NPJ Digital Medicine最新文献

{"title":"Large language models-enabled digital twins for precision medicine in rare gynecological tumors","authors":"Jacqueline Lammert, Nicole Pfarr, Leonid Kuligin, Sonja Mathes, Tobias Dreyer, Luise Modersohn, Patrick Metzger, Dyke Ferber, Jakob Nikolas Kather, Daniel Truhn, Lisa Christine Adams, Keno Kyrill Bressem, Sebastian Lange, Kristina Schwamborn, Martin Boeker, Marion Kiechle, Ulrich A. Schatz, Holger Bronger, Maximilian Tschochohei","doi":"10.1038/s41746-025-01810-z","DOIUrl":"https://doi.org/10.1038/s41746-025-01810-z","url":null,"abstract":"<p>Rare gynecological tumors (RGTs) present major clinical challenges due to their low incidence and heterogeneity. The lack of clear guidelines leads to suboptimal management and poor prognosis. Molecular tumor boards accelerate access to effective therapies by tailoring treatment based on biomarkers, beyond cancer type. Unstructured data that requires manual curation hinders efficient use of biomarker profiling for therapy matching. This study explores the use of large language models (LLMs) to construct digital twins for precision medicine in RGTs. Our proof-of-concept digital twin system integrates clinical and biomarker data from institutional and published cases (<i>n</i> = 21) and literature-derived data (<i>n</i> = 655 publications) to create tailored treatment plans for metastatic uterine carcinosarcoma, identifying options potentially missed by traditional, single-source analysis. LLM-enabled digital twins efficiently model individual patient trajectories. Shifting to a biology-based rather than organ-based tumor definition enables personalized care that could advance RGT management and thus enhance patient outcomes.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"12 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144586783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The overlooked role of exergames in cognitive-motor neurorehabilitation: a systematic review","authors":"Maria Grazia Maggio, Francesca Baglio, Raffaela Maione, Rosalia Calapai, Fulvia Di Iulio, Paulo Cezar Rocha dos Santos, Marcos Maldonado-Díaz, Giulia Pistorino, Antonio Cerasa, Angelo Quartarone, Rocco Salvatore Calabrò","doi":"10.1038/s41746-025-01843-4","DOIUrl":"https://doi.org/10.1038/s41746-025-01843-4","url":null,"abstract":"<p>Exergames are emerging tools for cognitive-motor neurorehabilitation, but their effects in adults with neurological conditions remain insufficiently explored. This systematic review examined studies from PubMed, Scopus, Embase, and Web of Science, assessing the impact of exergames on motor and cognitive outcomes. Eligible studies included randomized controlled trials (RCTs), non-RCTs, and experimental designs. The protocol was registered on PROSPERO (CRD420250655053) and followed PRISMA and Cochrane guidelines. Eleven studies (8 RCTs, 2 feasibility studies, 1 secondary analysis) reported improvements in balance, gait, executive function, and memory, with high adherence (85–100%) and minimal adverse effects. While the overall risk of bias was low, heterogeneity in interventions, populations, and outcome measures limited comparability and generalizability. Additionally, the absence of long-term follow-up hindered conclusions on sustained benefits. Exergames appear promising for cognitive-motor rehabilitation in neurological conditions. Future studies should adopt standardized protocols, include long-term follow-up, and explore neurophysiological mechanisms to support clinical implementation.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"8 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144586781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Population-scale cross-sectional observational study for AI-powered TB screening on one million CXRs","authors":"Prateek Munjal, Ahmed Al Mahrooqi, Ronnie Rajan, Andrew Jeremijenko, Iftikhar Ahmad, Muhammad Imran Akhtar, Marco A. F. Pimentel, Shadab Khan","doi":"10.1038/s41746-025-01832-7","DOIUrl":"https://doi.org/10.1038/s41746-025-01832-7","url":null,"abstract":"<p>Traditional tuberculosis (TB) screening involves radiologists manually reviewing chest X-rays (CXR), which is time-consuming, error-prone, and limited by workforce shortages. Our AI model, <b>AIRIS-TB</b> (<b>AI R</b>adiology <b>I</b>n <b>S</b>creening <b>TB</b>), aims to address these challenges by automating the reporting of all X-rays without any findings. AIRIS-TB was evaluated on over one million CXRs, achieving an AUC of 98.51% and overall false negative rate (FNR) of 1.57%, outperforming radiologists (1.85%) while maintaining a 0% TB-FNR. By selectively deferring only cases with findings to radiologists, the model has the potential to automate up to 80% of routine CXR reporting. Subgroup analysis revealed insignificant performance disparities across age, sex, HIV status, and region of origin, with sputum tests for suspected TB showing a strong correlation with model predictions. This large-scale validation demonstrates AIRIS-TB’s safety and efficiency in high-volume TB screening programs, reducing radiologist workload without compromising diagnostic accuracy.