphism interacts with extrinsic environmental pressures to shape clinical trajectory. We dissect this phenomenon in a unique cohort of 18 Old Order Mennonite individuals with a novel missense founder <i>RAG1</i> variant (p.C176F) with 25.8% recombinase activity of wild-type protein. After studying a myriad of markers, we demonstrate varying severity and progression of clinical disease (predominantly in females) that paralleled decline in CMV-specific immune competence, break in tolerance with hallmark presence of antibodies to IFNα, a TH1/IFNγ-driven inflammatory milieu with elevated serum markers (CXCL9/10), expansion of follicular helper T cells and clonal B cells enriched for autoreactive-prone VH4.34 B cell receptor. BCR and TCR repertoires carry the fingerprints of both extrinsic and RAG-driven intrinsic changes. Through this 'experiment of nature', our cohort demonstrated that partial RAG deficiency (pRD) is not a monolithic disease but the result of compounded RAG intrinsic and extrinsic events. Overall, this study uses multi-omics approaches to dissect the contribution of a hypomorphic RAG variant to variable severity of clinical disease and immunological dysregulation at cellular and molecular level. Emerging biomarkers of a multi-omic roadmap should allow monitoring of progression of clinical disease and guide who and when to treat with hematopoietic stem cell transplant in pRD.</p><p><strong>One sentence summary: </strong>A unique cohort of pediatric and adult US Mennonite patients with shared novel partial RAG defect and variable penetrance of disease allow correlation of clinical severity with RAG-intrinsic and extrinsic factors driving plasticity in T and B cell competence to eliminate infections and tolerance to self.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648806/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149171584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"NMR metabolomics for cardiovascular-kidney-metabolic risk stratification in an integrated healthcare system.","authors":"Jonathan Chacon-Barahona, Carlos Iribarren, Jamal Rana, Gina Rinetti Vargas, Sarah Rowell, Olivier Elemento, Karsten Suhre, Devin Absher, Jan Krumsiek","doi":"10.64898/2026.10.01.26364322","DOIUrl":"https://doi.org/10.64898/2026.10.01.26364322","url":null,"abstract":"<p><p>Metabolomic profiling predicts future disease in population cohorts, but its value in real-world healthcare remains uncertain. We profiled 250 nuclear magnetic resonance (NMR) biomarkers in 54,933 participants from a diverse integrated US healthcare system linked to longitudinal electronic health records, evaluating nine cardiovascular-kidney-metabolic disease endpoints. Adding NMR to standard-of-care predictors improved discrimination for nearly all endpoints, with the largest gains for type 2 diabetes, chronic kidney disease and metabolic dysfunction-associated steatotic liver disease. Beyond disease onset, NMR improved prediction of disease-specific complications and cross-organ outcomes. Gains persisted across demographic, anthropometric, and genetic-risk strata and were greatest where standard-of-care prediction was weakest. NMR-enhanced models generally preserved calibration, improved high-risk enrichment and increased decision-curve net benefit. Together, these findings support NMR metabolomics as a common approach to risk stratification across the cardiovascular-kidney-metabolic disease continuum and illustrate how NMR-enhanced risk estimates could be integrated with longitudinal clinical information.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648799/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149171741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging Shared Genetic Architecture Between Psychiatric and Other Medical Disorders Increases Power and Reveals Novel Psychiatric Risk Loci.","authors":"Qingyi Zhong, Apostolia Topaloudi, Peristera Paschou","doi":"10.64898/2026.10.01.26364453","DOIUrl":"10.64898/2026.10.01.26364453","url":null,"abstract":"<p><p>Psychiatric disorders frequently co-occur with a broad range of non-psychiatric conditions, suggesting shared biological mechanisms that transcend traditional diagnostic boundaries. Although recent studies have demonstrated substantial genetic overlap between psychiatric and other medical disorders, the genomic architecture underlying these relationships and the extent to which shared genetic liability can be leveraged for psychiatric gene discovery remain incompletely understood. Here, we systematically investigated the shared genetic architecture between eleven major psychiatric disorders and