Lancet Digital Health最新文献

{"title":"Development and external validation of a clinical prediction model for new-onset atrial fibrillation in intensive care: a multicentre, retrospective cohort study","authors":"Jonathan P Bedford DPhil ,&nbsp;Oliver Redfern PhD ,&nbsp;Stephen Gerry DPhil ,&nbsp;Robert Hatch BA ,&nbsp;Prof Liza Keating MSc ,&nbsp;Prof David Clifton DPhil ,&nbsp;Prof Gary S Collins PhD ,&nbsp;Prof Peter J Watkinson MD","doi":"10.1016/j.landig.2025.100896","DOIUrl":"10.1016/j.landig.2025.100896","url":null,"abstract":"<div><h3>Background</h3><div>New-onset atrial fibrillation, a condition associated with adverse outcomes in the short and long term, is common in patients admitted to intensive care units (ICUs). Identifying patients at high risk could inform trials of preventive interventions and help to target such interventions. We aimed to develop and externally validate a prediction model for new-onset atrial fibrillation in patients admitted to ICUs.</div></div><div><h3>Methods</h3><div>We conducted a multicentre, retrospective cohort study in three ICUs across the UK and four ICUs across the USA. Patients aged 16 years and older admitted to an ICU for more than 3 h without a history or presentation of clinically significant arrhythmia were eligible for inclusion. We analysed clinical variables to investigate the associations between predetermined candidate variables and risk of new-onset atrial fibrillation and to develop a model to estimate these risks. We developed the METRIC-AF model, a machine learning model incorporating dynamic variables. Model performance was assessed through internal–external cross-validation during model development and externally validated by use of multicentre data from ICUs across the UK. We then developed a simple graphical prediction tool using three important predictors.</div></div><div><h3>Findings</h3><div>Among 39 084 eligible patients admitted to an ICU between 2008 and 2019, 2797 (7·2%) developed new-onset atrial fibrillation during the first 7 days of their ICU stay. We identified multiple non-linear associations between candidate variables and risk of new-onset atrial fibrillation, including hypomagnesaemia at serum concentrations below 0·70 mmol/L. The final METRIC-AF model contained ten routinely collected clinical variables. Compared with a published logistic regression model, the METRIC-AF model showed superior calibration, net benefit across clinically relevant risk thresholds, and discriminative performance (C statistic 0·812 [95% CI 0·805–0·822] <em>vs</em> 0·786 [0·778–0·801]; p=0·0003). The simple graphical tool performed well in attributing the risk of new-onset atrial fibrillation in the external validation dataset (C statistic 0·727 [95% CI 0·716–0·739]).</div></div><div><h3>Interpretation</h3><div>The METRIC-AF model and its companion graphical tool could support the identification of patients at increased risk of developing new-onset atrial fibrillation during ICU admission, informing targeted prophylactic strategies and trial enrichment by use of routinely available clinical data. An online app also developed as part of the study allows for the exploration of prediction generation among individuals and external validation in prospective studies.</div></div><div><h3>Funding</h3><div>National Institute for Health and Care Research (NIHR) and NIHR Oxford Biomedical Research Centre.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100896"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145034409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital adherence technology interventions to reduce poor end-of-treatment outcomes and recurrence among adults with drug-sensitive tuberculosis in Ethiopia: a three-arm, pragmatic, cluster-randomised, controlled trial","authors":"Amare W Tadesse PhD ,&nbsp;Mamush Sahile MPH ,&nbsp;Nicola Foster PhD ,&nbsp;Christopher Finn McQuaid PhD ,&nbsp;Gedion Teferra Weldemichael MD ,&nbsp;Tofik Abdurhman MSc ,&nbsp;Zemedu Mohammed MPH ,&nbsp;Mahilet Belachew MSc ,&nbsp;Amanuel Shiferaw MPH ,&nbsp;Demelash Assefa MPH ,&nbsp;Demekech Gadissa MPH ,&nbsp;Hiwot Yazew MPH ,&nbsp;Nuria Yakob MPH ,&nbsp;Zewdneh Shewamene PhD ,&nbsp;Lara Goscé PhD ,&nbsp;Job van Rest MSc ,&nbsp;Norma Madden MSc ,&nbsp;Prof Salome Charalambous PhD ,&nbsp;Kristian van Kalmthout MSc ,&nbsp;Ahmed Bedru MD ,&nbsp;Prof Katherine L Fielding PhD","doi":"10.1016/j.landig.2025.100895","DOIUrl":"10.1016/j.landig.2025.100895","url":null,"abstract":"<div><h3>Background</h3><div>The effect of digital adherence technologies (DATs) on long-term tuberculosis treatment outcomes remains unclear. We aimed to assess the effectiveness of DATs in improving tuberculosis treatment outcomes and recurrence.</div></div><div><h3>Methods</h3><div>We did a pragmatic cluster-randomised trial in Ethiopia. 