NPJ Digital Medicine最新文献

Engaging an advisory board in discussions about the ethical relevance of algorithmic bias and fairness 聘请顾问委员会讨论算法偏见和公平性的伦理相关性

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-18 DOI: 10.1038/s41746-025-01711-1

Laura Brandt, Larry Au, Clinton Castro, Gabriel J. Odom

引用次数: 0

Fair ultrasound diagnosis via adversarial protected attribute aware perturbations on latent embeddings 基于潜在嵌入的对抗性保护属性感知扰动的公平超声诊断

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01641-y

Zikang Xu, Fenghe Tang, Quan Quan, Qingsong Yao, Qingpeng Kong, Jianrui Ding, Chunping Ning, S. Kevin Zhou

{"title":"Fair ultrasound diagnosis via adversarial protected attribute aware perturbations on latent embeddings","authors":"Zikang Xu, Fenghe Tang, Quan Quan, Qingsong Yao, Qingpeng Kong, Jianrui Ding, Chunping Ning, S. Kevin Zhou","doi":"10.1038/s41746-025-01641-y","DOIUrl":"https://doi.org/10.1038/s41746-025-01641-y","url":null,"abstract":"Deep learning techniques have significantly enhanced the convenience and precision of ultrasound image diagnosis, particularly in the crucial step of lesion segmentation. However, recent studies reveal that both train-from-scratch models and pre-trained models often exhibit performance disparities across sex and age attributes, leading to biased diagnoses for different subgroups. In this paper, we propose APPLE, a novel approach designed to mitigate unfairness without altering the parameters of the base model. APPLE achieves this by learning fair perturbations in the latent space through a generative adversarial network. Extensive experiments on both a publicly available dataset and an in-house ultrasound image dataset demonstrate that our method improves segmentation and diagnostic fairness across all sensitive attributes and various backbone architectures compared to the base models. Through this study, we aim to highlight the critical importance of fairness in medical segmentation and contribute to the development of a more equitable healthcare system.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"30 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067150","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}

引用次数: 0

FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting FedWeight：通过患者重加权减轻联邦学习对电子病历数据的协变量偏移

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01661-8

He Zhu, Jun Bai, Na Li, Xiaoxiao Li, Dianbo Liu, David L. Buckeridge, Yue Li

引用次数: 0

Hybrid machine learning for real-time prediction of edema trajectory in large middle cerebral artery stroke 混合机器学习实时预测大脑大中动脉卒中水肿轨迹

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01687-y

Ethan Phillips, Odhran O’Donoghue, Yumeng Zhang, Panos Tsimpos, Leigh Ann Mallinger, Stefanos Chatzidakis, Jack Pohlmann, Yili Du, Ivy Kim, Jonathan Song, Benjamin Brush, Stelios Smirnakis, Charlene J. Ong, Agni Orfanoudaki

{"title":"Hybrid machine learning for real-time prediction of edema trajectory in large middle cerebral artery stroke","authors":"Ethan Phillips, Odhran O’Donoghue, Yumeng Zhang, Panos Tsimpos, Leigh Ann Mallinger, Stefanos Chatzidakis, Jack Pohlmann, Yili Du, Ivy Kim, Jonathan Song, Benjamin Brush, Stelios Smirnakis, Charlene J. Ong, Agni Orfanoudaki","doi":"10.1038/s41746-025-01687-y","DOIUrl":"https://doi.org/10.1038/s41746-025-01687-y","url":null,"abstract":"In treating malignant cerebral edema after a large middle cerebral artery stroke, clinicians need quantitative tools for real-time risk assessment. Existing predictive models typically estimate risk at one, early time point, failing to account for dynamic variables. To address this, we developed Hybrid Ensemble Learning Models for Edema Trajectory (HELMET) to predict midline shift severity, an established indicator of malignant edema, over 8-h and 24-h windows. The HELMET models were trained on retrospective data from 623 patients and validated on 63 patients from a different hospital system, achieving mean areas under the receiver operating characteristic curve of 96.6% and 92.5%, respectively. By integrating transformer-based large language models with supervised ensemble learning, HELMET demonstrates the value of combining clinician expertise with multimodal health records in assessing patient risk. Our approach provides a framework for accurate, real-time estimation of dynamic clinical targets using human-curated and algorithm-derived inputs.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"9 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067146","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}

