Background:Recently,several cutting-edge experimental studies have directed chimeric antigen receptor(CAR)-T therapies toward specific renal diseases,revealing substantial renal benefits.Prior to widespread implementa...Background:Recently,several cutting-edge experimental studies have directed chimeric antigen receptor(CAR)-T therapies toward specific renal diseases,revealing substantial renal benefits.Prior to widespread implementation of these animal experiments and potentially clinical trials,it is crucial to assess the renal safety of CAR-T therapies using real-world safety evidence.Methods:Our focus was on utilizing 4 algorithms,including disproportionality analysis,based on the US Food and Drug Administration Adverse Event Reporting System database,to filter positive signals of acute and chronic renal injury associated with 6 CAR-T therapies.Further determination of causality was achieved through Mendelian randomization(MR)for drugs associated with renal injury events showing a correlation.Results:Six therapies were evaluated involving a total of 9,770 patients,with only acute kidney injury(AKI)identified as associated with idecabtagene vicleucel treatment using 4 algorithmic thresholds,including disproportionality analysis.Subsequently,MR revealed no causal relationship between the idecabtagene vicleucel target B cell maturation antigen and the risk of AKI(P=0.576),a finding validated in another independent dataset(P=0.734).Conclusion:CAR-T therapies do not directly cause renal damage and necessitate controlling adverse renal risks during or after treatment,such as cytokine release syndrome.Future research efforts should rigorously optimize these aspects to better cater to nephrologists’requirements.展开更多
Background:Clinical and biomedical research in low-resource settings often faces substantial challenges due to the need for high-quality data with sufficient sample sizes to construct effective models.These constraint...Background:Clinical and biomedical research in low-resource settings often faces substantial challenges due to the need for high-quality data with sufficient sample sizes to construct effective models.These constraints hinder robust model training and prompt researchers to seek methods for leveraging existing knowledge from related studies to support new research efforts.Transfer learning(TL),a machine learning technique,emerges as a powerful solution by utilizing knowledge from pretrained models to enhance the performance of new models,offering promise across various healthcare domains.Despite its conceptual origins in the 1990s,the application of TL in medical research has remained limited,especially beyond image analysis.This review aims to analyze TL applications,highlight overlooked techniques,and suggest improvements for future healthcare research.Methods:Following the PRISMA-ScR guidelines,we conducted a search for published articles that employed TL with structured clinical or biomedical data by searching the SCOPUS,MEDLINE,Web of Science,Embase,and CINAHL databases.Results:We screened 5,080 papers,with 86 meeting the inclusion criteria.Among these,only 2%(2 of 86)utilized external studies,and 5%(4 of 86)addressed scenarios involving multi-site collaborations with privacy constraints.Conclusions:To achieve actionable TL with structured medical data while addressing regional disparities,inequality,and privacy constraints in healthcare research,we advocate for the careful identification of appropriate source data and models,the selection of suitable TL frameworks,and the validation of TL models with proper baselines.展开更多
Background:There are few data on the comorbidity of diabetes in Chinese patients with depression.We aimed to calculate the prevalence and explore risk factors of type 2 diabetes mellitus(T2DM)among depression inpatien...Background:There are few data on the comorbidity of diabetes in Chinese patients with depression.We aimed to calculate the prevalence and explore risk factors of type 2 diabetes mellitus(T2DM)among depression inpatients from 2005 to 2018 in Beijing.Methods:This study is a cross-sectional study.The data collected from 19 specialized psychiatric hospitals in Beijing were analyzed.The prevalence of T2DM and its distribution were analyzed.The multivariable logistic regression was performed to explore the risk factors of T2DM.Results:A total of 20,899 depression inpatients were included.The prevalence of T2DM was 9.13%[95%confidence interval(CI),8.74%to 9.52%].The prevalence of T2DM showed an upward trend with year(P for trend<0.001)and age(P for trend<0.001).The prevalence of T2DM was higher among readmitted patients(12.97%)and patients with comorbid hypertension(26.16%),hyperlipidemia(21.28%),and nonalcoholic fatty liver disease(NAFLD)(18.85%).The prevalence of T2DM in females was lower than in males among patients aged 18 to 59 years,while the prevalence of T2DM in females was higher than in males among patients aged≥60 years.T2DM was associated with older age[adjusted odds ratios(aORs)ranged from 3.68 to 29.95,P<0.001],hypertension(aOR,3.01;95%CI,2.70 to 3.35;P<0.001),hyperlipidemia(aOR,1.69;95%CI,1.50 to 1.91;P<0.001),and NAFLD(aOR,1.58;95%CI,1.37 to 1.82;P<0.001).Conclusions:The prevalence of T2DM among depression inpatients from 2005 to 2018 in Beijing was high and increased with the year.Depression inpatients who were older and with hypertension,hyperlipidemia,and NAFLD had a higher prevalence and risk of T2DM.展开更多
Background:Major depressive disorder(MDD)and autism spectrum disorder(ASD)are complex and heterogeneous neuropsychiatric disorders with overlapping symptoms,presenting remarkable challenges for accurate diagnosis.Leve...Background:Major depressive disorder(MDD)and autism spectrum disorder(ASD)are complex and heterogeneous neuropsychiatric disorders with overlapping symptoms,presenting remarkable challenges for accurate diagnosis.Leveraging functional neuroimaging data offers an opportunity to develop more robust,data-driven approach for psychiatric disorder detection.However,existing methods often struggle to capture the long-term dependencies and dynamic patterns inherent in such data,particularly across diverse imaging sites.Methods:We propose Multiscale Contextual Mamba(MSC-Mamba),a Mamba-based model designed for capturing long-term dependencies in multivariate time-series data while maintaining linear scalability,allowing us to account for long-range interactions and subtle dynamic patterns within the brain’s functional networks.One of the main advantages of MSC-Mamba is its ability to leverage the distinct characteristics of time-series data,allowing it to generate meaningful contextual information across various scales.This method effectively addresses both channel-mixing and channel-independence scenarios,facilitating the selection of relevant features for prediction by considering both global and local contexts at multiple scales.Results:Two large-scale multisite functional magnetic resonance imaging datasets,including REST-meta-MDD(n=1,642)and Autism Brain Imaging Data Exchange(ABIDE)(n=1,022),were used to validate the performance of our proposed approach.MSC-Mamba has achieved stateof-the-art performance,with an accuracy of 69.91%for MDD detection and 73.08%for ASD detection.The results demonstrate the model’s robust generalization across imaging sites and its sensitivity to intricate brain network dynamics.Conclusions:This paper demonstrates the potential of state-space models in advancing psychiatric neuroimaging research.The findings suggest that such models can significantly enhance detection accuracy for MDD and ASD,pointing toward more reliable,data-driven diagnostic tools in psychiatric disorder detection.展开更多
Background:It is still uncertain how multimorbidity patterns affect transitions between fall states among middle-aged and older Chinese.Methods:Data were obtained from China Health and Retirement Longitudinal Study(CH...Background:It is still uncertain how multimorbidity patterns affect transitions between fall states among middle-aged and older Chinese.Methods:Data were obtained from China Health and Retirement Longitudinal Study(CHARLS)2011–2018.We utilized latent class analysis to categorize baseline multimorbidity patterns,Markov multi-state model to explore the impact of multimorbidity characterized by condition counts and multimorbidity patterns on subsequent fall transitions,and Cox proportional hazard models to assess hazard ratios of each transition.Results:A total of 14,244 participants aged 45 years and older were enrolled at baseline.Among these participants,11,956(83.9%)did not have a fall history in the last 2 years,1,054(7.4%)had mild falls,and 1,234(8.7%)had severe falls.Using a multi-state model,10,967 transitions were observed during a total follow-up of 57,094 person-times,6,527 of which had worsening transitions and 4,440 had improving transitions.Among 6,711 multimorbid participants,osteocardiovascular(20.5%),pulmonary-digestive-rheumatic(30.5%),metabolic-cardiovascular(22.9%),and neuropsychiatric-sensory(26.1%)patterns were classified.Multimorbid participants had significantly higher risks of transitions compared with other participants.Among 4 multimorbidity patterns,osteocardiovascular pattern had higher transition risks than other 3 patterns.Conclusions:Multimorbidity,especially the“osteo-cardiovascular pattern”identified in this study,was associated with higher risks of fall transitions among middle-aged and older Chinese.Generally,the effect of multimorbidity is more significant in older adults than in middle-aged adults.Findings from this study provide facts and evidence for fall prevention,and offer implications for clinicians to target on vulnerable population,and for public health policymakers to allocate healthcare resources.展开更多
Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capab...Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capabilities and limitations of these models is critical for determining their suitability in test preparation.Methods:This study evaluated the ability of 16 LLMs,including general-purpose models(e.g.,GPT-3.5,GPT-4,GPT-4o,GPT-o1,Google’s Bard,mistral-large,and Claude),domain-specific finetuned models(e.g.,DentalGPT,MedGPT,and BioGPT),and open-source models(e.g.,Llama2-7B,Llama2-13B,Llama2-70B,Llama3-8B,and Llama3-70B),to answer questions from a sample DAT.Quantitative analysis was performed to assess model accuracy in different sections,and qualitative thematic analysis by subject matter experts examined specific challenges encountered by the models.Results:GPT-4o and GPT-o1 outperformed others in text-based questions assessing knowledge and comprehension,with GPT-o1 achieving perfect scores in the natural sciences(NS)and reading comprehension(RC)sections.Open-source models such as Llama3-70B also performed competitively in RC tasks.However,all models,including GPT-4o,struggled substantially with perceptual ability(PA)items,highlighting a persistent limitation in handling image-based tasks requiring visual-spatial reasoning.Fine-tuned medical models(e.g.,DentalGPT,MedGPT,and BioGPT)demonstrated moderate success in text-based tasks but underperformed in areas requiring critical thinking and reasoning.Thematic analysis identified key challenges,including difficulties with stepwise problem-solving,transferring knowledge,comprehending intricate questions,and hallucinations,particularly on advanced items.Conclusions:While LLMs show potential for reinforcing factual knowledge and supporting learners,their limitations in handling higherorder cognitive tasks and image-based reasoning underscore the need for judicious integration with instructor-led guidance and targeted practice.This study provides valuable insights into the capabilities and limitations of current LLMs in preparing prospective dental students and highlights pathways for future innovations to improve performance across all cognitive skills assessed by the DAT.展开更多
Background:Magnetic resonance imaging(MRI)is of considerable importance due to its wide range of applications in both scientific research and clinical diagnostics.Acquiring high-quality MRI data is of paramount import...Background:Magnetic resonance imaging(MRI)is of considerable importance due to its wide range of applications in both scientific research and clinical diagnostics.Acquiring high-quality MRI data is of paramount importance.Super-resolution reconstruction serves as a post-acquisition method capable of improving MRI data quality.Current methods predominantly utilize convolutional neural networks in super-resolution reconstruction.However,convolutional layers have inherent limitations in capturing extensive spatial dependencies due to their localized nature.Methods:We developed a new methodology that enables rapid and high-quality MRI data acquisition through a novel super-resolution approach.We proposed an innovative architecture using transformers to exploit long-range spatial dependencies present in images,allowing for an unsupervised learning framework specifically designed for super-resolution tasks tailored to individual subject.We validated our approach using both simulated data and clinical data comprising 40 scans acquired with a 3-T MRI system.Results:We obtained images with T2 contrast at an isotropic spatial resolution of 500μm in just 4 min of imaging time,and simultaneously,the signal-to-noise ratio and contrast-to-noise ratio were improved by 13.23% and 18.45%,respectively,in comparison to current leading super-resolution techniques.Conclusions:The results demonstrated that incorporating long-range spatial dependencies substantially improved super-resolution reconstruction,thereby allowing for the acquisition of high-quality MRI data with reduced imaging time.展开更多
Background:Sleep traits have been suggested to correlate with various diseases,but most evidence is based on subjective sleep measurement.We investigated the associations of accelerometer-derived objective sleep trait...Background:Sleep traits have been suggested to correlate with various diseases,but most evidence is based on subjective sleep measurement.We investigated the associations of accelerometer-derived objective sleep traits with diseases throughout physiological systems to ascertain whether the disease spectrum related to objective sleep traits differs from that related to subjective sleep traits.Methods:In 88,461 UK Biobank(UKB)adults wearing accelerometers,multiple dimensions of sleep were objectively derived:(a)nocturnal sleep duration and onset timing,(b)sleep rhythm(relative amplitude and interdaily stability),and(c)sleep fragmentation(sleep efficiency and waking numbers).Associations with International Classification of Diseases,10th Revision-decoded diseases during follow-up were estimated using the Cox model,and the results were compared with those of a published literature search of subjectively measured sleep traits and diseases.National Health and Nutrition Examination Survey(NHANES)data were used to validate the newly identified associations unreported by previous studies.For the meta-analysis-reported associations(with subjective sleep traits)that were negative(with objective sleep traits)in our study,reanalysis was done in UKB with subjective sleep traits,stratified by objective measurements.Results:During the average 6.8-year follow-up,172 diseases were associated with sleep traits.Among them,42 showed at least doubled disease risk,including age-related physical debility(lowest versus highest quartile of relative amplitude,hazard ratio[HR]=3.36,95% confidence interval[CI]:2.25,5.02),gangrene(lowest versus highest quartile of interdaily stability,HR=2.61,95%CI:1.41,4.83),and fibrosis and cirrhosis of the liver(sleep onset timing≥0030 versus 2300 to 2330,HR=2.57,95%CI:1.42,4.67).A total of 92 diseases had>20%burden attributable to sleep,such as Parkinson’s disease(37.05%,95%CI:21.02%,49.83%),type 2 diabetes(36.12%,95%CI:29.00%,42.52%),and acute kidney failure(21.85%,95%CI:13.47%,29.42%).Notably,83(48.3%)disease associations were sleep rhythm specific,distinct from existing subjective-measure literature that focused on sleep duration.Reanalysis in UKB showed a contamination of objectively short sleepers in self-report long sleepers,which induced false-positive associations in subjective meta-analyses,including for ischemic heart disease and depressive disorder.Newly identified associations of sleep rhythm with 4 diseases including chronic obstructive pulmonary disease and diabetes were successfully replicated in NHANES.A mediation analysis showed that inflammatory factors including leukocytes,eosinophils,and C-reactive protein contributed significantly to all these newly identified sleep-disease associations.Conclusions:Objective sleep traits showed a disease spectrum similar to but not identical to that of subjective sleep traits.Objective measurement can be a useful complement to sleep-disease studies as it may help overcome false-positive associations caused by misclassification bias of some subjective measurement such as sleep duration.Comprehensive control of multiple sleep traits may be important for health as substantial disease burden was attributed to different sleep traits.展开更多
