PLOS digital health最新文献

{"title":"Image Imputation with conditional generative adversarial networks captures clinically relevant imaging features on computed tomography.","authors":"Joseph Rich, Jonathan Le, Ragheb Raad, Tapas Tejura, Ali Rastegarpour, Inderbir Gill, Vinay Duddalwar, Assad Oberai","doi":"10.1371/journal.pdig.0000970","DOIUrl":"10.1371/journal.pdig.0000970","url":null,"abstract":"<p><p>Kidney cancer is among the top 10 most common malignancies in adults, and is commonly evaluated with four-phase computed tomography (CT) imaging. However, the presence of missing or corrupted images remains a significant problem in medical imaging that impairs the detection, diagnosis, and treatment planning of kidney cancer. Deep learning approaches through conditional generative adversarial networks (cGANs) have recently shown technical promise in the task of imputing missing imaging data from these four-phase studies. In this study, we explored the clinical utility of these imputed images. We utilized a cGAN trained on 333 patients, with the task of the cGAN being to impute the image of any phase given the other three phases. We tested the clinical utility on the imputed images of the 37 patients in the test set by manually extracting 21 clinically relevant imaging features and comparing them to their ground truth counterpart. All 13 categorical clinical features had greater than 85% agreement rate between true images and their imputed counterparts. This high accuracy is maintained when stratifying across imaging phases. Imputed images also show good agreement with true images in select radiomic features including mean intensity and enhancement. Imputed images possess the features characteristic of benign or malignant diagnosis at an equivalent rate to true images. In conclusion, imputed images from cGANs have large potential for clinical use due to their ability to retain clinically relevant qualitative and quantitative features.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000970"},"PeriodicalIF":7.7,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12349720/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144850045","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":"Correction: Derivation and validation of an algorithm to predict transitions from community to residential long-term care among persons with dementia-A retrospective cohort study.","authors":"Wenshan Li, Luke Turcotte, Amy T Hsu, Robert Talarico, Danial Qureshi, Colleen Webber, Steven Hawken, Peter Tanuseputro, Douglas G Manuel, Greg Huyer","doi":"10.1371/journal.pdig.0000987","DOIUrl":"10.1371/journal.pdig.0000987","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1371/journal.pdig.0000441.].</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000987"},"PeriodicalIF":7.7,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12349011/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144850044","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":"Nursing activities and associated workload of nurses in virtual care centres: A multicentre observational study.","authors":"Jobbe P L Leenen, Jedidja Lok-Visser, Cindy Vollenbroek, Henk Sonneveld, Thijs Van Houwelingen, Gréanne Leeftink","doi":"10.1371/journal.pdig.0000974","DOIUrl":"10.1371/journal.pdig.0000974","url":null,"abstract":"<p><p>Virtual care centres (VCCs) are novel wards of hospitals and facilitate the provision of remote monitoring and home-based patient care by virtual care nurses. Whereas since the COVID-19 pandemic VCCs have rapidly emerged, there is a lack of insight in virtual care nurses' work and the associated work load. Therefore, the aim of this study was to identify the nursing activities performed in Virtual Care Centers (VCCs) and assess nurses' perceived workload associated with these activities. A multicentre descriptive, observational cross-sectional study was performed. Data collection (February - June 2024) involved three steps: establishing a list of nursing activities, defining and quantifying workload using the NASA-Task Load Index and Analytical Hierarchy Process (AHP), and measuring nursing activity-associated workload by a survey involving 19 virtual care nurses across six VCCs in the Netherlands who had been employed in VCCs for at least one year. Eventually, we identified 21 nursing activities categorized into five areas: education and training (n = 2), development and promotion of new care pathways (n = 4), patient contact (n = 4), clinical decision-making (n = 8), and administration (n = 2). The overall workload was predominantly rated as low to medium, with the development of protocols for new digital care pathways being the most demanding activity. Routine nursing activities, such as patient contact and clinical decision-making, resulted in low to very low workload ratings. In conclusion, we found VCC nurses engage in a broad spectrum of conventional and novel nursing tasks, of which we measured their associated workload using a novel approach integrating NASA-TLX and AHP. The highest associated workload suggest the need for task differentiation and/or additional training to support nurses in managing these high-demand tasks. The VCC model may offer a viable alternative for nurses experiencing high workloads in conventional wards, potentially alleviating some pressures on nursing staff in traditional healthcare settings, mostly in the shift from physical to mental demand.