引用次数: 0

The DRAGON benchmark for clinical NLP 临床NLP的DRAGON基准

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01626-x

Joeran S. Bosma, Koen Dercksen, Luc Builtjes, Romain André, Christian Roest, Stefan J. Fransen, Constant R. Noordman, Mar Navarro-Padilla, Judith Lefkes, Natália Alves, Max J. J. de Grauw, Leander van Eekelen, Joey M. A. Spronck, Megan Schuurmans, Bram de Wilde, Ward Hendrix, Witali Aswolinskiy, Anindo Saha, Jasper J. Twilt, Daan Geijs, Jeroen Veltman, Derya Yakar, Maarten de Rooij, Francesco Ciompi, Alessa Hering, Jeroen Geerdink, Henkjan Huisman

{"title":"The DRAGON benchmark for clinical NLP","authors":"Joeran S. Bosma, Koen Dercksen, Luc Builtjes, Romain André, Christian Roest, Stefan J. Fransen, Constant R. Noordman, Mar Navarro-Padilla, Judith Lefkes, Natália Alves, Max J. J. de Grauw, Leander van Eekelen, Joey M. A. Spronck, Megan Schuurmans, Bram de Wilde, Ward Hendrix, Witali Aswolinskiy, Anindo Saha, Jasper J. Twilt, Daan Geijs, Jeroen Veltman, Derya Yakar, Maarten de Rooij, Francesco Ciompi, Alessa Hering, Jeroen Geerdink, Henkjan Huisman","doi":"10.1038/s41746-025-01626-x","DOIUrl":"https://doi.org/10.1038/s41746-025-01626-x","url":null,"abstract":"Artificial Intelligence can mitigate the global shortage of medical diagnostic personnel but requires large-scale annotated datasets to train clinical algorithms. Natural Language Processing (NLP), including Large Language Models (LLMs), shows great potential for annotating clinical data to facilitate algorithm development but remains underexplored due to a lack of public benchmarks. This study introduces the DRAGON challenge, a benchmark for clinical NLP with 28 tasks and 28,824 annotated medical reports from five Dutch care centers. It facilitates automated, large-scale, cost-effective data annotation. Foundational LLMs were pretrained using four million clinical reports from a sixth Dutch care center. Evaluations showed the superiority of domain-specific pretraining (DRAGON 2025 test score of 0.770) and mixed-domain pretraining (0.756), compared to general-domain pretraining (0.734, p < 0.005). While strong performance was achieved on 18/28 tasks, performance was subpar on 10/28 tasks, uncovering where innovations are needed. Benchmark, code, and foundational LLMs are publicly available.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"57 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067148","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}

引用次数: 0

Development and prospective implementation of a large language model based system for early sepsis prediction 基于语言模型的脓毒症早期预测系统的开发和预期实现

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01689-w

Supreeth P. Shashikumar, Sina Mohammadi, Rishivardhan Krishnamoorthy, Avi Patel, Gabriel Wardi, Joseph C. Ahn, Karandeep Singh, Eliah Aronoff-Spencer, Shamim Nemati