78 health facilities (clusters) were randomised (1:1:1) to smart pillbox, medication labels, or standard of care. Adults aged 18 years or older with drug-sensitive pulmonary tuberculosis on a fixed-dose combination tuberculosis treatment regimen were enrolled and followed up for 12 months after treatment initiation. Those in the smart pillbox group received a pillbox with customisable audio-visual reminders, whereas participants in the medications label group received their tuberculosis medication with a weekly unique code label. Opening the box or texting the code prompted real-time dose logging on the adherence platform, facilitating differentiated response to an individual’s adherence by a health-care worker. The primary composite outcome comprised death, loss to follow-up, treatment failure, switch to drug-resistant tuberculosis treatment, or recurrence. Secondary outcomes were poor end-of-treatment outcome and loss to follow-up. Analysis accounted for clustered design with multiple imputation for the primary composite outcome. The trial is registered with Pan African Clinical Trials Registry (PACTR202008776694999) and is complete.</div></div><div><h3>Findings</h3><div>From May 24, 2021, to Aug 8, 2022, 8477 individuals undergoing tuberculosis treatment were assessed for eligibility. Of the 3885 participants enrolled, 3858 were included in the intention-to-treat population. 1567 (40·6%) of 3858 participants were women and the median age of all participants was 30 years (IQR 24–40). At 12 months, using multiple imputation, neither the smart pillbox group (adjusted odds ratio [OR] 1·04 [95% CI 0·74 to 1·45]; adjusted risk difference: 0·96 percentage points [95% CI –1·19 to 3·11]) nor the medication labels group (adjusted OR 1·14 [0·83 to 1·61]; adjusted risk difference: 0·42 percentage points [–1·75 to 2·59]) reduced the risk of the primary composite outcome. There was no evidence of effect on poor end-of-treatment outcomes or loss to follow-up in either intervention group, although the label intervention showed weak evidence of reduced loss to follow-up. Results were similar in complete case and per-protocol analyses.</div></div><div><h3>Interpretation</h3><div>The DAT interventions showed no reduction in unfavourable outcomes. This emphasises the necessity to optimise DATs to enhance tuberculosis management strategies and treatment outcomes.</div></div><div><h3>Funding</h3><div>Unitaid.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100895"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145201808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Navigating the landscape of medical artificial intelligence reporting guidelines","authors":"The Lancet Digital Health","doi":"10.1016/j.landig.2025.100925","DOIUrl":"10.1016/j.landig.2025.100925","url":null,"abstract":"","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100925"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145276446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the potential of generative artificial intelligence in medical image synthesis: opportunities, challenges, and future directions","authors":"Bardia Khosravi MD MPH ,&nbsp;Saptarshi Purkayastha PhD ,&nbsp;Prof Bradley J Erickson MD PhD ,&nbsp;Hari M Trivedi MD ,&nbsp;Judy W Gichoya MD MS","doi":"10.1016/j.landig.2025.100890","DOIUrl":"10.1016/j.landig.2025.100890","url":null,"abstract":"<div><div>Generative artificial intelligence has emerged as a transformative force in medical imaging since 2022, enabling the creation of derivative synthetic datasets that closely resemble real-world data. This Viewpoint examines key aspects of synthetic data, focusing on its advancements, applications, and challenges in medical imaging. Various generative artificial intelligence image generation paradigms, such as physics-informed and statistical models, and their potential to augment and diversify medical research resources are explored. The promises of synthetic datasets, including increased diversity, privacy preservation, and multifunctionality, are also discussed, along with their ability to model complex biological phenomena. Next, specific applications using synthetic data such as enhancing medical education, augmenting rare disease datasets, improving radiology workflows, and enabling privacy-preserving multicentre collaborations are highlighted. The challenges and ethical considerations surrounding generative artificial intelligence, including patient privacy, data copying, and potential biases that could impede clinical translation, are also addressed. Finally, future directions for research and development in this rapidly evolving field are outlined, emphasising the need for robust evaluation frameworks and responsible utilisation of generative artificial intelligence in medical imaging.