PLOS digital health最新文献

{"title":"Radiomics analysis for the early diagnosis of common sexually transmitted infections and skin lesions.","authors":"Jiajun Sun, Zhen Yu, Yingping Li, Janet M Towns, Lin Zhang, Jason J Ong, Zongyuan Ge, Christopher K Fairley, Lei Zhang","doi":"10.1371/journal.pdig.0000926","DOIUrl":"10.1371/journal.pdig.0000926","url":null,"abstract":"<p><p>Early identification of sexually transmitted infection (STI) symptoms can prevent subsequent complications and improve STI control. We analysed 597 images from STIAtlas and categorised the images into four typical STIs and two skin lesions by the anatomical sites of infections. We first applied nine image filters and 11 machine-learning image classifiers to the images. We then extracted radiomics features from the filtered images and trained them with 99 models that combined image filters and classifiers. Model performance was evaluated by area under curve (AUC) and permutation importance. When the information of infection sites was unspecified, a combined Gradient-Boosted Decision Trees (GBDT) classifier and Laplacian of Gaussian (LoG) filter model achieved the best overall performance with an average AUC of 0.681 (95% CI 0.628-0.734). This model predicted best for lichen sclerosus (AUC = 0.768, 0.740-0.796). The incorporation of infection site information led to a substantial improvement in the model's performance, with 22.3% improvement for anal infections (AUC = 0.833, 0.687-0.979) and 3.8% for skin infections (AUC = 0.707, 0.608-0.806). Lesion texture and statistical radiomics features were the most predictive for STIs. Combining machine learning and radiomics techniques is an effective method to categorise skin lesions associated with STIs clinically.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000926"},"PeriodicalIF":7.7,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12286352/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144700630","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: Neurological diagnoses in hospitalized COVID-19 patients associated with adverse outcomes: A multinational cohort study.","authors":"Meghan R Hutch, Jiyeon Son, Trang T Le, Chuan Hong, Xuan Wang, Zahra Shakeri Hossein Abad, Michele Morris, Alba Gutiérrez-Sacristán, Jeffrey G Klann, Anastasia Spiridou, Ashley Batugo, Riccardo Bellazzi, Vincent Benoit, Clara-Lea Bonzel, William A Bryant, Lorenzo Chiudinelli, Kelly Cho, Priyam Das, Tomás González González, David A Hanauer, Darren W Henderson, Yuk-Lam Ho, Ne Hooi Will Loh, Adeline Makoudjou, Simran Makwana, Alberto Malovini, Bertrand Moal, Danielle L Mowery, Antoine Neuraz, Malarkodi Jebathilagam Samayamuthu, Fernando J Sanz Vidorreta, Emily R Schriver, Petra Schubert, Jeffery Talbert, Amelia L M Tan, Byorn W L Tan, Bryce W Q Tan, Valentina Tibollo, Patric Tippman, Guillaume Verdy, William Yuan, Paul Avillach, Nils Gehlenborg, Gilbert S Omenn, Shyam Visweswaran, Tianxi Cai, Yuan Luo, Zongqi Xia","doi":"10.1371/journal.pdig.0000957","DOIUrl":"10.1371/journal.pdig.0000957","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1371/journal.pdig.0000484.].</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000957"},"PeriodicalIF":7.7,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12282889/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144692705","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":"Community-acquired pneumonia identification from electronic health records in the absence of a gold standard: A Bayesian latent class analysis.","authors":"Jia Wei, Kevin Yuan, Augustine Luk, A Sarah Walker, David W Eyre","doi":"10.1371/journal.pdig.0000936","DOIUrl":"10.1371/journal.pdig.0000936","url":null,"abstract":"<p><p>Community-acquired pneumonia (CAP) is common and a significant cause of mortality. However, CAP surveillance commonly relies on diagnostic codes from electronic health records (EHRs), with imperfect accuracy. We used Bayesian latent class models with multiple imputation to assess the accuracy of CAP diagnostic codes in the absence of a gold standard and to explore the contribution of various EHR data sources in improving CAP identification. Using 491,681 hospital admissions in Oxfordshire, UK, from 2016 to 2023, we investigated four EHR-based algorithms for CAP detection based on 1) primary diagnostic codes, 2) clinician-documented indications for antibiotic prescriptions, 3) radiology free-text reports, and 4) vital signs and blood tests. The estimated prevalence of CAP as the reason for emergency hospital admission was 13.6% (95% credible interval 13.3-14.0%). Primary diagnostic codes had low sensitivity but a high specificity (best fitting model, 0.275 and 0.997 respectively), as did vital signs with blood tests (0.348 and 0.963). Antibiotic indication text had a higher sensitivity (0.590) but a lower specificity (0.982), with radiology reports intermediate (0.485 and 0.960). Defining CAP as present when detected by any algorithm produced sensitivity and specificity of 0.873 and 0.905 respectively. Results remained consistent using alternative priors and in sensitivity analyses. Relying solely on diagnostic codes for CAP surveillance leads to substantial under-detection; combining EHR data across multiple algorithms enhances identification accuracy. Bayesian latent class analysis-based approaches could improve CAP surveillance and epidemiological estimates by integrating multiple EHR sources, even without a gold standard for CAP diagnosis.