Importance:Precisely decoding brain dysfunction from high-dimensional functional recordings is crucial for advancing our understanding of brain dysfunction in brain disorders.Self-supervised learning(SSL)models offer ...Importance:Precisely decoding brain dysfunction from high-dimensional functional recordings is crucial for advancing our understanding of brain dysfunction in brain disorders.Self-supervised learning(SSL)models offer a transformative approach for mapping dependencies in functional neuroimaging data.Leveraging the intrinsic organization of brain signals for comprehensive feature extraction,these models enable the analysis of critical neurofunctional features within a clinically relevant framework,overcoming challenges related to data heterogeneity and the scarcity of labeled data.Highlight:This paper provides a comprehensive overview of SSL techniques applied to functional neuroimaging data,such as functional magnetic resonance imaging and electroencephalography,with a specific focus on their applications in various neuropsychiatric disorders.We discuss 3 main categories of SSL methods:contrastive learning,generative learning,and generative-contrastive methods,outlining their basic principles and representative methods.Critically,we highlight the potential of SSL in addressing data scarcity,multimodal integration,and dynamic network modeling for disease detection and prediction.We showcase successful applications of these techniques in understanding and classifying conditions such as Alzheimer’s disease,Parkinson’s disease,and epilepsy,demonstrating their potential in downstream neuropsychological applications.Conclusion:SSL models provide a scalable and effective methodology for individual detection and prediction in brain disorders.Despite current limitations in interpretability and data heterogeneity,the potential of SSL for future clinical applications,particularly in the areas of transdiagnostic psychosis subtyping and decoding task-based brain functional recordings,is substantial.展开更多
Background:The traditional manual literature screening approach is limited by its time-consuming nature and high labor costs.A pressing issue is how to leverage large language models to enhance the efficiency and qual...Background:The traditional manual literature screening approach is limited by its time-consuming nature and high labor costs.A pressing issue is how to leverage large language models to enhance the efficiency and quality of evidence-based evaluations of drug efficacy and safety.Methods:This study utilized a manually curated reference literature database—comprising vaccine,hypoglycemic agent,and antidepressant evaluation studies—previously developed by our team through conventional systematic review methods.This validated database served as the gold standard for the development and optimization of LitAutoScreener.Following the PICOS(Population,Intervention,Comparison,Outcomes,Study Design)principles,a chain-of-thought reasoning approach with few-shot learning prompts was implemented to develop the screening algorithm.We subsequently evaluated the performance of LitAutoScreener using 2 independent validation cohorts,assessing both classification accuracy and processing efficiency.Results:For respiratory syncytial virus vaccine safety validation title-abstract screening,our tools based on GPT(GPT-4o),Kimi(moonshot-v1-128k),and DeepSeek(deepseek-chat 2.5)demonstrated high accuracy in inclusion/exclusion decisions(99.38%,98.94%,and 98.85%,respectively).Recall rates were 100.00%,99.13%,and 98.26%,with statistically significant performance differences(χ^(2)=5.99,P=0.048),where GPT outperformed the other models.Exclusion reason concordance rates were 98.85%,94.79%,and 96.47%(χ^(2)=30.22,P<0.001).In full-text screening,all models maintained perfect recall(100.00%),with accuracies of 100.00%(GPT),100.00%(Kimi),and 99.45%(DeepSeek).Processing times averaged 1 to 5 s per article for title–abstract screening and 60 s for full-text processing(including PDF preprocessing).Conclusions:LitAutoScreener offers a new approach for efficient literature screening in drug intervention studies,achieving high accuracy and significantly improving screening efficiency.展开更多
Background:China has the largest population with Alzheimer’s disease and related dementias(ADRDs)globally,and rapid population aging is expected to drive a substantial increase in cases.This study projects ADRD preva...Background:China has the largest population with Alzheimer’s disease and related dementias(ADRDs)globally,and rapid population aging is expected to drive a substantial increase in cases.This study projects ADRD prevalence and associated economic burdens across provinces in China from 2025 to 2060.Methods:Using data from the China Health and Retirement Longitudinal Study(CHARLS)supplemented by national demographic and provincial statistics,we projected the prevalence and care costs of ADRD for each of the 31 provinces in China from 2025 to 2060.Cost projections included formal care expenses and informal caregiving valued through replacement cost methods.We conducted uncertainty analysis to provide robust estimates for ADRD prevalence and costs.Results:By 2060,ADRD cases in China are projected to reach approximately 49.89 million,with the highest prevalence and economic burden concentrated in provinces such as Shandong,Sichuan,Jiangsu,Henan,and Guangdong.Formal care costs alone are expected to exceed$1 trillion annually,while the total economic value—including informal caregiving—could surpass$5 trillion.Geographic disparities highlight that Eastern and Central regions,with a higher proportions of older adults,will bear disproportionate costs.Informal caregiving is projected to constitute 60% to 80% of total ADRD-related costs.Conclusion:China faces an unprecedented rise in ADRD-related economic burden over the next 4 decades,with substantial regional disparities.Strengthening long-term care infrastructure,expanding financial and social support for caregivers,and implementing regionally tailored healthy aging policies are essential to ensuring equitable and sustainable ADRD care across China.展开更多
Background:The use of antidepressants in the treatment of bipolar depression remains controversial due to concerns about their potential to induce mood polarity switches.This multinational observational study aims to ...Background:The use of antidepressants in the treatment of bipolar depression remains controversial due to concerns about their potential to induce mood polarity switches.This multinational observational study aims to examine the association between the use of antidepressants and the risk of hypomanic/manic switch among bipolar depressive patients.Methods:Four electronic health record databases(IQVIA Disease Analyzer Germany,IQVIA Disease Analyzer France,IQVIA US Hospital Charge Data Master,and Beijing Anding Hospital)and one administrative claims database(IQVIA US Open Claims)were analyzed,and the study period covered from January 2013 until December 2017.Treatment patterns of patients with bipolar depression were collected.The hazard ratio(HR)was calculated by comparing the incidence of hypomanic/manic switch in patients who received antidepressants(AD group)with that in those who did not receive any antidepressant(non-AD group)in 730 days after the date of the first diagnosis of bipolar depression.Results:The analysis included a total of 122,843 patients from the 5 databases;60.6% of them received antidepressants for bipolar depression.Across the 5 data sources,the mean age at index date ranged from 37.50(15.72)to 52.10(16.22)years.After controlling potential confounders by propensity score matching,the AD group’s manic switch risk was not significantly higher than the non-AD group’s(HR 1.04[95%CI,0.96 to 1.13];P=0.989).Additionally,no statistically significant difference was observed between patients prescribed antimanic drugs and those who were not(HR 0.69[95%CI,0.38 to 1.25];P=0.535).Conclusions:This study indicated that antidepressants were widely used in clinical settings for managing bipolar depression.The use of antidepressants was not associated with the risk of mania/hypomania switch when compared to non-antidepressants treatment.Therefore,antidepressants could be considered a treatment option for bipolar depression.展开更多
Background:Hearing loss(HL)is one major cause of disability and can lead to social impairments.However,the relationship between loneliness and the risk of incident HL remains unclear.Our study aimed to investigate thi...Background:Hearing loss(HL)is one major cause of disability and can lead to social impairments.However,the relationship between loneliness and the risk of incident HL remains unclear.Our study aimed to investigate this association among adults in the UK.Methods:This cohort study was based on data from the UK Biobank study.Loneliness was assessed by asking participants if they often felt lonely.Incident HL was defined as a primary diagnosis,ascertained via linkage to electronic health records.Cox proportional hazard regression models were used to examine the association between loneliness and risk of incident HL.Results:Our analyses included 490,865 participants[mean(SD)age,56.5(8.1)years;54.4%female],among whom 90,893(18.5%)reported feeling lonely at baseline.Over a median follow-up period of 12.3 years(interquartile range,11.3 to 13.1),11,596 participants were diagnosed with incident HL.Compared to non-lonely participants,lonely individuals exhibited an increased risk of HL[hazard ratio(HR),1.36;95%confidence interval(CI),1.30 to 1.43].This association remained(HR,1.24;95%CI,1.17 to 1.31)after adjusting for potential confounders,including age,sex,socioeconomic status,biological and lifestyle factors,social isolation,depression,chronic diseases,use of ototoxic drugs,and genetic risk of HL.The joint analysis showed that loneliness was significantly associated with an increased risk of incident HL across all levels of genetic risks for HL.Conclusions:Loneliness was associated with the risk of incident HL independent of other prominent risk factors.Social enhancement strategies aimed at alleviating loneliness may prove beneficial in HL prevention.展开更多