{"title":"Development and prospective implementation of a large language model based system for early sepsis prediction","authors":"Supreeth P. Shashikumar, Sina Mohammadi, Rishivardhan Krishnamoorthy, Avi Patel, Gabriel Wardi, Joseph C. Ahn, Karandeep Singh, Eliah Aronoff-Spencer, Shamim Nemati","doi":"10.1038/s41746-025-01689-w","DOIUrl":"https://doi.org/10.1038/s41746-025-01689-w","url":null,"abstract":"Sepsis is a dysregulated host response to infection with high mortality and morbidity. Early detection and intervention have been shown to improve patient outcomes, but existing computational models relying on structured electronic health record data often miss contextual information from unstructured clinical notes. This study introduces COMPOSER-LLM, an open-source large language model (LLM) integrated with the COMPOSER model to enhance early sepsis prediction. For high-uncertainty predictions, the LLM extracts additional context to assess sepsis-mimics, improving accuracy. Evaluated on 2500 patient encounters, COMPOSER-LLM achieved a sensitivity of 72.1%, positive predictive value of 52.9%, F-1 score of 61.0%, and 0.0087 false alarms per patient hour, outperforming the standalone COMPOSER model. Prospective validation yielded similar results. Manual chart review found 62% of false positives had bacterial infections, demonstrating potential clinical utility. Our findings suggest that integrating LLMs with traditional models can enhance predictive performance by leveraging unstructured data, representing a significant advance in healthcare analytics.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"122 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067153","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}

引用次数: 0

Social determinants of health extraction from clinical notes across institutions using large language models 使用大型语言模型从各机构的临床记录中提取健康的社会决定因素

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-17 DOI: 10.1038/s41746-025-01645-8

Vipina K. Keloth, Salih Selek, Qingyu Chen, Christopher Gilman, Sunyang Fu, Yifang Dang, Xinghan Chen, Xinyue Hu, Yujia Zhou, Huan He, Jungwei W. Fan, Karen Wang, Cynthia Brandt, Cui Tao, Hongfang Liu, Hua Xu

{"title":"Social determinants of health extraction from clinical notes across institutions using large language models","authors":"Vipina K. Keloth, Salih Selek, Qingyu Chen, Christopher Gilman, Sunyang Fu, Yifang Dang, Xinghan Chen, Xinyue Hu, Yujia Zhou, Huan He, Jungwei W. Fan, Karen Wang, Cynthia Brandt, Cui Tao, Hongfang Liu, Hua Xu","doi":"10.1038/s41746-025-01645-8","DOIUrl":"https://doi.org/10.1038/s41746-025-01645-8","url":null,"abstract":"Detailed social determinants of health (SDoH) is often buried within clinical text in EHRs. Most current NLP efforts for SDoH have limitations, investigating limited factors, deriving data from a single institution, using specific patient cohorts/note types, with reduced focus on generalizability. We aim to address these issues by creating cross-institutional corpora and developing and evaluating the generalizability of classification models, including large language models (LLMs), for detecting SDoH factors using data from four institutions. Clinical notes were annotated with 21 SDoH factors at two levels: level 1 (SDoH factors only) and level 2 (SDoH factors and associated values). Compared to other models, instruction tuned LLM achieved top performance with micro-averaged F1 over 0.9 on level 1 corpora and over 0.84 on level 2 corpora. While models performed well when trained and tested on individual datasets, cross-dataset generalization highlighted remaining obstacles. Access to trained models will be made available at https://github.com/BIDS-Xu-Lab/LLMs4SDoH.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"41 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067159","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}

引用次数: 0

Monitoring the opioid epidemic via social media discussions 通过社交媒体讨论监测阿片类药物的流行

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-15 DOI: 10.1038/s41746-025-01642-x

Delaney A. Smith, Adam Lavertu, Aadesh Salecha, Tymor Hamamsy, Keith Humphreys, Anna Lembke, Mathew V. Kiang, Russ B. Altman, Johannes C. Eichstaedt

引用次数: 0

Reducing outpatient visits for FIT-positive participants of colorectal cancer screening programs with home-based digital counselling 通过家庭数字咨询减少结直肠癌筛查项目fit阳性参与者的门诊就诊

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-15 DOI: 10.1038/s41746-025-01683-2

