NPJ Digital Medicine最新文献_第6页

The FAIIR conversational AI agent assistant for youth mental health service provision 为青少年心理健康服务提供的fair会话人工智能代理助理

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-03 DOI: 10.1038/s41746-025-01647-6

Stephen Obadinma, Alia Lachana, Maia Leigh Norman, Jocelyn Rankin, Joanna Yu, Xiaodan Zhu, Darren Mastropaolo, Deval Pandya, Roxana Sultan, Elham Dolatabadi

{"title":"The FAIIR conversational AI agent assistant for youth mental health service provision","authors":"Stephen Obadinma, Alia Lachana, Maia Leigh Norman, Jocelyn Rankin, Joanna Yu, Xiaodan Zhu, Darren Mastropaolo, Deval Pandya, Roxana Sultan, Elham Dolatabadi","doi":"10.1038/s41746-025-01647-6","DOIUrl":"https://doi.org/10.1038/s41746-025-01647-6","url":null,"abstract":"Frontline crisis support plays a critical role in youth mental health services, where Crisis Responders (CRs) engage in conversations and assign issue tags to guide interventions. To enhance this process, we introduce FAIIR (Frontline Assistant: Issue Identification and Recommendation), an ensemble of domain-adapted transformer models trained on 780,000 conversations. FAIIR aims to reduce CR’s cognitive burden, enhance issue identification accuracy, and streamline post-conversation administrative tasks. Evaluated on retrospective data, FAIIR achieves an average AUC ROC of 94%, an average F1-score of 64%, and an average recall score of 81%. During the silent testing phase, its performance remained robust, with less than a 2% drop in all metrics. CRs exhibited 90.9% agreement with its predictions, and expert agreement with FAIIR exceeded their agreement with original labels. These findings highlight FAIIR’s potential to assist CRs in prioritizing urgent cases and ensuring appropriate resource allocation in crisis interventions.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"18 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143901346","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

Latent retinal structural patterns with aging 随年龄增长的潜在视网膜结构模式

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-03 DOI: 10.1038/s41746-025-01632-z

Kei Sano, Kota Fukai, Ryo Terauchi, Yuko Furuya, Shoko Nakazawa, Kosuke Sakai, Toru Honda, Yuya Watanabe, Takeshi Hayashi, Toru Nakagawa, Tadashi Nakano, Masayuki Tatemichi

{"title":"Latent retinal structural patterns with aging","authors":"Kei Sano, Kota Fukai, Ryo Terauchi, Yuko Furuya, Shoko Nakazawa, Kosuke Sakai, Toru Honda, Yuya Watanabe, Takeshi Hayashi, Toru Nakagawa, Tadashi Nakano, Masayuki Tatemichi","doi":"10.1038/s41746-025-01632-z","DOIUrl":"https://doi.org/10.1038/s41746-025-01632-z","url":null,"abstract":"Optical coherence tomography (OCT) is an efficient tool for non-invasively evaluating retinal structures. Retinal thinning changes assessed using OCT are recognized as potential biomarkers for systemic conditions such as Alzheimer’s disease, Parkinson’s disease, and chronic kidney disease. However, age-related retinal changes remain largely unexplored, complicating the differentiation between physiological and pathological alterations. Here, we introduced a highly granular approach to assess age-related spatial changes in the inner retina using latent retinal archetypes, identifying 36 retinal archetypes of macula and peripapillary sector images from 189,387 OCTs of 22,494 individuals. Subsequently, we evaluated the associations between these archetypes and age; age-related archetypes are characterized by total or superior thinning in the macula sector. Among individuals with myopia, the inferior thinning pattern in the macula ganglion cell–inner plexiform layer was associated with aging. The age-related effects in the peripapillary sector were primarily reflected in the shape of retinal artery trajectories. Overall, latent retinal archetypes would offer new avenues for the effective utilization of retinal biomarkers in age-related diseases.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"112 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143901353","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

Systematic evaluation of patient-reported outcomes in clinical trials of digital health in cardiovascular diseases 心血管疾病数字健康临床试验中患者报告结果的系统评估

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-02 DOI: 10.1038/s41746-025-01637-8