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000936"},"PeriodicalIF":0.0,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12279105/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144683755","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":"Morphology and calcification characterization in patients undergoing TAVI: A 3D statistical shape modelling study.","authors":"Raphaël Sivera, Ebba Montgomery-Liljeroth, Yaxi Chen, Silvia Schievano, Jan Brüning, Wouter Huberts, Anthony Mathur, Andrew Cook, Kush Patel, Claudio Capelli","doi":"10.1371/journal.pdig.0000564","DOIUrl":"10.1371/journal.pdig.0000564","url":null,"abstract":"<p><p>Aortic stenosis (AS) is a common valvular disease becoming more prevalent globally due to the aging of the population. Transcatheter aortic valve implantation (TAVI) is a minimally invasive intervention indicated for AS patients as alternative to surgical replacement. TAVI is to date an established procedure. However, it has been often associated with complications such as paravalvular leakage (PVL) or conduction abnormalities. Evidence of associations between morphological features of the aortic root, valve calcification measurements and suboptimal procedural outcomes have been suggested but the analyses were limited by availability and reproducibility of clinical measurements. In this work, we aim to enrich the characterization of AS patients referred for TAVI by analyzing the clinical findings in conjunction with advanced morphological analysis of the implantation site including aortic root, left ventricular outflow tract and 3D calcification patterns. A population of consecutive patients with AS (n = 130) who underwent TAVI at our clinical centre were retrospectively selected for this study. Demographic and clinical measurements were collected before and after TAVI. Pre-operative CT images were used to reconstruct 3D models of patient-specific anatomies. Statistical shape modelling was carried out and outcomes were analyzed in conjunction with clinical outcomes. The 3D modelling of the valve calcification rate matched previous clinical descriptions; including the crescent shapes visible on each leaflet and the higher calcification rate of the non-coronary cusp. Higher calcification rate was found in larger valves together with a positive association between each coronary height and the calcification of their respective leaflet. Sexual dimorphism, on both shape and calcification, was recorded beyond the size differences with straighter aortas and higher calcification rate at the junction between the left and right coronary leaflets for males compared to females. Morphological differences were significantly associated (p = 0.005) with PVL assessments based on post-operative echocardiograms. Larger aortas and shorter left coronary sinus were associated with less leakage. The outcome distribution appeared to be directly affected by sexual differences and device design. Female phenotypes, smaller and more conic aortic root, were associated with worse outcome. Different patterns in calcification distribution on the leaflets were identified but the association with outcomes is not conclusive. In the future, the presented morphological characterization of patients with AS could contribute to predict post-TAVI PVL and design and test improved TAVI devices.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000564"},"PeriodicalIF":0.0,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12279150/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144683757","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":"Emulating real-world GLP-1 efficacy in type 2 diabetes through causal learning and virtual patients.","authors":"Calum Robert MacLellan, Hristo Petkov, Conor McKeag, Feng Dong, David John Lowe, Roma Maguire, Sotiris Moschoyiannis, Jo Armes, Simon Skene, Alastair Finlinson, Christopher Sainsbury","doi":"10.1371/journal.pdig.0000927","DOIUrl":"10.1371/journal.pdig.0000927","url":null,"abstract":"<p><p>Randomized controlled trials (RCTs) remain the benchmark for assessing treatment effects but are limited to phenotypically narrow populations by design. We introduce a novel generative artificial intelligence (AI) driven emulation method that infers effect size through virtual clinical trials, which can emulate the RCT process and