Fleur E. Marijnissen, Elyse E. C. Rijnders, Merel M. Tielemans, Désiree van Noord, Leonieke M. M. Wolters, Jeroen M. Jansen, Ingrid Schot, Frank C. Bekkering, Agnes N. Reijm, Sophia M. van Baalen, Tingting Wang, Marijke Melles, Richard Goossens, Sohal Y. Ismail, Iris Lansdorp–Vogelaar, Pieter Jan F. de Jonge, Manon C. W. Spaander

{"title":"Reducing outpatient visits for FIT-positive participants of colorectal cancer screening programs with home-based digital counselling","authors":"Fleur E. Marijnissen, Elyse E. C. Rijnders, Merel M. Tielemans, Désiree van Noord, Leonieke M. M. Wolters, Jeroen M. Jansen, Ingrid Schot, Frank C. Bekkering, Agnes N. Reijm, Sophia M. van Baalen, Tingting Wang, Marijke Melles, Richard Goossens, Sohal Y. Ismail, Iris Lansdorp–Vogelaar, Pieter Jan F. de Jonge, Manon C. W. Spaander","doi":"10.1038/s41746-025-01683-2","DOIUrl":"https://doi.org/10.1038/s41746-025-01683-2","url":null,"abstract":"Digital counselling can alleviate the burden on healthcare systems and patients. While it has been evaluated as a supplement to standard care or a substitute for follow-up visits, its use for initial triaging and counselling remains unstudied. We developed a Digital Intake Tool (DIT) to facilitate the entire pre-colonoscopy counselling process for FIT-positive participants of a colorectal cancer screening program digitally, replacing the need for physicians. In this multicentre prospective non-inferiority study, we evaluated if the DIT could replace in-person counselling. DIT-counselling resulted in adequately prepared participants in 96.5%, compared to 97.6% after in-person counselling, demonstrating non-inferiority. Outpatient visits were significantly reduced, with only 3.4% requiring face-to-face consultations. Patient experiences were highly positive, without increased psychological distress or anxiety, and effective knowledge transfer. This approach benefits patients and healthcare systems, allowing patients to receive care at home, reducing travel and carbon emissions, while increasing outpatient capacity. ICTRP-registration: NL9315, March 8, 2021.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"52 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143979463","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}

引用次数: 0

EMR usability and patient safety: a national survey of physicians 电子病历可用性和患者安全：一项全国医生调查

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-15 DOI: 10.1038/s41746-025-01657-4

David Schwappach, Wolf Hautz, Gert Krummrey, Yvonne Pfeiffer, Raj M. Ratwani

{"title":"EMR usability and patient safety: a national survey of physicians","authors":"David Schwappach, Wolf Hautz, Gert Krummrey, Yvonne Pfeiffer, Raj M. Ratwani","doi":"10.1038/s41746-025-01657-4","DOIUrl":"https://doi.org/10.1038/s41746-025-01657-4","url":null,"abstract":"Despite widespread adoption of electronic medical records (EMRs), concerns persist regarding their usability and implications for patient safety. This national cross-sectional survey assessed physicians’ perceptions of EMR usability across safety-relevant domains. Among 1933 respondents from diverse care settings, 56% reported that their EMR did not enhance patient safety, and 50% perceived their system as inefficient. Usability ratings averaged 52% of the maximum score. Statistically significant differences were observed between EMRs in outpatient (η² = 0.13) and hospital (η² = 0.37) settings. Multilevel modeling attributed 38% of the variance in usability ratings to differences between EMRs, 51% to hospital-level variation within EMRs, and 11% to physician-level differences. Canonical discriminant analysis identified key differentiating usability features, including system response times, excessive alerts, prevention of data entry errors, and support for collaboration. These findings underscore substantial limitations in current EMR systems and reinforce the value of comparative usability assessments to inform targeted improvements in digital health infrastructure.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"29 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143979394","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}

引用次数: 0