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000974"},"PeriodicalIF":7.7,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12342328/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144839301","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":"Assessing the generalisation of artificial intelligence across mammography manufacturers.","authors":"Alistair J Hickman, Sandra Gomes, Lucy M Warren, Nadia A S Smith, Caroline Shenton-Taylor","doi":"10.1371/journal.pdig.0000973","DOIUrl":"10.1371/journal.pdig.0000973","url":null,"abstract":"<p><p>The aim of this study was to determine whether differences between manufacturer of mammogram images effects performance of artificial intelligence tools for classifying breast density. Processed mammograms from 10,156 women were used to train and validate three deep learning algorithms using three retrospective datasets: Hologic, General Electric, Mixed (equal numbers of Hologic, General Electric and Siemens images) and tested on four independent witheld test sets (Hologic, General Electric, Mixed and Siemens). The area under the receiver operating characteristic curve (AUC) was compared. Women aged 47-73 with normal breasts (routine recall - no cancer) and Volpara ground truth were selected from the OPTIMAM Mammography Image Database for the years 2012-2015. 95 % confidence intervals are used for significance testing in the results with a Bayesian Signed Rank test used to rank the overall performance of the models. Best single test performance is seen when a model is trained and tested on images from a single manufacturer (Hologic train/test: 0.98 and General Electric train/test: 0.97), however the same models performed significantly worse on any other manufacturer images (General Electric AUCs: 0.68 & 0.63; Hologic AUCs: 0.56 & 0.90). The model trained on the mixed dataset exhibited the best overall performance. Better performance occurs when training and test sets contain the same manufacturer distributions and better generalisation occurs when more manufacturers are included in training. Models in clinical use should be trained on data representing the different vendors of mammogram machines used across screening programs. This is clinically relevant as models will be impacted by changes and upgrades to mammogram machines in screening centres.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000973"},"PeriodicalIF":7.7,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12342238/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144839300","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":"Expression of Concern: Imbalanced class distribution and performance evaluation metrics: A systematic review of prediction accuracy for determining model performance in healthcare systems.","authors":"","doi":"10.1371/journal.pdig.0000984","DOIUrl":"10.1371/journal.pdig.0000984","url":null,"abstract":"","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000984"},"PeriodicalIF":7.7,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12333973/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144805419","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":"Social justice and social media: How medical schools display critical consciousness online.","authors":"Eray Yilmaz, Keegan D'Mello, Amrit Kirpalani","doi":"10.1371/journal.pdig.0000981","DOIUrl":"10.1371/journal.pdig.0000981","url":null,"abstract":"<p><p>Academic medical institutions have a pivotal role in addressing the inequalities faced by marginalized populations, especially by promoting values of social justice on online platforms that not only reach the medical sphere, but also the broader public. Central to this transformative agenda is the framework of critical consciousness (CC), which compels individuals to develop an acute awareness of societal inequalities and power dynamics and act as agents of change against inequalities across society. To investigate if and how medical schools use X (formerly Twitter) to display CC, tweets from March 22 - June 22, 2023 from all available Canadian medical school Twitter accounts were obtained and deductively coded. First, a content analysis was performed to collate and categorize tweets related to CC, followed by a critical discourse analysis with a CC framework to examine the role of language in conveying messages about equity and medical education. Of the 3442 tweets reviewed, 554 displayed CC (16.12%). The content analysis revealed that Empowerment of Marginalized Populations was the most prominent display of CC amongst tweets (n = 286), whereas there was a paucity of messaging around Intersectionality (n = 20). The critical discourse analysis revealed that language was purposefully used to positively spotlight equity-deserving individuals (e.g., \"celebrate\" and \"recognize\") with minimal dialogue framing institutions as agents of systemic power differentials. Medical schools ultimately advocate for positive change by sharing awareness-raising content that celebrate marginalized communities. However, the step beyond surface-level awareness-raising content towards critical self-reflection that acknowledged institutions' roles in perpetuating inequities was largely limited; this represents a missed opportunity to leverage the power of social media and engage in meaningful dialogue online to build trust between the healthcare sector and the public.