thirty-nine non-psychiatric disorders spanning cardiovascular, metabolic, autoimmune, neurological, gastrointestinal, respiratory, and other disease categories. To maximize statistical power, we meta-analyzed genome-wide association study (GWAS) summary statistics from large biobanks and international consortia. Genome-wide analyses identified 298 significant genetic correlations between psychiatric and non-psychiatric disorders, revealing pervasive shared liability across disease domains. Local genetic correlation analyses identified 180 genomic regions contributing to cross-disorder risk, while colocalization analyses revealed 1,470 candidate shared variants. Several highly pleiotropic loci, including regions containing FURIN and TCF4, contributed to risk across multiple psychiatric and other medical conditions. Genomic structural equation modeling identified recurrent dimensions of shared liability, including cardiovascular-kidney-metabolic and neuroimmune-related dimensions, suggesting that psychiatric-medical comorbidity is organized into biologically coherent systems. Leveraging these shared genetic architectures through cross-trait analyses identified 1,569 significant loci, including 382 psychiatric risk loci not reported in the individual psychiatric GWAS used in our analyses. Integrative positional and functional gene-mapping approaches prioritized genes involved in neuronal development, synaptic organization, neurotrophic signaling, and cellular homeostasis, with LRFN5, MED19, PITPNM2, CTNND1, TMEM106B, GLYCTK, and PCLO as the top implicated genes. Pathway and tissue-enrichment analyses further demonstrated convergence on neurodevelopmental and synaptic processes and significant enrichment across multiple brain regions. Our results provide a comprehensive characterization of the genetic architecture underlying comorbidity between psychiatric and other medical disorders and demonstrate that leveraging genetically correlated medical conditions across domains can substantially increase power for psychiatric risk-locus discovery.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648787/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149172306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncertainty-Aware Multiscale Treatment Response Forecasting in Advanced Non-Small Cell Lung Cancer Using Residual-Variance Conformal Prediction.","authors":"Faisal Yaseen, Daniel S Hippe, Sunan Cui, Jie Fu, Shouyi Wang, Lei Deng, Ting Ye, Paul E Kinahan, Jing Zeng, John H Gennari, Stephen R Bowen","doi":"10.64898/2026.09.30.26364431","DOIUrl":"10.64898/2026.09.30.26364431","url":null,"abstract":"<p><p>Voxel-level prediction of treatment response from longitudinal Fluorodexoyglucose (FDG) positron emission tomography / computed tomography (PET/CT) can enable spatially adaptive dose planning in advanced non-small cell lung cancer (NSCLC), but point predictions alone lack the uncertainty quantification needed for safe clinical decision-making. Standard conformal prediction (CP) provides distribution-free coverage but yields spatially uniform intervals and assumes exchangeability-an assumption violated by intra-tumoral spatial correlation. We propose a multiscale CP framework built on a re-optimized VoxelForecast generalized least squares model that explicitly captures voxel-level covariance. Variogram sensitivity analysis identified the Stable model as significantly superior to previouly used Matérn family. The framework combines hierarchical nested leave-one-patient-out calibration, which restores exchangeability for valid voxel-level coverage, with residual-variance CP (VarCP), which learns a heteroscedastic, location-dependent scale to produce adaptive intervals while preserving finite-sample marginal coverage guarantees. On two prospective NSCLC trial cohorts-FLARE-RT (N=25 locally advanced, chemoradio-therapy; 11,100 voxels) and BRIGHT (N=19 metastatic, chemoimmunotherapy; 26,980 voxels)-multiscale feature integration reduced voxel-level prediction error by 35% and 31%, respectively. VarCP maintained nominal coverage while producing 7-18% (FLARE-RT) and 4-8% (BRIGHT) narrower voxel-level intervals than a hierarchically calibrated standard CP baseline (paired Wilcoxon signed-rank, significant at most miscoverage levels), with comparable lesion-level gains (7-18% and 6-10%) via median-based aggregation of voxel residuals. VarCP delivers tighter yet mathematically guaranteed spatial uncertainty maps, supporting both voxel-level dose-adjustment and whole-lesion treatment planning in advanced NSCLC.