Background:A neuroimaging-derived biomarker termed the brain age is considered to capture the degree and diversity in the aging process of the brain,serving as a robust indicator of overall brain health.The impact of ...Background:A neuroimaging-derived biomarker termed the brain age is considered to capture the degree and diversity in the aging process of the brain,serving as a robust indicator of overall brain health.The impact of different levels of physical activity(PA)intensities on brain age is still not fully understood.This study aimed to investigate the associations between accelerometer-measured PA and brain age.Methods:A total of 16,972 eligible participants with both valid T1-weighted neuroimaging and accelerometer data from the UK Biobank was included.Brain age was estimated using an ensemble learning approach called Light Gradient-Boosting Machine(LightGBM).Over 1,400 image-derived phenotypes(IDPs)were initially chosen to undergo data-driven feature selection for brain age prediction.A measure of accelerated brain aging,the brain age gap(BAG)can be derived by subtracting the chronological age from the estimated brain age.A positive BAG indicates accelerated brain aging.PA was measured over a 7-day period using wrist-worn accelerometers,and time spent on light-intensity PA(LPA),moderate-intensity PA(MPA),vigorous-intensity PA(VPA),and moderate-to vigorous-intensity PA(MVPA)was extracted.The generalized additive model was applied to examine the nonlinear association between PA and BAG after adjusting for potential confounders.Results:The brain age estimated by LightGBM achieved an appreciable performance(r=0.81,mean absolute error[MAE]=3.65),which was further improved by age bias correction(r=0.90,MAE=3.03).We found that LPA(F=2.47,P=0.04),MPA(F=6.49,P<1×10^(-300)),VPA(F=4.92,P=2.58×10^(-5)),and MVPA(F=6.45,P<1×10^(-300))exhibited an approximate U-shaped relationship with BAG,demonstrating that both insufficient and excessive PA levels adversely impact brain aging.Furthermore,mediation analysis suggested that BAG partially mediated the associations between PA and cognitive functions as well as brain-related disorders.Conclusions:Our study revealed a U-shaped association between accelerometer-measured PA and BAG,highlighting that advanced brain health may be attainable through engaging in moderate amounts of objectively measured PA irrespectively of intensities.展开更多
Background:Source-free unsupervised domain adaptation(SFUDA)methods aim to address the challenge of domain shift while preserving data privacy.Existing SFUDA approaches construct reliable and confident pseudo-labels f...Background:Source-free unsupervised domain adaptation(SFUDA)methods aim to address the challenge of domain shift while preserving data privacy.Existing SFUDA approaches construct reliable and confident pseudo-labels for target-domain data through denoising methods,thereby guiding the training of the target-domain model.The effectiveness of denoising approaches is influenced by the degree of domain gap between the source and target domains.A marked shift can cause the pseudo-labels to be unreliable,even after applying denoising.Methods:We propose a novel 2-stage framework for SFUDA called visual prompt source-free domain adaptation(VP-SFDA).We propose input-specific visual prompt in the first stage,prompting process,which bridges the target-domain data to source-domain distribution.Our method utilizes visual prompts and batch normalization constraint to enable the alignment model to learn domainspecific knowledge and align the target-domain data with the source-domain contribution.The second stage is the adaptation process,which aims at optimizing the segmentation model from the source domain to the target domain.This is accomplished through the denoising techniques,ultimately enhancing the performance.Results:Our study presents a comparative analysis of several SFUDA techniques in the VPSFDA framework across 4 tasks:abdominal magnetic resonance imaging(MRI)to computed tomography(CT),abdominal CT to MRI,cardiac MRI to CT,and cardiac CT to MRI.Notably,in the abdominal MRI to CT adaptation task,the VP-OS method achieved a remarkable improvement,increasing the average DICE score from 0.658 to 0.773(P<0.01)and reducing the average surface distance(ASD)from 3.489 to 2.961(P<0.01).Similarly,the VP-LD and VP-DPL methods also showed significant improvements over their base algorithms in both abdominal and cardiac MRI to CT tasks.Conclusions:This paper proposes VP-SFDA,a novel 2-stage framework for SFUDA in medical imaging,which achieves superior performance through input-specific visual prompts and batch normalization constraint for domain adaptation,coupled with denoising methods for enhanced results.Comparative experiments on 4 medical SFUDA tasks demonstrate that VO-SFDA surpasses existing methods,with ablation studies confirming the benefits of domain-specific patterns.展开更多
We sincerely appreciate the insightful comments and suggestions by Meng and colleagues[1]for our recently published paper,entitled“In-hospital mortality prediction among intensive care unit patients with acute ischem...We sincerely appreciate the insightful comments and suggestions by Meng and colleagues[1]for our recently published paper,entitled“In-hospital mortality prediction among intensive care unit patients with acute ischemic stroke:A machine learning approach”[2].We are pleased that our emphasis on model fairness,interpretability,and comprehensive benchmarking were found valuable,and their perspectives are very inspirational on expanding and enhancing our work.展开更多
End-stage renal disease(ESRD)significantly impacts patients’quality of life and poses substantial socioeconomic burdens.Dietary interventions are crucial for managing ESRD,yet high-quality evidence and analysis speci...End-stage renal disease(ESRD)significantly impacts patients’quality of life and poses substantial socioeconomic burdens.Dietary interventions are crucial for managing ESRD,yet high-quality evidence and analysis specifically linking diet to mortality outcomes is scarce.Methods:We conducted a comprehensive study involving 656 peritoneal dialysis(PD)patients over 12 years,with an average follow-up every 3 months.Dietary intake was meticulously recorded using a 3-day dietary record method,integrated with detailed health records and outcomes.We employed a 2-stage model to evaluate nonlinear relationships between dietary nutrients and mortality risk,accounting for various confounding factors.Findings:Our analysis revealed that 14 out of 26 nutritional elements lack guidelines for ESRD and PD patients,with 13 showing significant associations with mortality.For example,while guidelines suggest a dietary protein intake of 1.0 to 1.2 g/kg/d,our findings indicate an optimal range of 0.88 to 1.13 g/kg/d.Similarly,the recommended dietary energy intake of 25 to 35 kcal/kg/d was refined to 26 to 42 kcal/kg/d.We identified that 69% of dietary intake-outcome relationships are nonlinear,especially in patients with poor health status.Interpretations:Our study provides detailed dietary intake thresholds that correlate with improved prognosis in ESRD patients,enhancing current guidelines.The findings highlight the importance of personalized nutritional management and underscore the nonlinear nature of nutrient–disease relationships,particularly in severely ill patients.This approach can refine dietary recommendations and improve patient care in ESRD.展开更多
Background:The electrocardiogram(ECG)is a valuable,noninvasive tool for monitoring heart-related conditions,providing critical insights.However,the interpretation of ECG data alongside patient information demands subs...Background:The electrocardiogram(ECG)is a valuable,noninvasive tool for monitoring heart-related conditions,providing critical insights.However,the interpretation of ECG data alongside patient information demands substantial medical expertise and resources.While deep learning methods help streamline this process,they often fall short in integrating patient data with ECG readings and do not provide the nuanced clinical suggestions and insights necessary for accurate diagnosis.Methods:Although recent advancements in multi-modal large language modeling have propelled their application scope beyond the natural language processing domain,their applicability to ECG processing remains largely unexplored,partly due to the lack of text–ECG data.To this end,we develop ECG-Language Model(ECG-LM),the first multi-modal large language model able to process natural language and understand ECG signals.The model employs a specialized ECG encoder that transforms raw ECG signals into a high-dimensional feature space,which is then aligned with the textual feature space derived from the large language model.To address the scarcity of text–ECG data,we generated text–ECG pairs by leveraging detailed ECG pattern descriptions from medical guidelines,creating a robust dataset for pre-training ECG-LM.Additionally,we fine-tune ECG-LM with public clinical conversation datasets and build an additional supervised fine-tuning dataset based on real clinical data from the hospital,aiming to provide a more comprehensive and customized user experience.Results:ECG-LM outperforms existing few-shot and zero-shot solutions in cardiovascular disease detection across all 3 tasks(diagnostic,rhythm,and form)while also demonstrating strong potential in ECG-related question answering.Conclusions:The results across various tasks demonstrate that ECG-LM effectively captures the intricate features of ECGs,showcasing its versatility in applications such as disease prediction and advanced question answering.展开更多