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000981"},"PeriodicalIF":7.7,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12331073/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144801166","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":"Associations between text communication engagement and maternal-neonatal outcomes in the Mobile WACh NEO Trial.","authors":"James Peng, Erica Wetzler, Brenda Wandika, Peninah Kithao, June Moraa, Jenna I Udren, Olivia Schultes, Esther Akinyi, Lusi Osborn, Anna Hedstrom, Barbra A Richardson, Manasi Kumar, Dalton Wamalwa, John Kinuthia, Keshet Ronen, Jennifer A Unger","doi":"10.1371/journal.pdig.0000968","DOIUrl":"10.1371/journal.pdig.0000968","url":null,"abstract":"<p><p>Despite a global reduction in neonatal deaths in the last few decades, high neonatal mortality rates persist in low- to middle-income countries. Mobile health interventions offer a promising solution to promote early newborn care (ENC) practices and improve neonatal health. The Mobile WACh NEO randomized controlled trial evaluated the effect of a text messaging communication intervention on neonatal health outcomes in Kenya from 2020 to 2023. Perinatal participants received automated messages from enrollment at 28-36 weeks gestation until six weeks postpartum and could message with a study nurse. This secondary analysis aimed to characterize participant text engagement and examine associations between engagement and maternal-neonatal health outcomes. Among 2,470 intervention participants retained through follow-up, median time in the intervention was 14 weeks. Participants received a median of 58 automated messages (average 0.58 per day), sent a median of 24 messages (average 0.25 per day), and received a median of 14 nurse responses (average 0.14 per day). Younger, more educated, unmarried, unemployed, and first-time mothers sent more messages, while those who had a lower social support score at baseline messaged less. Increased participant messaging was associated with greater increase in neonatal danger sign knowledge from baseline to six-week follow-up (Adj Est: 0.39; 95% CI: 0.09-0.68) and lower odds of early initiation of breastfeeding (aOR: 0.62; 95% CI: 0.45-0.86). Our findings contribute to the understanding of who can benefit from mobile health programs and how these interventions might impact behaviors and outcomes.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000968"},"PeriodicalIF":7.7,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12331090/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144801137","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":"Impak Sihat: A telehealth system development and feasibility evaluation to empower rural population in Malaysia on the quality use of medicines.","authors":"Nor Ilyani Mohamed Nazar, Norny Syafinaz Ab Rahman, Nor Elina Alias, Syahrir Zaini, Tg Karmila Tg Mohd Kamil, Nurjasmine Aida Jamani, Mohamed Hassan Elnaem","doi":"10.1371/journal.pdig.0000937","DOIUrl":"10.1371/journal.pdig.0000937","url":null,"abstract":"<p><p>The escalating global burden of chronic diseases necessitates innovative approaches to enhance medication adherence and quality use of medicines (QUM), particularly in underserved rural populations. This study developed and evaluated Impak Sihat, a telehealth system tailored to address systemic healthcare barriers in rural Malaysia through a three-phase mixed-methods design. Phase 1 involved qualitative interviews with 15 villagers, revealing smartphone ownership, inconsistent internet connectivity, high social media engagement, and limited critical appraisal of online health information. Phase 2 utilised these insights to create a dual-component system: a public portal with Malay-language educational materials, appointment booking, and a practitioner platform featuring secured patient data management. Phase 3 assessed feasibility via community demonstrations with 77 participants (mean age 53.4 ± 11.8 years), showing high acceptance scores (73-87%) across six domains. Key findings included strong usability (87.0 ± 16.3) and interface design (74.8 ± 23.9), though older adults scored significantly lower on interface design for learnability (ρ=-0.29, p < 0.01). The system's offline functionality and WhatsApp integration mitigated rural connectivity constraints, yet challenges persisted in data confidentiality (lowest score: 73.1 ± 26.7). Healthy participants consistently rated the system significantly higher across multiple domains (Interface Design: p = 0.003, User Experience: p = 0.018, Healthcare Delivery: p = 0.002, and Overall Satisfaction: p = 0.003). These results underscore the potential of context-specific telehealth systems to bridge urban-rural health disparities while highlighting critical implementation barriers. This work highlights the importance of engaging key stakeholders, such as healthcare providers and community leaders, to ensure system sustainability and scalability. Overall, the study demonstrates that digital health interventions, when appropriately tailored to the specific needs of rural populations, can significantly contribute to reducing healthcare disparities and promoting patient empowerment.