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"10 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144586827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of intervention design on engagement and outcomes in digital self-help for insomnia - factorial RCT.","authors":"Amira Hentati,Nils Hentati Isacsson,Ann Rosén,Susanna Jernelöv,Viktor Kaldo,Brjánn Ljótsson,Erik Forsell,Nils Lindefors,Martin Kraepelien","doi":"10.1038/s41746-025-01839-0","DOIUrl":"https://doi.org/10.1038/s41746-025-01839-0","url":null,"abstract":"Digital self-help can improve access to mental health care, but poor engagement limits effectiveness. This single-blind 2 x 2 x 2 factorial randomized controlled trial examined whether an optimized graphical user interface (GUI), automated reminders (AR), and an adaptive treatment strategy (ATS) improved engagement and outcomes in a digital self-help insomnia intervention. Adults (N = 447) with moderate to severe insomnia were randomized to combinations of the factors. The GUI improved self-rated engagement, sleep log activity, login frequency, and usability. AR increased sleep log activity and logins, while ATS improved satisfaction. All three combined significantly improved insomnia symptoms (d = 0.50). No severe adverse effects were reported. Clinicians spent 13.74 min on average on the ATS. Statistical analyses included linear and multilevel regression. Factors were effect coded. Intervention design can enhance engagement and outcomes, requiring minimal clinician time. Pre-registered 2023-04-11 (ClinicalTrials.gov, NCT05826002). Funded by the Swedish Ministry of Health and Social Affairs, grant number: S2018/03855/FS.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"72 1","pages":"416"},"PeriodicalIF":15.2,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144586510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital transformation with clinical alerts and personalized care systems in an integrated value based model","authors":"Dinh Nguyen, Sinjin Lee, Ronil Synghal, Leon Chan, Ferdinand Justus, Mark Moromisato, Tianyuan Shao, Caleb Wang, Mason Kellogg, Brett Anwar, Julie Creech, Kari Ochoa, Danette Gigliotti, Omar Ortiz, Kien La, Khang Nguyen","doi":"10.1038/s41746-025-01838-1","DOIUrl":"https://doi.org/10.1038/s41746-025-01838-1","url":null,"abstract":"<p>Patient portals are widely available to facilitate self-service interactions, including appointment booking, offering convenience to users. Promoting clinically appropriate care pathways, however, is complex; portals must recognize patient intent while striving for an intuitive experience. In October 2024, the Southern California Permanente Medical Group, which serves 4.9 M patients, deployed the Kaiser Permanente Intelligent Navigator (KPIN), a system that augments the patient portal to enhance care navigation and the patient experience. It applies natural language processing to generate alerts for high-acuity cases, and it recommends suitable care offerings. Early findings are promising, with the AUC for clinical alerts and clinical navigation at 0.977 and 0.889, respectively. KPIN’s adjusted successful booking rate was 53.68%, with an abandonment rate of 2.94% (IQR: 2.77–3.11%), aligning with patient survey results showing an 8.63 percentage point increase for positive sentiment. These results highlight the success of KPIN in an integrated value-based care delivery model.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"21 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144578130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Framework for bias evaluation in large language models in healthcare settings","authors":"Tara Templin, Sophia Fort, Prasanna Padmanabham, Pratyush Seshadri, Ram Rimal, Junier Oliva, Kristin Hassmiller Lich, Sean Sylvia, Nasa Sinnott-Armstrong","doi":"10.1038/s41746-025-01786-w","DOIUrl":"https://doi.org/10.1038/s41746-025-01786-w","url":null,"abstract":"<p>A critical gap in the adoption of large language models for AI-assisted clinical decisions is the lack of a standardized audit framework to evaluate models for accuracy and bias. Our framework introduces a five-step framework that guides practitioners through stakeholder engagement, model calibration to specific patient populations, and rigorous testing through clinically relevant scenarios. We provide open-access tools for stakeholder engagement and an example of an audit. As the regulation of models becomes more critical, we believe adoption of an audit framework that tests model outputs, rather than regulating specific hyperparameters or inputs, will encourage the responsible use of AI in clinical settings.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"21 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144578262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characteristics, licensing, and ethical considerations of openly accessible oral-maxillofacial imaging datasets: a systematic review","authors":"Jing Hao, Andrew Nalley, Andy Wai Kan Yeung, Ray Tanaka, Qi Yong H. Ai, Walter Yu Hang Lam, Zhiyi Shan, Yiu Yan Leung, Abeer AlHadidi, Michael M. Bornstein, James Kit Hon Tsoi, Colman McGrath, Kuo Feng Hung","doi":"10.1038/s41746-025-01818-5","DOIUrl":"https://doi.org/10.1038/s41746-025-01818-5","url":null,"abstract":"<p>Several open-source oral-maxillofacial imaging datasets have been created but their characteristics, ethical clearance, and licensing for reuse remain unclear. This study aimed to systematically identify these datasets and investigate their characteristics, ethical approvals, and licensing requirements for reuse. Open-source oral-maxillofacial imaging datasets were identified through electronic databases and dataset platforms. 105 datasets with 437538 images and 100 intraoral videos from patients across twenty-one countries were included. The datasets comprise imaging modalities, including photographs, periapical, panoramic, and cephalometric radiographs, CBCT, MRI, surface scans, videos, and histopathological images. Nearly 80% of them provide annotations, but only 25.7% specified the annotators’ qualification. The majority (83.8%) did not disclose whether ethical approval was obtained, while 61.9% specified terms or licenses for dataset reuse. There is an urgent need to develop standardized guidelines for reusing image datasets and to establish AI-specific consents to fully inform patients about potential uses of their data in AI projects.