Background:Multimodal large language models(LLMs)have shown potential in various health-related fields.However,many healthcare studies have raised concerns about the reliability and biases of LLMs in healthcare applic...Background:Multimodal large language models(LLMs)have shown potential in various health-related fields.However,many healthcare studies have raised concerns about the reliability and biases of LLMs in healthcare applications.Methods:To explore the practical application of multimodal LLMs in skin disease identification,and to evaluate sex and age biases,we tested the performance of 2 popular multimodal LLMs,ChatGPT-4 and LLaVA-1.6,across diverse sex and age groups using a subset of a large dermatoscopic dataset containing around 10,000 images and 3 skin diseases(melanoma,melanocytic nevi,and benign keratosis-like lesions).Results:In comparison to 3 deep learning models(VGG16,ResNet50,and Model Derm)based on convolutional neural network(CNN),one vision transformer model(Swin-B),we found that ChatGPT-4 and LLaVA-1.6 demonstrated overall accuracies that were 3% and 23% higher(and F1-scores that were 4% and 34% higher),respectively,than the best performing CNN-based baseline while maintaining accuracies that were 38% and 26% lower(and F1-scores that were 38% and 19% lower),respectively,than Swin-B.Meanwhile,ChatGPT-4 is generally unbiased in identifying these skin diseases across sex and age groups,while LLaVA-1.6 is generally unbiased across age groups,in contrast to Swin-B,which is biased in identifying melanocytic nevi.Conclusions:This study suggests the usefulness and fairness of LLMs in dermatological applications,aiding physicians and practitioners with diagnostic recommendations and patient screening.To further verify and evaluate the reliability and fairness of LLMs in healthcare,experiments using larger and more diverse datasets need to be performed in the future.展开更多
We value Cummins et al.’s study on an XGBoost model for in-hospital mortality prediction in intensive care unit(ICU)patients with acute ischemic stroke(AIS)[1].Its focus on model fairness across demographics and in-d...We value Cummins et al.’s study on an XGBoost model for in-hospital mortality prediction in intensive care unit(ICU)patients with acute ischemic stroke(AIS)[1].Its focus on model fairness across demographics and in-depth feature analysis gives useful references for clinical predictive analytics.To boost the model’s reproducibility,we suggest clarifying specific inclusion criteria and data preprocessing methods for the electronic ICU(eICU)database in future external validation.展开更多
基金supported by grants from the National Natural Science Foundation of China(82470736,22321005,824B2015,82170711,82070733)Beijing Nova Program(20220484147,20240484677)+4 种基金Beijing Natural Science Foun dation(7242144)the National Key Research and Development Program of China(2024YFC2511000)the Funda mental Research Funds for the Central Universities(Peking University Clinical Scientist Training Program)(BMU2024 PYJH021)the National High Level Hospital Clinical Research Funding(Interdisciplinary Research Project of Peking University First Hospital)(2023IR12)CAMS Innovation Fund for Medical Sciences(2019-I2M-5-046).
文摘Background:Recently,several cutting-edge experimental studies have directed chimeric antigen receptor(CAR)-T therapies toward specific renal diseases,revealing substantial renal benefits.Prior to widespread implementation of these animal experiments and potentially clinical trials,it is crucial to assess the renal safety of CAR-T therapies using real-world safety evidence.Methods:Our focus was on utilizing 4 algorithms,including disproportionality analysis,based on the US Food and Drug Administration Adverse Event Reporting System database,to filter positive signals of acute and chronic renal injury associated with 6 CAR-T therapies.Further determination of causality was achieved through Mendelian randomization(MR)for drugs associated with renal injury events showing a correlation.Results:Six therapies were evaluated involving a total of 9,770 patients,with only acute kidney injury(AKI)identified as associated with idecabtagene vicleucel treatment using 4 algorithmic thresholds,including disproportionality analysis.Subsequently,MR revealed no causal relationship between the idecabtagene vicleucel target B cell maturation antigen and the risk of AKI(P=0.576),a finding validated in another independent dataset(P=0.734).Conclusion:CAR-T therapies do not directly cause renal damage and necessitate controlling adverse renal risks during or after treatment,such as cytokine release syndrome.Future research efforts should rigorously optimize these aspects to better cater to nephrologists’requirements.
基金supported by the Duke/Duke-NUS Collaboration grant.
文摘Background:Clinical and biomedical research in low-resource settings often faces substantial challenges due to the need for high-quality data with sufficient sample sizes to construct effective models.These constraints hinder robust model training and prompt researchers to seek methods for leveraging existing knowledge from related studies to support new research efforts.Transfer learning(TL),a machine learning technique,emerges as a powerful solution by utilizing knowledge from pretrained models to enhance the performance of new models,offering promise across various healthcare domains.Despite its conceptual origins in the 1990s,the application of TL in medical research has remained limited,especially beyond image analysis.This review aims to analyze TL applications,highlight overlooked techniques,and suggest improvements for future healthcare research.Methods:Following the PRISMA-ScR guidelines,we conducted a search for published articles that employed TL with structured clinical or biomedical data by searching the SCOPUS,MEDLINE,Web of Science,Embase,and CINAHL databases.Results:We screened 5,080 papers,with 86 meeting the inclusion criteria.Among these,only 2%(2 of 86)utilized external studies,and 5%(4 of 86)addressed scenarios involving multi-site collaborations with privacy constraints.Conclusions:To achieve actionable TL with structured medical data while addressing regional disparities,inequality,and privacy constraints in healthcare research,we advocate for the careful identification of appropriate source data and models,the selection of suitable TL frameworks,and the validation of TL models with proper baselines.
文摘Background:There are few data on the comorbidity of diabetes in Chinese patients with depression.We aimed to calculate the prevalence and explore risk factors of type 2 diabetes mellitus(T2DM)among depression inpatients from 2005 to 2018 in Beijing.Methods:This study is a cross-sectional study.The data collected from 19 specialized psychiatric hospitals in Beijing were analyzed.The prevalence of T2DM and its distribution were analyzed.The multivariable logistic regression was performed to explore the risk factors of T2DM.Results:A total of 20,899 depression inpatients were included.The prevalence of T2DM was 9.13%[95%confidence interval(CI),8.74%to 9.52%].The prevalence of T2DM showed an upward trend with year(P for trend<0.001)and age(P for trend<0.001).The prevalence of T2DM was higher among readmitted patients(12.97%)and patients with comorbid hypertension(26.16%),hyperlipidemia(21.28%),and nonalcoholic fatty liver disease(NAFLD)(18.85%).The prevalence of T2DM in females was lower than in males among patients aged 18 to 59 years,while the prevalence of T2DM in females was higher than in males among patients aged≥60 years.T2DM was associated with older age[adjusted odds ratios(aORs)ranged from 3.68 to 29.95,P<0.001],hypertension(aOR,3.01;95%CI,2.70 to 3.35;P<0.001),hyperlipidemia(aOR,1.69;95%CI,1.50 to 1.91;P<0.001),and NAFLD(aOR,1.58;95%CI,1.37 to 1.82;P<0.001).Conclusions:The prevalence of T2DM among depression inpatients from 2005 to 2018 in Beijing was high and increased with the year.Depression inpatients who were older and with hypertension,hyperlipidemia,and NAFLD had a higher prevalence and risk of T2DM.
