Communications medicine最新文献

{"title":"Circulating EV-microRNAs are dynamic biomarkers of resistance to therapeutic immunomodulation in metastatic melanoma.","authors":"Elena Splendiani, Zein Mersini Besharat, Claudia Sabato, Teresa Maria Rosaria Noviello, Maria Fortunata Lofiego, Vincenzo D'Alonzo, Monica Valente, Laura Solmonese, Alessia Covre, Sandra Coral, Francesca Pia Caruso, Tanja Milena Autilio, Ines Simeone, Agnese Po, Anna Citarella, Giuseppina Catanzaro, Roberta Mortarini, Andrea Anichini, Michele Maio, Michele Ceccarelli, Anna Maria Di Giacomo, Elisabetta Ferretti","doi":"10.1038/s43856-026-01629-z","DOIUrl":"https://doi.org/10.1038/s43856-026-01629-z","url":null,"abstract":"<p><strong>Background: </strong>Melanoma is the most aggressive skin cancer, with a 50% five-year mortality in metastatic or unresectable cases. Non-invasive biomarkers are crucial for guiding treatment. MicroRNAs (miRNAs), especially those encapsulated in extracellular vesicles (EVs), are stable in plasma and hold promise as biomarkers.</p><p><strong>Methods: </strong>This study analyzed EV-associated miRNAs (EV-miRNAs) from 50 blood samples of 18 stage III-IV melanoma patients treated with anti-CTLA-4 immunotherapy and a DNA hypomethylating agent. Samples were collected at baseline, week 4, and week 12. Patients were classified as responders (R) or non-responders (NR).</p><p><strong>Results: </strong>A baseline signature of four EV-miRNAs predicted primary resistance. Treatment altered 15 EV-miRNAs at week 4 and 51 at week 12; nine were consistently modulated. At week 12, 27 EV-miRNAs differed between NR and R, with miR-1203 and miR-566-3p up-regulated in NR, linked to resistance and poor survival.</p><p><strong>Conclusions: </strong>These results highlight EV-miRNAs as non-invasive biomarkers for predicting and monitoring therapy response.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Combining ADA, IFN-γ, and IL-6 as host biomarkers enhances diagnosis of paucibacillary and extrapulmonary tuberculosis.","authors":"Zhan Qiu Mao, Jia Mao, Huilie Zheng, Qian Zhong Liu, Xiao Mei Tang, Qi Long Zhang","doi":"10.1038/s43856-026-01580-z","DOIUrl":"https://doi.org/10.1038/s43856-026-01580-z","url":null,"abstract":"<p><strong>Background: </strong>Accurate diagnosis of paucibacillary and extrapulmonary tuberculosis remains a major global challenge, as conventional pathogen-focused methods often lack sensitivity. This study aimed to evaluate whether combining host-derived immune biomarkers could improve diagnostic accuracy.</p><p><strong>Methods: </strong>In a retrospective diagnostic accuracy study of 605 participants, nucleic acid amplification tests served as the reference standard. We assessed the performance of adenosine deaminase, interferon-gamma, and interleukin-6 individually and in combination using receiver operating characteristic analysis. Serial (rule-in) and parallel (rule-out) testing algorithms were compared to define strategies for confirmation or exclusion of disease.</p><p><strong>Results: </strong>Here we show that while interferon-gamma was the strongest single discriminator, combinations of biomarkers significantly outperform any single marker. The dual-panel of interferon-gamma and interleukin-6 achieves the highest overall accuracy. Serial testing of adenosine deaminase and interferon-gamma maximizes specificity to 97.4% (95% CI: 95.6-98.7%), providing a reliable rule-in tool. Conversely, parallel testing of all three biomarkers achieves 94.9% (95% CI: 95.6-98.7%)sensitivity, enabling an effective rule-out strategy.</p><p><strong>Conclusions: </strong>The integration of complementary host biomarkers overcomes the inherent limitations of single analyte tests. By enabling adaptable, context-specific algorithms tailored for either high-specificity confirmation or high-sensitivity screening, this approach provides a scalable and cost-effective diagnostic approach, tailored for resource-constrained settings.