potentially extrapolate into wider populations. We validate the virtual trials by comparing the predicted impact of glucagon-like peptide-1 (GLP-1) agonists on HbA1c in type-2 diabetes (T2DM) with its true efficacy established in the LEAD-5 trial. Our emulation model learns treatment effects from real-world evidence data by a combined generative AI and causal learning approach. Training data comprised pre- and post-treatment outcomes for 5,476 people with T2DM. We considered three treatment arms: GLP-1 (Liraglutide), basal insulin (glargine), and placebo. After training, virtual trials were conducted by sampling 232 virtual patients per arm (according to the LEAD-5 inclusion criteria) and predicting post-treatment outcomes. We used difference-in-differences (DiD) for pairwise comparisons between arms. Our goal was to emulate LEAD-5 by demonstrating a significant DiD in post-treatment HbA1c reduction for GLP-1 compared to basal insulin and placebo. We found significant differences in HbA1c reduction for GLP-1 vs basal insulin (-1.21 mmol/mol (-0.11%); p < 0.001) and GLP-1 vs placebo (-2.58 mmol/mol (-0.24%); p < 0.001) in our virtual populations, consistent with LEAD-5 (Liraglutide vs glargine: -2.62mmol/mol (-0.24%); p = 0.0015, Liraglutide vs placebo: -11.91 mmol/mol (-1.09%); p < 0.0001). The causal AI-powered clinical trials can emulate LEAD-5 in important measurements for T2DM. Our algorithm is specialty agnostic and can explore counterfactual questions, making it suitable for further study in the generalizability of RCT results in real-world populations to support clinical decision-making and policy recommendations.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000927"},"PeriodicalIF":0.0,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12279107/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144683756","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":"A mapping survey of digital clinical consultations in maternity care in England.","authors":"Georgia Clancy, Kerry Evans, Helen Spiby, Victoria Barrett, Candice Sunney, Catrin Evans","doi":"10.1371/journal.pdig.0000944","DOIUrl":"10.1371/journal.pdig.0000944","url":null,"abstract":"<p><p>Many areas of healthcare are exploring the use of digital technologies with the aim of improving and expanding care to service users. In England, maternity care is currently undergoing a digital transformation in line with the National Health Service's (NHS) Long Term Plan, which seeks to enhance care delivery and accessibility through digital tools. However, there is a lack of data on the current use and practice of digital consultations across the country. This project aimed to map and explore how digital clinical consultations are currently being used by NHS maternity care services in England. An online survey was designed to capture data on current practice, guidance and procedures to address potential inequalities. The survey was distributed to each NHS Trust (n = 121) that provides maternity care (specifically to senior maternity care professionals and digital maternity leaders who could provide an overview of how digital consultations were being used where they worked). The survey was open between January and March 2024. 53 completed surveys were received representing 39 different organisations (32% of those currently providing maternity care in England). Quantitative summary statistics indicated that telephone consultations were the most commonly used digital modality across all stages of the maternity care pathway. Thematic analysis identified barriers such as a lack of staff consultation and lack of staff training on the use of digital consultations. It was uncommon for women to be asked about their consultation preferences or assessed for individual needs. In conclusion, the findings reveal significant variation in the use of digital consultations, highlighting a gap between policy intentions and practice. Key areas for improvement in the delivery and implementation of digital consultations include staff training, systems to record women's consultation preferences/needs, and more research to support digital inclusion. The findings of this survey have the potential to have applications beyond maternity care and in different geographical contexts.