基金supported by grants from the National Natural Science Foundation of P.R.China(62276081 and 62106113)the Guangdong Basic and Applied Basic Research Foundation(2023A1515010792 and 2023B1515120065)the Shenzhen Science and Technology Program(GXWD20231129121139001 and JCYJ20240813110522029).
文摘Background:Major depressive disorder(MDD)and autism spectrum disorder(ASD)are complex and heterogeneous neuropsychiatric disorders with overlapping symptoms,presenting remarkable challenges for accurate diagnosis.Leveraging functional neuroimaging data offers an opportunity to develop more robust,data-driven approach for psychiatric disorder detection.However,existing methods often struggle to capture the long-term dependencies and dynamic patterns inherent in such data,particularly across diverse imaging sites.Methods:We propose Multiscale Contextual Mamba(MSC-Mamba),a Mamba-based model designed for capturing long-term dependencies in multivariate time-series data while maintaining linear scalability,allowing us to account for long-range interactions and subtle dynamic patterns within the brain’s functional networks.One of the main advantages of MSC-Mamba is its ability to leverage the distinct characteristics of time-series data,allowing it to generate meaningful contextual information across various scales.This method effectively addresses both channel-mixing and channel-independence scenarios,facilitating the selection of relevant features for prediction by considering both global and local contexts at multiple scales.Results:Two large-scale multisite functional magnetic resonance imaging datasets,including REST-meta-MDD(n=1,642)and Autism Brain Imaging Data Exchange(ABIDE)(n=1,022),were used to validate the performance of our proposed approach.MSC-Mamba has achieved stateof-the-art performance,with an accuracy of 69.91%for MDD detection and 73.08%for ASD detection.The results demonstrate the model’s robust generalization across imaging sites and its sensitivity to intricate brain network dynamics.Conclusions:This paper demonstrates the potential of state-space models in advancing psychiatric neuroimaging research.The findings suggest that such models can significantly enhance detection accuracy for MDD and ASD,pointing toward more reliable,data-driven diagnostic tools in psychiatric disorder detection.
基金funded by the National Key R&D Program of China(2023YFB4603200 and 2023YFC3606400)National Natural Science Foundation of China(72374013)Capital’s Funds for Health Improvement and Research(CFH 2024-1G-3014).
文摘Background:It is still uncertain how multimorbidity patterns affect transitions between fall states among middle-aged and older Chinese.Methods:Data were obtained from China Health and Retirement Longitudinal Study(CHARLS)2011–2018.We utilized latent class analysis to categorize baseline multimorbidity patterns,Markov multi-state model to explore the impact of multimorbidity characterized by condition counts and multimorbidity patterns on subsequent fall transitions,and Cox proportional hazard models to assess hazard ratios of each transition.Results:A total of 14,244 participants aged 45 years and older were enrolled at baseline.Among these participants,11,956(83.9%)did not have a fall history in the last 2 years,1,054(7.4%)had mild falls,and 1,234(8.7%)had severe falls.Using a multi-state model,10,967 transitions were observed during a total follow-up of 57,094 person-times,6,527 of which had worsening transitions and 4,440 had improving transitions.Among 6,711 multimorbid participants,osteocardiovascular(20.5%),pulmonary-digestive-rheumatic(30.5%),metabolic-cardiovascular(22.9%),and neuropsychiatric-sensory(26.1%)patterns were classified.Multimorbid participants had significantly higher risks of transitions compared with other participants.Among 4 multimorbidity patterns,osteocardiovascular pattern had higher transition risks than other 3 patterns.Conclusions:Multimorbidity,especially the“osteo-cardiovascular pattern”identified in this study,was associated with higher risks of fall transitions among middle-aged and older Chinese.Generally,the effect of multimorbidity is more significant in older adults than in middle-aged adults.Findings from this study provide facts and evidence for fall prevention,and offer implications for clinicians to target on vulnerable population,and for public health policymakers to allocate healthcare resources.
基金partially supported by the National Institutes of Health’s National Center for Complementary and Integrative Health under grant number R01AT009457National Institute on Aging under grant number R01AG078154National Cancer Institute under grant number R01CA287413.
文摘Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capabilities and limitations of these models is critical for determining their suitability in test preparation.Methods:This study evaluated the ability of 16 LLMs,including general-purpose models(e.g.,GPT-3.5,GPT-4,GPT-4o,GPT-o1,Google’s Bard,mistral-large,and Claude),domain-specific finetuned models(e.g.,DentalGPT,MedGPT,and BioGPT),and open-source models(e.g.,Llama2-7B,Llama2-13B,Llama2-70B,Llama3-8B,and Llama3-70B),to answer questions from a sample DAT.Quantitative analysis was performed to assess model accuracy in different sections,and qualitative thematic analysis by subject matter experts examined specific challenges encountered by the models.Results:GPT-4o and GPT-o1 outperformed others in text-based questions assessing knowledge and comprehension,with GPT-o1 achieving perfect scores in the natural sciences(NS)and reading comprehension(RC)sections.Open-source models such as Llama3-70B also performed competitively in RC tasks.However,all models,including GPT-4o,struggled substantially with perceptual ability(PA)items,highlighting a persistent limitation in handling image-based tasks requiring visual-spatial reasoning.Fine-tuned medical models(e.g.,DentalGPT,MedGPT,and BioGPT)demonstrated moderate success in text-based tasks but underperformed in areas requiring critical thinking and reasoning.Thematic analysis identified key challenges,including difficulties with stepwise problem-solving,transferring knowledge,comprehending intricate questions,and hallucinations,particularly on advanced items.Conclusions:While LLMs show potential for reinforcing factual knowledge and supporting learners,their limitations in handling higherorder cognitive tasks and image-based reasoning underscore the need for judicious integration with instructor-led guidance and targeted practice.This study provides valuable insights into the capabilities and limitations of current LLMs in preparing prospective dental students and highlights pathways for future innovations to improve performance across all cognitive skills assessed by the DAT.
基金supported in part by the Beijing Natural Science Foundation under Award Number L258055in part by the Major Program of the National Natural Science Foundation of China under Award Numbers 62394310 and 62394312in part by the National Institutes of Health(NIH)under Award Numbers R01 EB019483,R01 NS106030,R01 NS124212,R01 LM013608,R01 HD109395,R01 EB031849,R01 NS133228,R01 NS121657,R21 EB036105,and S10OD025111.
文摘Background:Magnetic resonance imaging(MRI)is of considerable importance due to its wide range of applications in both scientific research and clinical diagnostics.Acquiring high-quality MRI data is of paramount importance.Super-resolution reconstruction serves as a post-acquisition method capable of improving MRI data quality.Current methods predominantly utilize convolutional neural networks in super-resolution reconstruction.However,convolutional layers have inherent limitations in capturing extensive spatial dependencies due to their localized nature.Methods:We developed a new methodology that enables rapid and high-quality MRI data acquisition through a novel super-resolution approach.We proposed an innovative architecture using transformers to exploit long-range spatial dependencies present in images,allowing for an unsupervised learning framework specifically designed for super-resolution tasks tailored to individual subject.We validated our approach using both simulated data and clinical data comprising 40 scans acquired with a 3-T MRI system.Results:We obtained images with T2 contrast at an isotropic spatial resolution of 500μm in just 4 min of imaging time,and simultaneously,the signal-to-noise ratio and contrast-to-noise ratio were improved by 13.23% and 18.45%,respectively,in comparison to current leading super-resolution techniques.Conclusions:The results demonstrated that incorporating long-range spatial dependencies substantially improved super-resolution reconstruction,thereby allowing for the acquisition of high-quality MRI data with reduced imaging time.