Qianying Wang, Xiqian Huo, Jianlan Cui, Jie Li, Xueke Bai, Chaoqun Wu, Qian Zhang, Dianjianyi Sun, Jie Zhao

{"title":"Systematic evaluation of patient-reported outcomes in clinical trials of digital health in cardiovascular diseases","authors":"Qianying Wang, Xiqian Huo, Jianlan Cui, Jie Li, Xueke Bai, Chaoqun Wu, Qian Zhang, Dianjianyi Sun, Jie Zhao","doi":"10.1038/s41746-025-01637-8","DOIUrl":"https://doi.org/10.1038/s41746-025-01637-8","url":null,"abstract":"The integration of patient-reported outcomes (PROs) in digital health (DH) trials for cardiovascular diseases (CVDs) remains unknown. We searched ClinicalTrials.gov for randomized trials that tested DH interventions in CVDs from 2004 to 2024. The search identified 8037 trials, with 673 eligible trials included in the analysis. Among these, 321 trials (48%) incorporated PROs. The number of DH trials and the use of PROs have shown a significant upward trend. Phone-based interventions predominated (38%), mostly targeting hypertension (38%) and heart failure (27%). Behavioral interventions showed higher prevalence of PROs’ usage (1.24 [1.04–1.48]), while trials for diagnostic or screening purpose (0.39 [0.20–0.77]) utilized PROs less frequently. Only 15% of trials reported results on ClinicalTrials.gov, while 58% were published in PubMed after completion. Despite DH trial expansion, PRO integration remains insufficient, especially in trials where clinical and patient perspectives are important in informing treatment decisions. Timely results dissemination is critical to improving transparency.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"5 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143901355","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

Leveraging long context in retrieval augmented language models for medical question answering 在医学问题回答的检索增强语言模型中利用长上下文

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-02 DOI: 10.1038/s41746-025-01651-w

Gongbo Zhang, Zihan Xu, Qiao Jin, Fangyi Chen, Yilu Fang, Yi Liu, Justin F. Rousseau, Ziyang Xu, Zhiyong Lu, Chunhua Weng, Yifan Peng

{"title":"Leveraging long context in retrieval augmented language models for medical question answering","authors":"Gongbo Zhang, Zihan Xu, Qiao Jin, Fangyi Chen, Yilu Fang, Yi Liu, Justin F. Rousseau, Ziyang Xu, Zhiyong Lu, Chunhua Weng, Yifan Peng","doi":"10.1038/s41746-025-01651-w","DOIUrl":"https://doi.org/10.1038/s41746-025-01651-w","url":null,"abstract":"While holding great promise for improving and facilitating healthcare through applications of medical literature summarization, large language models (LLMs) struggle to produce up-to-date responses on evolving topics due to outdated knowledge or hallucination. Retrieval-augmented generation (RAG) is a pivotal innovation that improves the accuracy and relevance of LLM responses by integrating LLMs with a search engine and external sources of knowledge. However, the quality of RAG responses can be largely impacted by the rank and density of key information in the retrieval results, such as the “lost-in-the-middle” problem. In this work, we aim to improve the robustness and reliability of the RAG workflow in the medical domain. Specifically, we propose a map-reduce strategy, BriefContext, to combat the “lost-in-the-middle” issue without modifying the model weights. We demonstrated the advantage of the workflow with various LLM backbones and on multiple QA datasets. This method promises to improve the safety and reliability of LLMs deployed in healthcare domains by reducing the risk of misinformation, ensuring critical clinical content is retained in generated responses, and enabling more trustworthy use of LLMs in critical tasks such as medical question answering, clinical decision support, and patient-facing applications.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"55 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143901356","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

A systematic review of quantum machine learning for digital health 量子机器学习用于数字健康的系统综述

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-02 DOI: 10.1038/s41746-025-01597-z

Riddhi S. Gupta, Carolyn E. Wood, Teyl Engstrom, Jason D. Pole, Sally Shrapnel

引用次数: 0

Insights into endometriosis symptom trajectories and assessment of surgical intervention outcomes using longitudinal actigraphy 子宫内膜异位症的症状轨迹和手术干预结果的纵向活动成像评估