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000937"},"PeriodicalIF":7.7,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12331124/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144801140","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":"Development and evaluation of large-language models (LLMs) for oncology: A scoping review.","authors":"Namya Mehan, Teshan Dias Desinghe, Ashirbani Saha","doi":"10.1371/journal.pdig.0000980","DOIUrl":"10.1371/journal.pdig.0000980","url":null,"abstract":"<p><p>Large language models (LLMs), a significant development in artificial intelligence (AI), are continuing to demonstrate seminal improvement in performance for various text analysis and generation tasks. There are limited systematic studies on LLM applications that were developed/evaluated in relevance to oncology. Our scoping review explores applications of LLMs in oncology to determine (1) the nature of LLM applications relevant to a cancer/tumor type, (2) the phases of cancer care addressed by the LLMs, (3) which LLMs were used in these applications, (4) the sources and pre-processing of datasets used, (5) the techniques used to optimize the performance of LLMs, (6) the methods of evaluation, and (7) the common limitations noted by the authors of these LLM applications and to study their implications in research and practice. A librarian-assisted search was performed across the following databases: Association for Computing Machinery (ACM), Embase, Engineering Village, IEEE Xplore, Medline, Scopus, SPIE and Web of Science till Jan 12, 2024. Pre-prints from this search were considered if they were published/accepted by Feb 29, 2024. From the initial search of 14863 articles, 60 were finally included. Our results demonstrated that LLMs were mostly evaluated across a diverse set of oncology-related applications. Generative pre-trained transformer (GPT)-based LLMs were mostly used. In the subset of studies where the phase(s) of cancer care was/were provided or implied, treatment and diagnosis were the most included phases. Data for development and evaluation extended from patient health records, synthetic patient records, research and professional society publications to social media. Prompt-designing and engineering were performed as data pre-processing steps in several studies. Clinicians, trainees, researchers, and patients were among the variety of users targeted by the applications. In the17% studies that developed LLMs for oncological aspects, domain adaptation through pre-training and fine-tuning were often performed and resulted in performance improvement. The evaluation of an LLM's performance involved usage of both standard, validated, non-standardized, and/or customized performance measures considering a variety of constructs, other than accuracy. Six primary themes emerged as limitations including limitation of generalizability/applicability, sample size, bias and subjectivity, and evaluation metrics. This review highlights that LLMs, specific to oncological aspects, are less common than general-purpose LLMs. The application areas were heterogeneous, used diverse data sources, were directed towards a variety of users, and resulted in variety of evaluation methods. Despite the diversity of LLM applications in oncology, future research needs to address the limited generalizability of these applications, mitigation of bias and subjectivity, and standardization of evaluation methodologies. Future applications of LLMs in oncology should","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000980"},"PeriodicalIF":7.7,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12331086/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144801138","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":"Health literacy is much more than knowing about health; it also involves the emotions experienced during illness.","authors":"Felipe Cezar Cabral, Maria Eulália Vinadé Chagas, Gabriela de Oliveira Laguna Silva, Guilherme Alcides Flores Soares Rollin, Mariana Giordano Cordoni, Lindayane Debom Motta, Leo Anthony Celi, Taís de Campos Moreira","doi":"10.1371/journal.pdig.0000979","DOIUrl":"10.1371/journal.pdig.0000979","url":null,"abstract":"","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 8","pages":"e0000979"},"PeriodicalIF":7.7,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12331045/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144801139","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}