基金supported by the National Key R&D Program of China(grant numbers:2022YFC3602900 and 2022YFC2702900)the National Natural Science Foundation of China(grant number:82273747)the Beijing Municipal Health Development Research Fund(grant number:2022-1G-1021).
文摘Background:Sleep traits have been suggested to correlate with various diseases,but most evidence is based on subjective sleep measurement.We investigated the associations of accelerometer-derived objective sleep traits with diseases throughout physiological systems to ascertain whether the disease spectrum related to objective sleep traits differs from that related to subjective sleep traits.Methods:In 88,461 UK Biobank(UKB)adults wearing accelerometers,multiple dimensions of sleep were objectively derived:(a)nocturnal sleep duration and onset timing,(b)sleep rhythm(relative amplitude and interdaily stability),and(c)sleep fragmentation(sleep efficiency and waking numbers).Associations with International Classification of Diseases,10th Revision-decoded diseases during follow-up were estimated using the Cox model,and the results were compared with those of a published literature search of subjectively measured sleep traits and diseases.National Health and Nutrition Examination Survey(NHANES)data were used to validate the newly identified associations unreported by previous studies.For the meta-analysis-reported associations(with subjective sleep traits)that were negative(with objective sleep traits)in our study,reanalysis was done in UKB with subjective sleep traits,stratified by objective measurements.Results:During the average 6.8-year follow-up,172 diseases were associated with sleep traits.Among them,42 showed at least doubled disease risk,including age-related physical debility(lowest versus highest quartile of relative amplitude,hazard ratio[HR]=3.36,95% confidence interval[CI]:2.25,5.02),gangrene(lowest versus highest quartile of interdaily stability,HR=2.61,95%CI:1.41,4.83),and fibrosis and cirrhosis of the liver(sleep onset timing≥0030 versus 2300 to 2330,HR=2.57,95%CI:1.42,4.67).A total of 92 diseases had>20%burden attributable to sleep,such as Parkinson’s disease(37.05%,95%CI:21.02%,49.83%),type 2 diabetes(36.12%,95%CI:29.00%,42.52%),and acute kidney failure(21.85%,95%CI:13.47%,29.42%).Notably,83(48.3%)disease associations were sleep rhythm specific,distinct from existing subjective-measure literature that focused on sleep duration.Reanalysis in UKB showed a contamination of objectively short sleepers in self-report long sleepers,which induced false-positive associations in subjective meta-analyses,including for ischemic heart disease and depressive disorder.Newly identified associations of sleep rhythm with 4 diseases including chronic obstructive pulmonary disease and diabetes were successfully replicated in NHANES.A mediation analysis showed that inflammatory factors including leukocytes,eosinophils,and C-reactive protein contributed significantly to all these newly identified sleep-disease associations.Conclusions:Objective sleep traits showed a disease spectrum similar to but not identical to that of subjective sleep traits.Objective measurement can be a useful complement to sleep-disease studies as it may help overcome false-positive associations caused by misclassification bias of some subjective measurement such as sleep duration.Comprehensive control of multiple sleep traits may be important for health as substantial disease burden was attributed to different sleep traits.
基金supported by grants from the National Natural Science Foundation of P.R.China(62276081 and 62106113)Guangdong Basic and Applied Basic Research Foundation(2023A1515010792 and 2023B1515120065)Shenzhen Science and Technology Program(GXWD20231129121139001 and JCYJ20240813110522029).
文摘Importance:Precisely decoding brain dysfunction from high-dimensional functional recordings is crucial for advancing our understanding of brain dysfunction in brain disorders.Self-supervised learning(SSL)models offer a transformative approach for mapping dependencies in functional neuroimaging data.Leveraging the intrinsic organization of brain signals for comprehensive feature extraction,these models enable the analysis of critical neurofunctional features within a clinically relevant framework,overcoming challenges related to data heterogeneity and the scarcity of labeled data.Highlight:This paper provides a comprehensive overview of SSL techniques applied to functional neuroimaging data,such as functional magnetic resonance imaging and electroencephalography,with a specific focus on their applications in various neuropsychiatric disorders.We discuss 3 main categories of SSL methods:contrastive learning,generative learning,and generative-contrastive methods,outlining their basic principles and representative methods.Critically,we highlight the potential of SSL in addressing data scarcity,multimodal integration,and dynamic network modeling for disease detection and prediction.We showcase successful applications of these techniques in understanding and classifying conditions such as Alzheimer’s disease,Parkinson’s disease,and epilepsy,demonstrating their potential in downstream neuropsychological applications.Conclusion:SSL models provide a scalable and effective methodology for individual detection and prediction in brain disorders.Despite current limitations in interpretability and data heterogeneity,the potential of SSL for future clinical applications,particularly in the areas of transdiagnostic psychosis subtyping and decoding task-based brain functional recordings,is substantial.
基金supported by the National Natural Science Foundation of China(grant no.72074011)the Research Project of China Drug Supervision and Administration Research Association(2025-Y-Y-012).
文摘Background:The traditional manual literature screening approach is limited by its time-consuming nature and high labor costs.A pressing issue is how to leverage large language models to enhance the efficiency and quality of evidence-based evaluations of drug efficacy and safety.Methods:This study utilized a manually curated reference literature database—comprising vaccine,hypoglycemic agent,and antidepressant evaluation studies—previously developed by our team through conventional systematic review methods.This validated database served as the gold standard for the development and optimization of LitAutoScreener.Following the PICOS(Population,Intervention,Comparison,Outcomes,Study Design)principles,a chain-of-thought reasoning approach with few-shot learning prompts was implemented to develop the screening algorithm.We subsequently evaluated the performance of LitAutoScreener using 2 independent validation cohorts,assessing both classification accuracy and processing efficiency.Results:For respiratory syncytial virus vaccine safety validation title-abstract screening,our tools based on GPT(GPT-4o),Kimi(moonshot-v1-128k),and DeepSeek(deepseek-chat 2.5)demonstrated high accuracy in inclusion/exclusion decisions(99.38%,98.94%,and 98.85%,respectively).Recall rates were 100.00%,99.13%,and 98.26%,with statistically significant performance differences(χ^(2)=5.99,P=0.048),where GPT outperformed the other models.Exclusion reason concordance rates were 98.85%,94.79%,and 96.47%(χ^(2)=30.22,P<0.001).In full-text screening,all models maintained perfect recall(100.00%),with accuracies of 100.00%(GPT),100.00%(Kimi),and 99.45%(DeepSeek).Processing times averaged 1 to 5 s per article for title–abstract screening and 60 s for full-text processing(including PDF preprocessing).Conclusions:LitAutoScreener offers a new approach for efficient literature screening in drug intervention studies,achieving high accuracy and significantly improving screening efficiency.
基金funded by the National Natural Science Foundation of China(Grant Nos.72404183,72293585,72293580,and 72125009).
文摘Background:China has the largest population with Alzheimer’s disease and related dementias(ADRDs)globally,and rapid population aging is expected to drive a substantial increase in cases.This study projects ADRD prevalence and associated economic burdens across provinces in China from 2025 to 2060.Methods:Using data from the China Health and Retirement Longitudinal Study(CHARLS)supplemented by national demographic and provincial statistics,we projected the prevalence and care costs of ADRD for each of the 31 provinces in China from 2025 to 2060.Cost projections included formal care expenses and informal caregiving valued through replacement cost methods.We conducted uncertainty analysis to provide robust estimates for ADRD prevalence and costs.Results:By 2060,ADRD cases in China are projected to reach approximately 49.89 million,with the highest prevalence and economic burden concentrated in provinces such as Shandong,Sichuan,Jiangsu,Henan,and Guangdong.Formal care costs alone are expected to exceed$1 trillion annually,while the total economic value—including informal caregiving—could surpass$5 trillion.Geographic disparities highlight that Eastern and Central regions,with a higher proportions of older adults,will bear disproportionate costs.Informal caregiving is projected to constitute 60% to 80% of total ADRD-related costs.Conclusion:China faces an unprecedented rise in ADRD-related economic burden over the next 4 decades,with substantial regional disparities.Strengthening long-term care infrastructure,expanding financial and social support for caregivers,and implementing regionally tailored healthy aging policies are essential to ensuring equitable and sustainable ADRD care across China.