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-02 DOI: 10.1038/s41746-025-01629-8

Katherine Edgley, Philippa T. K. Saunders, Lucy H. R. Whitaker, Andrew W. Horne, Athanasios Tsanas

{"title":"Insights into endometriosis symptom trajectories and assessment of surgical intervention outcomes using longitudinal actigraphy","authors":"Katherine Edgley, Philippa T. K. Saunders, Lucy H. R. Whitaker, Andrew W. Horne, Athanasios Tsanas","doi":"10.1038/s41746-025-01629-8","DOIUrl":"https://doi.org/10.1038/s41746-025-01629-8","url":null,"abstract":"Endometriosis is a common, chronic condition associated with debilitating pain, fatigue, and heterogeneous symptom presentation. In this exploratory study, 68 participants with confirmed endometriosis were monitored for up to three 4–6-week smartwatch cycles. We collected daily self-reports of pain and fatigue as well as retrospective questionnaires assessing quality of life, and we extracted daily measures of physical activity (PA), sleep, and diurnal rhythms from wrist-worn actigraphy data. We found that daily PA was strongly negatively correlated with self-reported fatigue (repeated measures correlations (R < -0.3)) and that participants with more severe or variable symptom trajectories displayed lower levels of PA, greater sleep disturbance, and more disrupted sleep and activity rhythms (Spearman’s ({|R|} > 0.3)). Lastly, we found evidence of sleep and PA changes following surgery for endometriosis that reflected change in self-reported symptoms. Collectively, our findings suggest that passive data collection using wrist-worn wearables in endometriosis could facilitate individualized objective insights into symptom trajectories.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"95 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898136","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

Small language models learn enhanced reasoning skills from medical textbooks 小语言模型从医学教科书中学习增强的推理技能

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-02 DOI: 10.1038/s41746-025-01653-8

Hyunjae Kim, Hyeon Hwang, Jiwoo Lee, Sihyeon Park, Dain Kim, Taewhoo Lee, Chanwoong Yoon, Jiwoong Sohn, Jungwoo Park, Olga Reykhart, Thomas Fetherston, Donghee Choi, Soo Heon Kwak, Qingyu Chen, Jaewoo Kang

{"title":"Small language models learn enhanced reasoning skills from medical textbooks","authors":"Hyunjae Kim, Hyeon Hwang, Jiwoo Lee, Sihyeon Park, Dain Kim, Taewhoo Lee, Chanwoong Yoon, Jiwoong Sohn, Jungwoo Park, Olga Reykhart, Thomas Fetherston, Donghee Choi, Soo Heon Kwak, Qingyu Chen, Jaewoo Kang","doi":"10.1038/s41746-025-01653-8","DOIUrl":"https://doi.org/10.1038/s41746-025-01653-8","url":null,"abstract":"Small language models (SLM) offer promise for medical applications by addressing the privacy and hardware constraints of large language models; however, their limited parameters (often fewer than ten billion) hinder multi-step reasoning for complex medical tasks. This study presents Meerkat, a new family of medical SLMs designed to be lightweight while enhancing reasoning capabilities. We begin by designing an effective and efficient training method. This involves extracting high-quality chain-of-thought reasoning paths from 18 medical textbooks, which are then combined with diverse instruction-following datasets within the medical domain, totaling 441K training examples. Fine-tuning was conducted on open-source SLMs using this curated dataset. Our Meerkat-7B and Meerkat-8B models outperformed their counterparts by 22.3% and 10.6% across six exam datasets, respectively. They also improved scores on the NEJM Case Challenge from 7 to 16 and from 13 to 20, surpassing the human score of 13.7. Additionally, they demonstrated superiority in expert evaluations, excelling in all metrics—completeness, factuality, clarity, and logical consistency—of reasoning abilities.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"34 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143901354","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

Deep learning HRNet FCN for blood vessel identification in laparoscopic pancreatic surgery 深度学习HRNet FCN用于腹腔镜胰腺手术血管识别

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-01 DOI: 10.1038/s41746-025-01663-6