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000944"},"PeriodicalIF":0.0,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12279142/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144683754","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":"An artificial intelligence-based approach to identify volume status in patients with severe dengue using wearable PPG data.","authors":"Ngan Nguyen Lyle, Ho Quang Chanh, Hao Nguyen Van, James Anibal, Stefan Karolcik, Damien Ming, Giang Nguyen Thi, Huyen Vu Ngo Thanh, Huy Nguyen Quang, Hai Ho Bich, Khoa Le Dinh Van, Van Hoang Minh Tu, Khanh Phan Nguyen Quoc, Huynh Trung Trieu, Qui Tu Phan, Tho Phan Vinh, Tai Luong Thi Hue, Pantelis Georgiou, Louise Thwaites, Sophie Yacoub","doi":"10.1371/journal.pdig.0000924","DOIUrl":"10.1371/journal.pdig.0000924","url":null,"abstract":"<p><p>Dengue shock syndrome (DSS) is a serious complication of dengue infection which occurs when critical plasma leakage results in haemodynamic shock. Treatment is challenging as fluid therapy must balance the risk of hypoperfusion with volume overload. In this study, we investigate the potential utility of wearable photoplethysmography (PPG) to determine volume status in DSS. In this prospective observational study, we enrolled 250 adults and children with a clinical diagnosis of dengue admitted to the Hospital for Tropical Diseases, Ho Chi Minh City. PPG monitoring using a wearable device was applied for a 24-hour period. Clinical events were then matched to the PPG data by date and time. We predefined two clinical states for comparison: (1) the 2-hour period before a shock event was an \"empty\" volume state and (2) the 2-hour period between 1 and 3 hours after a fluid initiation event was a \"full\" volume state. PPG data were sampled from these states for analysis. Variability and waveform morphology features were extracted and analyzed using principal components analysis and random forest. Waveform images were used to develop a computer vision model. Of the 250 patients enrolled, 90 patients experienced the predefined outcomes, and had sufficient data for the analysis. Principal components analysis identified four principal components (PCs), from the 23 pulse wave features. Logistic regression using these PCs showed that the empty state is associated with PCs 1 (p = 0.016) and 4 (p = 0.036) with both PCs denoting increased sympathetic activity. Random forest showed that heart rate and the LF-HF ratio are the most important features. A computer vision model had a sensitivity of 0.81 and a specificity of 0.70 for the empty state. These results provide proof of concept that an artificial intelligence-based approach using continuous PPG monitoring can provide information on volume states in DSS.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000924"},"PeriodicalIF":0.0,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12273927/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144664054","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":"The ActiveText@T2D text messaging behavioural intervention to increase physical activity in adults with type 2 diabetes: A prospective single-arm feasibility trial.","authors":"Holly Blake, Mohammed Jameen Alsahli, Wendy J Chaplin, Stathis Th Konstantinidis","doi":"10.1371/journal.pdig.0000953","DOIUrl":"10.1371/journal.pdig.0000953","url":null,"abstract":"<p><p>Physical activity is a core aspect of type 2 diabetes (T2DM) self-management, but most Saudi adults do not meet physical activity recommendations and there are no culturally tailored interventions to promote physical activity in Saudi adults with T2DM. This study is a prospective single-centre, single-arm feasibility study of a mobile SMS text messaging intervention with a nested qualitative study. The aim was to explore the feasibility and acceptability of ActiveText@T2D, a 6-week theory-based mobile text messaging intervention to promote physical activity in people with T2DM in Saudi Arabia. Intervention development was informed by the Behaviour Change Wheel (BCW) framework and COM-B model. ActiveText@T2D consisted of 2 one-way SMS text messages per week, for 6 weeks. All participants were offered the intervention and assessed at baseline (Time 0: T0) and 3-month follow-up (Time 1: T1). Data collection included feasibility outcomes (recruitment and retention), clinical outcomes (body mass index and glycaemic control from clinic records at T0), and self-reported outcomes (self-efficacy, physical activity, and barriers to exercise at T0, T1). Qualitative interview data (n = 19) were collected at T1 with 11 patients (7 male, 4 female, mean age 54.5 years) and 8 female nurses (mean age 31.8 years). Quantitative data were analysed descriptively, qualitative data were analysed thematically. Of 98 participants approached, 62 were eligible, and 52 consented (84% participation rate; 23 women, 29 men; mean age 54.82 years), 44 (85%) completed baseline measures and received the intervention. Thirty-nine participants completed follow-up measures (75% retention to T1). All outcome measures were sensitive to change: The Arabic version of the CDC Barriers to Being Active Quiz (BBAQ), The Arabic version of Exercise Self-Efficacy scale (ESE-A), The Arabic International Physical Activity Questionnaire (A-IPAQ). Patients and healthcare professionals perceived the intervention to be broadly acceptable. Qualitative findings identified three overarching themes: \"use of text messaging as a health intervention\", \"engagement with physical activity\" and \"instilling knowledge about physical activity and diabetes control\". This study demonstrates the feasibility and acceptability of ActiveText@T2D, a theory-based culturally tailored SMS text messaging intervention, to Saudi adult patients with T2DM and healthcare professionals involved in their care. The next step would be a full-scale definitive randomised controlled trial to assess the effectiveness and cost-effectiveness of ActiveText@T2D. Protocol registration: Protocols.io, DOI: dx.doi.org/10.17504/protocols.io.261ger217l47/v1 (registered on 08.01.2025).