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"12 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144565782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic literature review on integrating AI-powered smart glasses into digital health management for proactive healthcare solutions","authors":"Boyuan Wang, Ying Zheng, Xihao Han, Liang Kong, Gexin Xiao, Zunxiong Xiao, Shanji Chen","doi":"10.1038/s41746-025-01715-x","DOIUrl":"https://doi.org/10.1038/s41746-025-01715-x","url":null,"abstract":"<p>AI-powered smart glasses are emerging as a highly promising advancement in the field of digital health management, owing to their capabilities in real-time monitoring, chronic disease management, and personalized treatment planning. To comprehensively understand the current state of development, we systematically searched multiple databases, including Web of Science, PubMed, and IEEE Xplore, to collect relevant literature. This paper provides a systematic analysis of the current applications of smart glasses in healthcare, focusing on their potential benefits and limitations. Key issues discussed include user engagement, treatment adherence, data privacy, standardization, battery efficiency, clinical validation, and medical ethics. Our findings suggest that, supported by emerging clinical evidence, smart glasses have demonstrated significant improvements in areas such as assisted medical services, health management, anxiety alleviation in children, and telemedicine. By integrating multi-modal sensors, these devices are capable of accurately tracking certain physiological indicators and synchronizing real-time visual input, thereby enhancing the accuracy and timeliness of health interventions and medical services. Notably, some cutting-edge smart glasses have adopted advanced artificial intelligence algorithms, particularly large language models (LLMs) with context awareness and human-like interaction capabilities. These AI-powered glasses can offer real-time, personalized dietary and health management recommendations tailored to users’ daily life scenarios. Building on these findings, this study further proposes a conceptual framework for proactive health management using smart glasses and explores future directions in technological development and practical applications. Overall, AI-enhanced smart glasses show great potential as a critical interface between healthcare providers and patients, poised to play a vital role in the future of personalized medicine and continuous health management.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"26 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144565818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review of machine learning approaches in cochlear implant outcomes","authors":"Anu Prasad Sreenivasan Nair, Srikanta K. Mishra, Pedro Andres Alba Diaz","doi":"10.1038/s41746-025-01733-9","DOIUrl":"https://doi.org/10.1038/s41746-025-01733-9","url":null,"abstract":"<p>Cochlear implants (CIs) have transformed the lives of over one million individuals with hearing impairment, including children as young as nine months. This systematic review critically examines the current literature on the application of machine learning (ML) techniques for predicting CI outcomes. A comprehensive search identified 20 relevant studies. Imaging-based studies demonstrated high predictive accuracy for language and speech perception outcomes. Neural function measures provided a feasible way to assess the functional status of the auditory nerve, while clinical and audiological predictors were extensively explored through data mining techniques. Additionally, ML-based speech enhancement algorithms showed promise in improving speech recognition in noisy environments, a major challenge for CI users. Despite these advancements, a significant gap remains in developing models that can be directly integrated into CI programming. Integrating ML into CIs— in areas like signal processing and device programming—holds immense potential to support personalized patient care for hearing-impaired individuals.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"17 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144565784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The relation between passively collected data and PTSD: a systematic review and meta-analysis","authors":"Ningzhe Zhu, Anjali Sarawgi, Markus Bühner, Harald Baumeister, Patricia Garatva, Thomas Ehring, Yannik Terhorst","doi":"10.1038/s41746-025-01825-6","DOIUrl":"https://doi.org/10.1038/s41746-025-01825-6","url":null,"abstract":"<p>Post-traumatic stress disorder (PTSD) is a common mental disorder. This systematic review and meta-analysis examined the association between mobile sensing features and PTSD symptoms. Studies were sourced from the Database for Mobile Sensing Studies in Mental Healthcare (DAMOS), with inclusion criteria requiring correlations between mobile sensing data and PTSD symptoms assessed by validated tools. Seventeen studies encompassing 1847 participants (mean age = 38.68, 63.18% female) remained after study selection. Of 18 features across sleep, mobility, activity, and social activity, only wake after sleep onset (<i>r</i> = 0.14, 95% CI = [0.03, 0.25]) and relative amplitude of physical activity (<i>r</i> = −0.10, 95% CI = [−0.17, −0.03]) were significantly associated with PTSD symptoms. Findings were consistent across PTSD measurements, populations, demographics, and sensing durations. Although mobile sensing offers unobtrusive, objective, and ecologically valid insights into PTSD, confirmatory studies and research to optimize sensor assessment are needed before clinical practice.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"273 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144565781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}