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100890"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144859803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Causal deep learning for real-time detection of cardiac surgery-associated acute kidney injury: derivation and validation in seven time-series cohorts","authors":"Qin Zhong PhD ,&nbsp;Yuxiao Cheng BEng ,&nbsp;Zongren Li PhD ,&nbsp;Prof Dongjin Wang MD ,&nbsp;Chongyou Rao PhD ,&nbsp;Yi Jiang MD ,&nbsp;Lianglong Li BEng ,&nbsp;Ziqian Wang BEng ,&nbsp;Pan Liu PhD ,&nbsp;Hebin Che MSc ,&nbsp;Pei Li PhD ,&nbsp;Prof Xin Lu PhD ,&nbsp;Jinli Suo PhD ,&nbsp;Kunlun He PhD","doi":"10.1016/j.landig.2025.100901","DOIUrl":"10.1016/j.landig.2025.100901","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Background&lt;/h3&gt;&lt;div&gt;Cardiac surgery-associated acute kidney injury (CSA-AKI) is a complex complication substantially contributing to an increased risk of mortality. Effective CSA-AKI management relies on timely diagnosis and interventions. However, many cases are detected too late. Despite the advancements in novel biomarkers and data-driven predictive models, existing practices are primarily constrained due to the limited discriminative and generalisation capabilities and stringent application requirements, presenting major challenges to the timely and effective diagnosis and interventions in CSA-AKI management. This study aimed to develop a causal deep learning architecture, named REACT, to achieve precise and dynamic predictions of CSA-AKI within the subsequent 48 h.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods&lt;/h3&gt;&lt;div&gt;In this retrospective model development and prospective validation study, we included adult patients (aged ≥18 years) from seven distinct cohorts undergoing major open-heart surgery for model training and validation. Data for model development and internal validation were sourced from electronic health records of two large centres in Beijing, China, between Jan 1, 2000, and Dec 31, 2022. External validation was conducted on three independent centres in China between Jan 1, 2000, and Dec 31, 2022&lt;strong&gt;,&lt;/strong&gt; along with cross-national data from the public databases MIMIC-IV and eICU in the USA. To facilitate implementation, we also developed a publicly accessible web calculator and applet. The model’s prospective application was validated from June 1, to Oct 31, 2023, at two centres in Beijing and Nanjing, China.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Findings&lt;/h3&gt;&lt;div&gt;The final derivation cohort included 14 513 eligible patients with a median age of 56 years (IQR 45–65), 5515 (38·0%) patients were female, and 3047 (21·0%) developed CSA-AKI. The external validation dataset included 20 813 patients from China and 28 023 from the USA. REACT reduced 1328 input variables to six essential causal factors for CSA-AKI prediction. In internal validation, REACT achieved an average area under the receiver operating characteristic curve (AUROC) of 0·930 (SD 0·032), outperforming state-of-the-art deep learning architectures, specifically transformer-based and long short-term memory-based models, which rely on more complex variables. The model consistently outperformed in external validation across different centres (average AUROC 0·920 [SD 0·036]) and regions (0·867 [0·073]), as well as in prospective validation (0·896 [0·023]). Compared with guideline-recommended pathways, REACT detected CSA-AKI on average 16·35 h (SD 2·01) earlier in external validation.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Interpretation&lt;/h3&gt;&lt;div&gt;We proposed a causal deep learning approach to predict CSA-AKI risk within 48 h, distilling the complex temporal interactions between variables into only a few universal, relatively cost-effective inputs. The approach shows great potential for deployment across hospit","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100901"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145151393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Value of artificial intelligence in neuro-oncology","authors":"Sebastian Voigtlaender MSc ,&nbsp;Thomas A Nelson MD ,&nbsp;Philipp Karschnia MD ,&nbsp;Eugene J Vaios MD ,&nbsp;Prof Michelle M Kim MD ,&nbsp;Philipp Lohmann PhD ,&nbsp;Prof Norbert Galldiks MD ,&nbsp;Prof Mariella G Filbin MD ,&nbsp;Shekoofeh Azizi PhD ,&nbsp;Vivek Natarajan MSc ,&nbsp;Prof Michelle Monje MD ,&nbsp;Prof Jorg Dietrich MD ,&nbsp;Sebastian F Winter MD","doi":"10.1016/j.landig.2025.100876","DOIUrl":"10.1016/j.landig.2025.100876","url":null,"abstract":"<div><div>CNS cancers are complex, difficult-to-treat malignancies that remain insufficiently understood and mostly incurable, despite decades of research efforts. Artificial intelligence (AI) is poised to reshape neuro-oncological practice and research, driving advances in medical image analysis, neuro–molecular–genetic characterisation, biomarker discovery, therapeutic target identification, tailored management strategies, and neurorehabilitation. This Review examines key opportunities and challenges associated with AI applications along the neuro-oncological care trajectory. We highlight emerging trends in foundation models, biophysical modelling, synthetic data, and drug development and discuss regulatory, operational, and ethical hurdles across data, translation, and implementation gaps. Near-term clinical translation depends on scaling validated AI solutions for well defined clinical tasks. In contrast, more experimental AI solutions offer broader potential but require technical refinement and resolution of data and regulatory challenges. Addressing both general and neuro-oncology-specific issues is essential to unlock the full potential of AI and ensure its responsible, effective, and needs-based integration into neuro-oncological practice.