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648795/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149171923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Differentiating nonfluent/agrammatic and logopenic primary progressive aphasia in Catalan-Spanish bilinguals by applying multilingual multimodal machine learning to connected speech.","authors":"Lokesha Srinivas Pugalenthi, Andrew Parker Collins, Núria Montagut Colomer, Sonia-Karin Marqués-Kiderle, Camille A Wagner Rodriguez, Jan Christian Holst Chaires, Whendy Avila Motta, Julia Filella-Mercè, Junyi Jessy Li, Fernando Llanos, Nuole Zhu, Sara Rubio-Guerra, Ignacio Illan-Gala, Sergi Borrego-Écija, Albert Lladó, Juan Fortea, Alberto Lleó, Raquel Sánchez-Valle, Maya L Henry, Miguel Ángel Santos Santos, Stephanie M Grasso","doi":"10.64898/2026.09.18.26363435","DOIUrl":"https://doi.org/10.64898/2026.09.18.26363435","url":null,"abstract":"<p><p>Background The nonfluent/agrammatic (nfv) and logopenic (lv) variants of primary progressive aphasia (PPA) disrupt fluency through distinct underlying neurocognitive mechanisms. Differential diagnosis currently requires hours of cognitive-linguistic testing, with additional barriers for bilingual patients due to a shortage of bilingual service providers and a lack of well-established assessment methods. In English speakers, a promising automated approach for differentiating nfvPPA and lvPPA is to derive speech-timing measures and linguistic features from connected speech as input to machine learning (ML) classification algorithms. To our knowledge, this approach has not been evaluated in the context of bilingualism. Methods Thirty four Catalan Spanish simultaneous bilingual patients (lv = 24, nfv = 10) were asked to describe a picture (Western Aphasia Battery Picnic Scene) in both their dominant and non-dominant language. From the participant's recorded response, we derived four feature sets: speech-timing measures, derived with PRAAT; word-level parameters, derived from corpora; linguistic features, derived with the natural language processing tools SpaCy and CLAN; image-text congruence scores, derived with the vision-language encoder Multilingual-CLIP. Each feature set was fed into classification algorithms for differentiating nfv from lv in participants' non-dominant and dominant samples. Then, we combined each feature set's classifier into an ensemble model. We used the McNemar test to determine the statistical significance of differences in classification performance between responses in the non-dominant and dominant language. Results The best-performing classifier achieved F1 macro scores of 93% (word-level parameters) and 92% (ensemble) in the non-dominant and dominant language, respectively. For all feature sets and ensemble models, classification performance did not significantly differ between the non-dominant and dominant language. Ensemble modeling did not significantly improve classification performance in either language. Conclusions Taking advantage of recent advances in multilingual multimodal machine learning, we accurately differentiate Catalan-Spanish bilingual individuals with nfvPPA and lvPPA using a largely automated, time-efficient (1-2 minutes), and ecologically valid connected-speech-based approach. Future directions include evaluating this approach on larger datasets balanced by PPA subtype, using automated transcriptions of connected speech. Our study represents a step towards addressing current inequities in PPA differential diagnosis for non-English-speaking bilingual speakers. Trial registration Data from the clinical trial NCT05741853 was retrospectively analyzed.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13622319/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149121450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mental Stress-Induced Myocardial Ischemia and Cardiovascular Events in Women after a Premature Myocardial Infarction: A Prospective Cohort Study.","authors":"Viola Vaccarino, Amit J Shah, Hua She, Lisa Elon, Tené T Lewis, Yi-An Ko, Nancy Murrah, Lucy H Shallenberger, Tatum Roberts, Lewam Stefanos, Eric Fan-Lou, Bradley Pearce, Zachary T Martin, Jose Medina Inojosa, Kristen Harris, Taha Ahmed, Alexander Razavi, Puja K Mehta, J Douglas Bremner, Paolo Raggi, Arshed Quyyumi","doi":"10.64898/2026.09.30.26364456","DOIUrl":"https://doi.org/10.64898/2026.09.30.26364456","url":null,"abstract":"<p><strong>Background: </strong>Women with premature myocardial infarction (MI) have worse outcomes than similarly aged men despite less severe disease, and stress-related factors have been implicated. We examined sex differences in the association of mental stress-induced myocardial ischemia (MSIMI) with cardiovascular outcomes after premature MI and explored vascular and inflammatory mechanisms.