Jile Shi, Ruohan Cui, Zhihong Wang, Qi Yan, Lu Ping, Hu Zhou, Junyi Gao, Chihua Fang, Xianlin Han, Surong Hua, Wenming Wu

{"title":"Deep learning HRNet FCN for blood vessel identification in laparoscopic pancreatic surgery","authors":"Jile Shi, Ruohan Cui, Zhihong Wang, Qi Yan, Lu Ping, Hu Zhou, Junyi Gao, Chihua Fang, Xianlin Han, Surong Hua, Wenming Wu","doi":"10.1038/s41746-025-01663-6","DOIUrl":"https://doi.org/10.1038/s41746-025-01663-6","url":null,"abstract":"Laparoscopic pancreatic surgery remains highly challenging due to the complexity of the pancreas and surrounding vascular structures, with risk of injuring critical blood vessels such as the Superior Mesenteric Vein (SMV)-Portal Vein (PV) axis and splenic vein. Here, we evaluated the High Resolution Network (HRNet)-Full Convolutional Network (FCN) model for its ability to accurately identify vascular contours and improve surgical safety. Using 12,694 images from 126 laparoscopic distal pancreatectomy (LDP) videos and 35,986 images from 138 Whipple procedure videos, the model demonstrated robust performance, achieving a mean Dice coefficient of 0.754, a recall of 85.00%, and a precision of 91.10%. By combining datasets from LDP and Whipple procedures, the model showed strong generalization across different surgical contexts and achieved real-time processing speeds of 11 frames per second during surgery process. These findings highlight HRNet-FCN’s potential to recognize anatomical landmarks, enhance surgical precision, reduce complications, and improve laparoscopic pancreatic outcomes.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"34 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143893485","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

Impact of missing electronic fetal monitoring signals on perinatal asphyxia: a multicohort analysis 缺失胎儿电子监测信号对围产期窒息的影响：一项多队列分析

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-01 DOI: 10.1038/s41746-025-01665-4

Debjyoti Karmakar, Lochana Mendis, Emerson Keenan, Marimuthu Palaniswami, Roxanne Hastie, Enes Makalic, Fiona Brownfoot

{"title":"Impact of missing electronic fetal monitoring signals on perinatal asphyxia: a multicohort analysis","authors":"Debjyoti Karmakar, Lochana Mendis, Emerson Keenan, Marimuthu Palaniswami, Roxanne Hastie, Enes Makalic, Fiona Brownfoot","doi":"10.1038/s41746-025-01665-4","DOIUrl":"https://doi.org/10.1038/s41746-025-01665-4","url":null,"abstract":"Cardiotocography (CTG) is essential for monitoring high-risk pregnancies, yet perinatal asphyxia prediction accuracy remains limited to 50–55%. Regions of artifacts (missing valid signals)-including signal processing aberrations-possibly contribute to this limitation, highlighted by 40% of FDA reports on intrapartum stillbirths. This cohort study applied causal inference to two digitized CTG databases, analyzing 36,792 labor episodes (>36 weeks) at a tertiary Australian hospital (2010–2021) and externally validating on a Czech dataset (n = 552).High rates of missing valid signals (>30% fetal heart rate signal dropout or >1% maternal-fetal heart rate coincidence) was independently associated with asphyxia (aOR 1.47, 95% CI 1.19–1.81); dropout >30% showing stronger link (aOR 1.58, 95% CI 1.13–2.20 Australian dataset; aOR 2.30, 95% CI 1.08–4.91 Czech dataset). Risk of asphyxia increased with higher dropout (>37.45%, aOR 2.21 Australian dataset; >34.01%, aOR 4.08 Czech dataset). Integrating measures of missing valid signals into fetal monitoring algorithms may improve decision-making and neonatal outcomes.","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"11 1","pages":""},"PeriodicalIF":15.2,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143893484","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

Collection of the digital data from the neurological examination 收集神经系统检查的数字数据

IF 15.2 1区医学

NPJ Digital Medicine Pub Date : 2025-05-01 DOI: 10.1038/s41746-025-01659-2

Bruno Kusznir Vitturi, Patrik Theodor Nerdal, Walter Maetzler

引用次数: 0