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":"6 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosing and defining MASLD in people living with chronic hepatitis B.","authors":"Emily Martyn, Alejandro Arenas-Pinto, Richard Gilson, Nomathemba Chandiwana, Stuart Flanagan, Douglas MacDonald, Emmanuel A Tsochatzis, W D Francois Venter, Jennifer Manne-Goehler, Philippa C Matthews","doi":"10.1038/s43856-026-01383-2","DOIUrl":"https://doi.org/10.1038/s43856-026-01383-2","url":null,"abstract":"<p><p>Chronic hepatitis B (CHB) infection and metabolic dysfunction-associated steatotic liver disease (MASLD) are important contributors to the growing worldwide burden of liver disease. There is limited understanding regarding the interaction between CHB and MASLD. This is a consequence of the changing terminology for liver disease, inconsistent application of diagnostic tools, and poor understanding of global populations. In this review, we collate data on the use of diagnostic tests for identifying MASLD and associated liver inflammation or fibrosis in people living with CHB. We advocate for improved consensus on diagnosis, evidence-based monitoring and risk stratification, enhanced access to interventions and reduced health inequity.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":"6 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning prediction of nocturnal hypertension for patients intolerant to ambulatory blood pressure monitoring.","authors":"Yifan Lin, Mingwei Wen, Peiying Sun, Junjun Sun, Hao Yang, Yanfang Wang","doi":"10.1038/s43856-026-01639-x","DOIUrl":"https://doi.org/10.1038/s43856-026-01639-x","url":null,"abstract":"<p><strong>Background: </strong>Ambulatory blood pressure monitoring (ABPM) plays an irreplaceable role in the diagnosis and management of hypertension. However, more than 100 million of the 1.4 billion people with hypertension worldwide cannot tolerate nighttime ABPM due to noise and arm compression. Previous prediction methods relying on demographic factors and home blood pressure measurements are time-consuming and burdensome while exhibiting limited accuracy for nocturnal hypertension. There is a need for a more accurate, low-burden approach to identify high-risk patients intolerant to nighttime ABPM monitoring.</p><p><strong>Methods: </strong>We collected 2,874 ABPM records at a regional medical center to conduct a retrospective cohort study. Kernel density estimation based preprocessing was applied to stabilize data fluctuations. A variational autoencoder based deep learning model was developed using daytime blood pressure and heart rate combined with full-day activity and posture states to predict nocturnal hypertension.</p><p><strong>Results: </strong>Here we show that the ABPM-VAE model achieves an AUC of 0.82 (95% CI 0.77-0.88) on the test set, outperforming the ablation model (AUC 0.67; 95% CI 0.61-0.74; p < 0.001) and prior methods based on demographic and home blood pressure data (AUC 0.69). For nocturnal hypertension prediction, the model yields a PPV of 92.12%, NPV of 55.20%, sensitivity of 0.73, and specificity of 0.84.</p><p><strong>Conclusions: </strong>The entropy reduction preprocessing-enhanced deep learning model predicts nocturnal hypertension risk from ABPM without adding burden to patients or physicians. It serves as an effective screening tool to identify high-risk individuals intolerant to nighttime monitoring, serving as a valuable complement to conventional ABPM.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147857841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sequential versus upfront oxaliplatin-based therapy in metastatic colorectal cancer: long-term outcomes of a randomized phase 3 trial.","authors":"Makoto Okawaki, Mototsugu Shimokawa, Ryo Inada, Hitoshi Ojima, Hiroaki Tanioka, Shingo Noura, Yoshinori Munemoto, Keiichiro Ishibashi, Yoshiaki Shindo, Madoka Hamada, Masasumi Okajima, Yoshiyuki Yamaguchi, Takeshi Yamada, Yasuhiro Shimada, Takeshi Nagasaka","doi":"10.1038/s43856-026-01633-3","DOIUrl":"https://doi.org/10.1038/s43856-026-01633-3","url":null,"abstract":"<p><strong>Background: </strong>Whether oxaliplatin should be given sequentially or upfront in metastatic colorectal cancer remains clinically relevant, especially for older patients often underrepresented in trials.