</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000953"},"PeriodicalIF":0.0,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12273986/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144664055","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":"Urban environmental and population factors as determinants of COVID-19 severity: A spatially-resolved probabilistic modeling approach.","authors":"Jacob Roxon, Marie-Sophie Dumont, Eric Vilain, Mircea T Sofonea, Roland J-M Pellenq","doi":"10.1371/journal.pdig.0000921","DOIUrl":"10.1371/journal.pdig.0000921","url":null,"abstract":"<p><p>COVID-19 is caused by a severe acute respiratory syndrome due to the SARS-CoV-2 coronavirus. It has reshaped the world with the way our communities interact, people work, commute, and spend their leisure time. While different mitigation solutions for controlling COVID-19 virus transmission have already been established, global models that would explain and predict the impact of urban environments on the case fatality ratio CFR of COVID-19 (defined as the number of deaths divided by the number of cases over a time window) are missing. Here, with readily available data from public sources, we study the CFR of the coronavirus for 118 locations (city zip-codes, city boroughs, and cities) worldwide to identify the links between the CFR and outdoor, indoor and personal urban factors. We show that a probabilistic model, optimized on the sample of 20 districts from 4 major US cities, provides an accurate predictive tool for the CFR of COVID-19 regardless of the geographical location. Furthermore, we show that the validity of the model extends to other infectious diseases such as flu and pneumonia with pre-COVID-19 pandemic data for 3 US cities indicating that the first COVID-19 wave severity corresponds to that of pneumonia while other COVID-19 waves have the severity of influenza.When adjusted for the population, our model can be used to evaluate risk and severity of the disease within different parts of the city for different waves of the pandemic. Our results suggest that although disease screening and vaccination policies to containment and lockdowns remain critical in controlling the spread of airborne diseases, urban factors such as population density, humidity, or order of buildings, should all be taken into consideration when identifying resources and planning targeted responses to mitigate the impact and severity of the viruses transmitted through air.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000921"},"PeriodicalIF":0.0,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12274012/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144664056","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":"The role of artificial intelligence in maternal and child health: Progress, controversies, and future directions.","authors":"Audêncio Victor","doi":"10.1371/journal.pdig.0000938","DOIUrl":"10.1371/journal.pdig.0000938","url":null,"abstract":"<p><p>This debate paper examines the transformative potential of Artificial Intelligence (AI), specifically through Machine Learning (ML), in enhancing preventive measures in maternal and child health (MCH). With the proliferation of Big Data, ML has become crucial in handling complex, non-linear interactions among health determinants to not only predict but also prevent adverse outcomes. This paper underscores AI's applications in early interventions that could decrease the incidence of MCH issues. It reviews technological advancements while addressing ethical, practical, and data-related challenges in applying AI in preventive healthcare. Emphasis is placed on recent supervised, unsupervised, and reinforcement learning applications that significantly advance preventive care, particularly in low-resource settings. The manuscript discusses the development of AI models for early diagnosis, comprehensive risk assessments, and customized preventive interventions, while highlighting challenges like data diversity, privacy issues, and integrating multimodal health data.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"4 7","pages":"e0000938"},"PeriodicalIF":0.0,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12270093/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144661231","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}