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100876"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144812596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How CHART (Chatbot Assessment Reporting Tool) can help to advance clinical artificial intelligence research through clearer task definition and robust validation","authors":"Arun James Thirunavukarasu ,&nbsp;Ernest Lim ,&nbsp;Bright Huo","doi":"10.1016/j.landig.2025.100910","DOIUrl":"10.1016/j.landig.2025.100910","url":null,"abstract":"","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 9","pages":"Article 100910"},"PeriodicalIF":24.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterising non-household contact patterns relevant to respiratory transmission in the USA: analysis of a cross-sectional survey","authors":"Juliana C Taube AB ,&nbsp;Zachary Susswein BS ,&nbsp;Vittoria Colizza PhD ,&nbsp;Prof Shweta Bansal PhD","doi":"10.1016/j.landig.2025.100888","DOIUrl":"10.1016/j.landig.2025.100888","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Background&lt;/h3&gt;&lt;div&gt;Interpersonal contact has a crucial role in the transmission of infectious diseases. Characterising heterogeneity in contact patterns across individuals, time, and space is necessary to inform accurate estimates of transmission risk, particularly to explain superspreading, predict differences in vulnerability by age, and inform physical distancing policies. Current respiratory disease models often rely on data from the 2008 POLYMOD study conducted in Europe, which is now outdated and is potentially unrepresentative of behaviour in other geographical regions. We aimed to understand the variation in contact patterns in the USA across time, spatial scales, and demographic and social classifications during the COVID-19 pandemic, and to estimate what social behaviour looks like at baseline, in the absence of an ongoing pandemic.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods&lt;/h3&gt;&lt;div&gt;For this study of contact patterns relevant to respiratory transmission during a pandemic, we examined 10·7 million responses to the US COVID-19 Trends and Impact Survey between June 1, 2020, and April 30, 2021 (ie, during the COVID-19 pandemic); the survey recruited participants aged 18 years and older in the USA through Facebook. Data were post-stratified by age and gender to correct for sample representation. We used generalised additive models to characterise spatiotemporal heterogeneity in respiratory contact patterns during the pandemic at the county-week scale; we established how contact patterns vary by urbanicity, age (18–54 years, 55–64 years, 65–74 years, or ≥75 years), gender (male or female), race or ethnicity (Asian, Black or African American, Hispanic, White, or other), and contact setting (work, shopping for essentials, social gatherings, or other). We used a regression approach to estimate baseline (non-pandemic) contact patterns.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Findings&lt;/h3&gt;&lt;div&gt;Although contact patterns varied over time during the COVID-19 pandemic, the average number of daily contacts was relatively stable after controlling for the effect of incidence-mediated risk perception and disease-related policy. The mean number of non-household contacts was spatially heterogeneous, varying across urban versus rural settings, regardless of the presence of disease. Additional heterogeneity was observed across age, gender, race or ethnicity, and contact setting. Mean number of contacts decreased with age for individuals older than 55 years and was lower in women than in men. During periods of increased national incidence of disease, the contacts of White individuals and contacts at work or social gatherings showed the greatest change.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Interpretation&lt;/h3&gt;&lt;div&gt;Our findings indicate that US adult baseline contact patterns show little variability over time after controlling for disease, but high spatial variability regardless of disease, with implications for understanding the seasonality of respiratory infectious diseases. The highly structured spat","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 8","pages":"Article 100888"},"PeriodicalIF":24.1,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital health interventions for mental health disorders: an umbrella review of meta-analyses of randomised controlled trials","authors":"Cristina Crocamo PhD ,&nbsp;Dario Palpella MD ,&nbsp;Daniele Cavaleri MD ,&nbsp;Christian Nasti MD ,&nbsp;Susanna Piacenti MD ,&nbsp;Pietro Morello MD ,&nbsp;Giada Lauria MD ,&nbsp;Oliviero Villa MD ,&nbsp;Ilaria Riboldi PhD ,&nbsp;Francesco Bartoli PhD ,&nbsp;John Torous MD ,&nbsp;Prof Giuseppe Carrà PhD","doi":"10.1016/j.landig.2025.100878","DOIUrl":"10.1016/j.landig.2025.100878","url":null,"abstract":"<div><div>Digital