</p><p><strong>Methods: </strong>We report updated overall survival and prespecified quality-of-life outcomes from the open-label, randomized, multicenter phase 3 C-cubed trial in Japan. This trial was registered in the University Hospital Medical Information Network Clinical Trials Registry UMIN000015405 (October 14, 2014) and UMIN000036690 (May 10, 2019). The primary endpoint was reported previously. Of 311 randomized patients with previously untreated metastatic colorectal cancer, 300 were included in the full analysis set. Patients received fluoropyrimidine plus bevacizumab, with oxaliplatin added at progression, or received fluoropyrimidine, oxaliplatin, and bevacizumab from the start. Overall survival was analyzed using Cox regression and time-restricted survival methods. Quality of life was assessed prospectively using patient-reported questionnaires.</p><p><strong>Results: </strong>Here we show that overall survival is similar between strategies. Median overall survival is 27.2 months with sequential treatment and 27.4 months with upfront treatment (hazard ratio, 1.00; 95% confidence interval, 0.76-1.33; p = 0.98). Additional time-restricted and milestone analyses show no between-group differences. There is no evidence that age modifies the relative treatment effect. Patient-reported outcomes indicate a lower early treatment burden with sequential treatment, including smaller early declines in physical functioning and less sensory neuropathy.</p><p><strong>Conclusions: </strong>Updated overall survival is comparable between sequential and upfront oxaliplatin-based strategies, while sequential treatment is associated with lower early treatment burden. These findings support individualized selection of first-line treatment based on age, vulnerability, and treatment goals.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846976","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Networked SIRS model with Kalman filter state estimation for epidemic monitoring in Europe.","authors":"Atte Aalto, Daniele Proverbio, Giulia Giordano, Alexander Skupin, Jorge Gonçalves","doi":"10.1038/s43856-026-01611-9","DOIUrl":"https://doi.org/10.1038/s43856-026-01611-9","url":null,"abstract":"<p><strong>Background: </strong>Metapopulation models, which consider epidemic spread across interconnected regions, can provide more accurate epidemic predictions compared to isolated models for the corresponding regions. Still, their added complexity and data requirements raise questions about their tangible benefits over simpler, localized models.</p><p><strong>Methods: </strong>We develop and validate two networked compartmental metapopulation models for predicting influenza-like illness across Europe: a detailed network-based model, including international travel dynamics, and a simpler mean-field model, aggregating average regional data. The network is constructed using public mobility data and complemented with population densities at border regions. Incidence data of influenza-like illnesses from 28 countries are integrated using an Extended Kalman filter.</p><p><strong>Results: </strong>We show that networked epidemic models effectively capture epidemic dynamics across regions and epidemic phases. The models enable accurate forecasts, missing data imputation, and actionable insights: network models outperform isolated models in forecasting epidemic progression, particularly during critical periods such as wave onsets and peaks, and maintain reliability in scenarios with missing data.</p><p><strong>Conclusions: </strong>The findings unveil and quantify the advantages of metapopulation models for epidemic forecasting in interconnected regions, and pave the way to the integration of mobility and epidemic surveillance to improve the monitoring and prediction of spreading diseases.