</p><p><strong>Methods: </strong>We studied 602 individuals (mean age 51 years; 46% women) within 8 months of MI who underwent 99mTc-sestamibi myocardial perfusion imaging at rest and after mental stress. MSIMI was defined as a summed difference score ≥3. We measured reactive hyperemia index (RHI) as peripheral microvascular reactivity to stress and interleukin-6 (IL-6) as inflammatory response to stress. The primary endpoint was cardiovascular death or nonfatal MI; the secondary endpoint added heart failure hospitalization. Cox proportional hazards models adjusted for sociodemographic factors, clinical, and psychiatric history.</p><p><strong>Results: </strong>MSIMI was more common in women than men (19.2% vs. 13.2%, p=0.04). Over a median follow-up of 34.2 months, 51 primary and 91 secondary endpoint events occurred. The hazard ratio (HR) of MSIMI for the primary endpoint was 3.2 (95% CI, 1.3, 7.6) in women and 1.9 (95% CI, 0.8, 4.5) in men (p for interaction=0.41), which remained similar after multivariable adjustment. The adjusted population attributable fraction was 27.6% (99% CI, 3.9%, 36.9%) in women and 11.1% (95% CI, -5.8%, 18.4%) in men. In exploratory analyses, all MSIMI-related primary endpoint events among women occurred in Black women (HR, 4.3; 95% CI, 1.7-11.1). Associations with the secondary endpoint were weaker and attenuated after adjustment. IL-6 and RHI responses did not attenuate associations.</p><p><strong>Conclusions: </strong>Among women, MSIMI was associated with about threefold increased risk and over one quarter attributable fraction for cardiovascular death or nonfatal MI. Black women were at especially high risk. While larger studies are warranted to confirm sex differences and explore mechanisms, these data underscore the impact of emotional factors on morbidity and mortality in post-MI women.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648805/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149172243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Beyond Walking Speed: Estimating and Classifying Post-Stroke Gait Propulsion From Standardized Clinical Measures.","authors":"Jie Fei, Ruoxi Wang, Louis N Awad","doi":"10.64898/2026.09.30.26364361","DOIUrl":"https://doi.org/10.64898/2026.09.30.26364361","url":null,"abstract":"<p><strong>Background: </strong>Gait propulsion (anterior ground reaction force) is commonly impaired after stroke. Force plates and wearable sensors can quantify propulsion but are not routinely available in clinical settings. Walking speed alone does not adequately capture the heterogeneity of propulsion deficits.</p><p><strong>Objective: </strong>To determine whether standardized clinical measures can estimate post- stroke propulsion metrics and classify propulsion impairment.</p><p><strong>Methods: </strong>Forty individuals post-stroke participated: 33 for model development and cross-validation (11 with repeated observations for longitudinal validation) and 7 for independent-cohort validation. Clinical measures included the 6-Minute Walk Test (6MWT), comfortable walking speed (CWS), Functional Gait Assessment (FGA), Fugl- Meyer Assessment of the Lower Extremity (FMA-LE), Short Physical Performance Battery (SPPB), Timed Up and Go test (TUG), and stroke chronicity. Regression models were developed to estimate three propulsion metrics: paretic propulsion peak and impulse, and interlimb propulsion symmetry. Model performance was evaluated using R ² and RMSE. Propulsion impairment was classified from symmetry estimates using a prespecified cutoff of 31.5%, with classification accuracy reported.</p><p><strong>Results: </strong>Models using the 6MWT, SPPB, TUG, and chronicity estimated paretic propulsion peak (R ² = 0.77, RMSE = 3.00 %body weight) and impulse (R ² = 0.81, RMSE = 0.65 %body weight·%stance phase). FMA-LE, SPPB, TUG, and chronicity estimated propulsion symmetry (R ² = 0.75, RMSE = 8.56%) and classified propulsion impairment with 84.8% accuracy. Classification accuracy was 91.7% in longitudinal validation and 85.7% in independent-cohort validation.</p><p><strong>Conclusions: </strong>A multidomain clinical assessment framework may support practical and accurate estimation and classification of post-stroke propulsion impairment in clinical settings where specialized instrumentation is unavailable.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13648812/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149167687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}