health interventions (DHIs) show promise for the treatment of mental health disorders. However, existing meta-analytical research is methodologically heterogeneous, with studies including a mix of clinical, non-clinical, and transdiagnostic populations, hindering a comprehensive understanding of DHI effectiveness. Thus, we conducted an umbrella review of meta-analyses of randomised controlled trials investigating the effectiveness of DHIs for specific mental health disorders and evaluating the quality of evidence. We searched three public electronic databases from inception to February, 2024 and included 16 studies. DHIs were effective compared with active interventions for schizophrenia spectrum disorders, major depressive disorder, social anxiety disorder, and panic disorder. Notable treatment effects compared with a waiting list were also observed for specific phobias, generalised anxiety disorder, obsessive-compulsive disorder, post-traumatic stress disorder, and bulimia nervosa. Certainty of evidence was rated as very low or low in most cases, except for generalised anxiety disorder-related outcomes, which showed a moderate rating. To integrate DHIs into clinical practice, further high-quality studies with clearly defined target populations and robust comparators are needed.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 8","pages":"Article 100878"},"PeriodicalIF":24.1,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144561613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Re-engineering a machine learning phenotype to adapt to the changing COVID-19 landscape: a machine learning modelling study from the N3C and RECOVER consortia","authors":"Miles Crosskey PhD ,&nbsp;Tomas McIntee PhD ,&nbsp;Sandy Preiss MS ,&nbsp;Daniel Brannock MS ,&nbsp;John M Baratta MD ,&nbsp;Yun Jae Yoo BS ,&nbsp;Emily Hadley MS ,&nbsp;Frank Blanceró BA ,&nbsp;Robert Chew MS ,&nbsp;Johanna Loomba MS ,&nbsp;Abhishek Bhatia MS ,&nbsp;Prof Christopher G Chute MD ,&nbsp;Prof Melissa Haendel PhD ,&nbsp;Richard Moffitt PhD ,&nbsp;Emily R Pfaff PhD ,&nbsp;N3C Consortium and the RECOVER EHR cohort","doi":"10.1016/j.landig.2025.100887","DOIUrl":"10.1016/j.landig.2025.100887","url":null,"abstract":"<div><h3>Background</h3><div>In 2021, we used the National COVID Cohort Collaborative (N3C) as part of the National Institutes of Health RECOVER Initiative to develop a machine learning pipeline to identify patients with a high probability of having post-acute sequelae of SARS-CoV-2 infection or long COVID. However, the increased home testing, missing documentation, and reinfections that characterise the pandemic beyond 2022 necessitated the re-engineering of our original model to account for these changes in the COVID-19 research landscape.</div></div><div><h3>Methods</h3><div>Trained on 72 745 patient records (36 238 with long COVID and 36 507 with no evidence of long COVID), our updated XGBoost model gathered data for each patient in overlapping 100-day periods that progressed through time and issued a probability of long COVID for each 100-day period. We ran the model on patients in N3C (n=5 875 065) who met at least one of the following criteria from Jan 1, 2020, to June 22, 2023: a U07·1 (COVID-19) diagnosis code; a positive SARS-CoV-2 test; a U09·9 (post-acute sequelae of SARS-CoV-2 infection) diagnosis code; a prescription for nirmatrelvir–ritonavir or remdesivir; or an M35·81 (multisystem inflammatory syndrome in children [MIS-C]) diagnosis code. Each patient was given a model score that predicted long COVID status for each 100-day window in which they were aged ≥18 years. If a patient had known acute COVID-19 during any 100-day window (including reinfections), we censored the data from 7 days before the diagnosis or positive test date to 28 days after. We ran the model on controls selected from pre-2020 data to assess the likelihood of false positives.</div></div><div><h3>Findings</h3><div>The updated model had an area under the receiver operating characteristic curve of 0·90. Precision and recall could be adjusted according to a given use case, depending on whether greater sensitivity or specificity was warranted. Using our model, we estimate the overall prevalence of long COVID among the COVID-19 positive cohort within N3C repository to be 10.4%.</div></div><div><h3>Interpretation</h3><div>By eschewing the COVID-19 index date as an anchor point for analysis, we can assess the probability of long COVID among patients who might have tested at home, or with suspected (but untested) cases of COVID-19, or multiple SARS-CoV-2 reinfections. We view this exercise as a model for maintaining and updating any machine learning pipeline used for clinical research and operations.</div></div><div><h3>Funding</h3><div>National Institutes of Health RECOVER Initiative.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 8","pages":"Article 100887"},"PeriodicalIF":24.1,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974537","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}