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intra-operative application of ultra-high frequency ultrasound facilitates differentiation of bowel wall characteristics between ganglionic and aganglionic segments during transanal endorectal pull-through for Hirschsprung disease.","authors":"Christina Granéli, Maria Evertsson, Tobias Erlöv, Tebin Hawez, Kristine Hagelsteen, Louise Tofft, Tomas Jansson, Magnus Cinthio, Pernilla Stenström","doi":"10.1038/s43856-026-01632-4","DOIUrl":"https://doi.org/10.1038/s43856-026-01632-4","url":null,"abstract":"<p><strong>Background: </strong>The hypothesis was that ultra-high-frequency (UHF) ultrasound can distinguish ganglionic from aganglionic bowel during surgery for Hirschsprung disease. The aim was to assess UHF ultrasound for differentiating ganglionosis from aganglionosis in children undergoing rectosigmoid Hirschsprung disease surgery.</p><p><strong>Methods: </strong>The child's intestine was examined intra-operatively using the Vevo MD (Fujifilm VisualSonics, Toronto, ON, Canada) system equipped with the UHF70 transducer; 50 MHz in center frequency. Collected images were analyzed and results presented as median.</p><p><strong>Results: </strong>Twenty-one patients were examined intra-operatively with UHF ultrasound. The muscularis interna was thicker in ganglionic bowel compared to aganglionic: 0.298 vs 0.599 (p < 0.001), and the ratio of the muscularis interna/muscularis externa was greater; 0.621 vs 1.225 (p < 0.001). The echogenicity was higher, i.e., whiter, in the aganglionic submucosa 104.5 vs 81.6 (p < 0.016).</p><p><strong>Conclusion: </strong>The use of intra-operative UHF ultrasound shows great promise in the determination of ganglionic versus aganglionic bowel.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":"6 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive cross-cohort analysis reveals global gut microbiome signatures of celiac disease.","authors":"Peter J Prendergast, Haig V Bishop, Craig W Herbold, Elena F Verdu, Renwick C J Dobson, Andrew S Day, Olivia J Ogilvie","doi":"10.1038/s43856-026-01627-1","DOIUrl":"https://doi.org/10.1038/s43856-026-01627-1","url":null,"abstract":"<p><strong>Background: </strong>Celiac disease affects ~1-2% of people and remains incurable, requiring lifelong dietary restriction. The gut microbiome is thought to contribute to the development and progression of celiac disease. However, findings across previous studies are fragmented, making it difficult to understand exactly how the gut microbiome is involved.</p><p><strong>Methods: </strong>We integrate over 900 samples from global datasets spanning different disease stages (before onset, during active disease, and after treatment), body sites, and research methods. Datasets produced using both 16S rRNA gene sequencing and shotgun metagenomics profile the gut microbiome. Alpha and beta diversity analyses and differential abundance testing identify consistent changes in bacterial communities linked to celiac disease. Machine learning tests how well microbiome data predicts disease status.</p><p><strong>Results: </strong>Here, we show that celiac disease is not marked by large changes in gut microbiome diversity. Instead, there are subtle, consistent changes in specific bacteria, including a reduction in beneficial butyrate producers (Faecalibacterium, Prevotella, Agathobacter, Gemmiger), changes in mucin-associated microbes (Akkermansia muciniphila), and an increase in potentially harmful bacteria (Helicobacter, Campylobacter, Haemophilus parainfluenzae). These changes are seen before and during active disease and persist on a gluten-free diet. Microbiome-based disease prediction is moderately accurate for active disease and weaker for prospective performance, likely constrained by training data.</p><p><strong>Conclusions: </strong>Our findings suggest that celiac disease is linked to specific changes in gut bacteria that are not fully resolved by diet alone. Future treatments may need to focus on restoring healthy gut bacteria, not just avoiding gluten, to better manage the disease.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Temporal trends in pregnancy outcomes during a health system shock.","authors":"Florence Tydeman, Kathryn V Dalrymple, Alice McGreevy, Lucilla Poston, Tisha Dasgupta, Abigail Easter, Asma Khalil, Lisa Long, Hiten D Mistry, Daghni Rajasingam, Sergio A Silverio, Marina Soley-Bori, Yanzhong Wang, Sara L White, Peter von Dadelszen, Laura A Magee","doi":"10.1038/s43856-026-01493-x","DOIUrl":"https://doi.org/10.1038/s43856-026-01493-x","url":null,"abstract":"<p><strong>Background: </strong>To better understand reported COVID-19 pandemic effects on pregnancy, we examined temporal trends in pregnancy outcomes in a diverse population from South London, United Kingdom.</p><p><strong>Methods: </strong>We included 31,411 singleton pregnancies with complete registration and birth outcomes across pre-pandemic (May 1, 2019-March 22, 2020, 24.5%), pandemic lockdowns (March 23, 2020-July 17, 2021, 32.3%), and pandemic without lockdown epochs (July 18, 2021-April 22, 2023, 43.2%). Multivariable regression was employed to evaluate outcomes by study epoch, adjusting for potential confounders (e.g., ethnicity, deprivation, site), followed by generalized additive modelling to visualise monthly trends.</p><p><strong>Results: </strong>Here we show that of 17 outcomes: six have stable trends (e.g., preterm birth, stillbirth); eight show linear trends, either decreasing (e.g., gestational age at birth, vaginal tears) or increasing (e.g., Caesareans, postpartum haemorrhage); and three show quadratic (complex) trends (e.g., secondary mental health services, labour induction).</p><p><strong>Conclusions: </strong>Overall, most outcomes during the pandemic mirror pre-pandemic trends, with fluctuations observed likely due to site-specific responses. Our findings highlight the need for temporal analysis of pregnancy outcomes.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Validation of remote multimodal AI screening for Parkinson disease across diverse settings.","authors":"Md Saiful Islam, Tariq Adnan, Abdelrahman Abdelkader, Zipei Liu, Evelyn Ma, Sooyong Park, Asif Azad, Pai Liu, Meghan Pawlik, Emily Hartman, Erin Shelton, Kristina B Larson, M Saifur Rahman, Cathe Schwartz, Karen Jaffe, Jamie L Adams, Ruth B Schneider, Jan Freyberg, E Ray Dorsey, Ehsan Hoque","doi":"10.1038/s43856-026-01606-6","DOIUrl":"https://doi.org/10.1038/s43856-026-01606-6","url":null,"abstract":"<p><strong>Background: </strong>Timely detection of Parkinson's disease (PD) remains limited by reliance on in-person neurological evaluations that are often costly and geographically inaccessible. To address these barriers, we develop PARK (Parkinson's Analysis with Remote Kinetic-tasks) - a web-based artificial intelligence (AI) tool that screens for PD using short webcam recordings of facial expression, motor, and speech tasks.</p><p><strong>Methods: </strong>Across eight independent studies (n = 1,865 participants; 670 with PD), participants completed three standardized tasks (smile mimicry, finger tapping, and pangram utterance) via webcam. Task-specific neural networks estimate PD risk and uncertainty, which are integrated through an uncertainty-calibrated fusion model (UFNet). Model performance is evaluated on one internal and two external test sets representing supervised and unsupervised real-world environments. Three movement disorder specialists also reviewed videos from 30 participants to benchmark clinical agreement of the PARK tool. User experience is assessed through structured surveys containing open-ended or multiple-choice questions.</p><p><strong>Results: </strong>PARK achieves accuracies of 80.2-80.6% and AUROC of 0.85-0.87 across all evaluation cohorts, with 83.3-86.5% sensitivity and 71.2-78.4% specificity. Predictive performance remains stable across sex, age, and ethnicity. Agreement with clinician judgments reaches Cohen's κ = 0.59. Uncertainty estimates reflect diagnostic confidence, and performance declines at high-uncertainty levels. Usability is rated highly (System Usability Scale  > 70) in both supervised and unsupervised settings, with low perceived risk and strong user preference for remote screening.</p><p><strong>Conclusions: </strong>PARK demonstrates promising accuracy and favorable user acceptance for remote PD screening, highlighting its potential as an accessible, equitable, and uncertainty-aware tool for neurological assessment when traditional care is challenging to obtain.</p>","PeriodicalId":72646,"journal":{"name":"Communications medicine","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147846297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}