基金partially supported by the Beijing Municipal Administration of Hospitals Incubating Program(PX20211903 and PX2019071)Capital’s Funds for Health Improvement and Research(2024-4-2129)+1 种基金the Beijing Municipal Science&Technology Commission(Z221100007422010)the Beijing High Level Public Health Professionals Training Plan(xuekegugan-01-12).
文摘Background:The use of antidepressants in the treatment of bipolar depression remains controversial due to concerns about their potential to induce mood polarity switches.This multinational observational study aims to examine the association between the use of antidepressants and the risk of hypomanic/manic switch among bipolar depressive patients.Methods:Four electronic health record databases(IQVIA Disease Analyzer Germany,IQVIA Disease Analyzer France,IQVIA US Hospital Charge Data Master,and Beijing Anding Hospital)and one administrative claims database(IQVIA US Open Claims)were analyzed,and the study period covered from January 2013 until December 2017.Treatment patterns of patients with bipolar depression were collected.The hazard ratio(HR)was calculated by comparing the incidence of hypomanic/manic switch in patients who received antidepressants(AD group)with that in those who did not receive any antidepressant(non-AD group)in 730 days after the date of the first diagnosis of bipolar depression.Results:The analysis included a total of 122,843 patients from the 5 databases;60.6% of them received antidepressants for bipolar depression.Across the 5 data sources,the mean age at index date ranged from 37.50(15.72)to 52.10(16.22)years.After controlling potential confounders by propensity score matching,the AD group’s manic switch risk was not significantly higher than the non-AD group’s(HR 1.04[95%CI,0.96 to 1.13];P=0.989).Additionally,no statistically significant difference was observed between patients prescribed antimanic drugs and those who were not(HR 0.69[95%CI,0.38 to 1.25];P=0.535).Conclusions:This study indicated that antidepressants were widely used in clinical settings for managing bipolar depression.The use of antidepressants was not associated with the risk of mania/hypomania switch when compared to non-antidepressants treatment.Therefore,antidepressants could be considered a treatment option for bipolar depression.
基金supported by the Young Elite Scientists Sponsorship Program by China Association for Science and Technology(grant number 2020QNRC001 to Y.X.)the LiaoNing Revitalization Talents Program(grant number XLYC2203168 to Y.X.)+1 种基金the 345 Talent Project of Shengjing Hospital of China Medical University(grant number M0294 to Y.X.)the Scientific Research Project of the Liaoning Province Education Department(grant number LJKMZ20221149 to Y.X.).
文摘Background:Hearing loss(HL)is one major cause of disability and can lead to social impairments.However,the relationship between loneliness and the risk of incident HL remains unclear.Our study aimed to investigate this association among adults in the UK.Methods:This cohort study was based on data from the UK Biobank study.Loneliness was assessed by asking participants if they often felt lonely.Incident HL was defined as a primary diagnosis,ascertained via linkage to electronic health records.Cox proportional hazard regression models were used to examine the association between loneliness and risk of incident HL.Results:Our analyses included 490,865 participants[mean(SD)age,56.5(8.1)years;54.4%female],among whom 90,893(18.5%)reported feeling lonely at baseline.Over a median follow-up period of 12.3 years(interquartile range,11.3 to 13.1),11,596 participants were diagnosed with incident HL.Compared to non-lonely participants,lonely individuals exhibited an increased risk of HL[hazard ratio(HR),1.36;95%confidence interval(CI),1.30 to 1.43].This association remained(HR,1.24;95%CI,1.17 to 1.31)after adjusting for potential confounders,including age,sex,socioeconomic status,biological and lifestyle factors,social isolation,depression,chronic diseases,use of ototoxic drugs,and genetic risk of HL.The joint analysis showed that loneliness was significantly associated with an increased risk of incident HL across all levels of genetic risks for HL.Conclusions:Loneliness was associated with the risk of incident HL independent of other prominent risk factors.Social enhancement strategies aimed at alleviating loneliness may prove beneficial in HL prevention.
基金supported by the National Natural Science Foundation of China(grant number 72204071)the Zhejiang Provincial Natural Science Foundation of China(grant number LY23G030005)the Scientific Research Foundation for Scholars of HZNU(grant number 4265C50221204119).
文摘Background:A neuroimaging-derived biomarker termed the brain age is considered to capture the degree and diversity in the aging process of the brain,serving as a robust indicator of overall brain health.The impact of different levels of physical activity(PA)intensities on brain age is still not fully understood.This study aimed to investigate the associations between accelerometer-measured PA and brain age.Methods:A total of 16,972 eligible participants with both valid T1-weighted neuroimaging and accelerometer data from the UK Biobank was included.Brain age was estimated using an ensemble learning approach called Light Gradient-Boosting Machine(LightGBM).Over 1,400 image-derived phenotypes(IDPs)were initially chosen to undergo data-driven feature selection for brain age prediction.A measure of accelerated brain aging,the brain age gap(BAG)can be derived by subtracting the chronological age from the estimated brain age.A positive BAG indicates accelerated brain aging.PA was measured over a 7-day period using wrist-worn accelerometers,and time spent on light-intensity PA(LPA),moderate-intensity PA(MPA),vigorous-intensity PA(VPA),and moderate-to vigorous-intensity PA(MVPA)was extracted.The generalized additive model was applied to examine the nonlinear association between PA and BAG after adjusting for potential confounders.Results:The brain age estimated by LightGBM achieved an appreciable performance(r=0.81,mean absolute error[MAE]=3.65),which was further improved by age bias correction(r=0.90,MAE=3.03).We found that LPA(F=2.47,P=0.04),MPA(F=6.49,P<1×10^(-300)),VPA(F=4.92,P=2.58×10^(-5)),and MVPA(F=6.45,P<1×10^(-300))exhibited an approximate U-shaped relationship with BAG,demonstrating that both insufficient and excessive PA levels adversely impact brain aging.Furthermore,mediation analysis suggested that BAG partially mediated the associations between PA and cognitive functions as well as brain-related disorders.Conclusions:Our study revealed a U-shaped association between accelerometer-measured PA and BAG,highlighting that advanced brain health may be attainable through engaging in moderate amounts of objectively measured PA irrespectively of intensities.
基金supportted by the Natural Science Foundation of China(62394311,62394310)Beijing Natural Science Foundation(QY24034)National Biomedical Imaging Facility Grant and from the startup funds of Peking University Health Science Center.
文摘Background:Source-free unsupervised domain adaptation(SFUDA)methods aim to address the challenge of domain shift while preserving data privacy.Existing SFUDA approaches construct reliable and confident pseudo-labels for target-domain data through denoising methods,thereby guiding the training of the target-domain model.The effectiveness of denoising approaches is influenced by the degree of domain gap between the source and target domains.A marked shift can cause the pseudo-labels to be unreliable,even after applying denoising.Methods:We propose a novel 2-stage framework for SFUDA called visual prompt source-free domain adaptation(VP-SFDA).We propose input-specific visual prompt in the first stage,prompting process,which bridges the target-domain data to source-domain distribution.Our method utilizes visual prompts and batch normalization constraint to enable the alignment model to learn domainspecific knowledge and align the target-domain data with the source-domain contribution.The second stage is the adaptation process,which aims at optimizing the segmentation model from the source domain to the target domain.This is accomplished through the denoising techniques,ultimately enhancing the performance.Results:Our study presents a comparative analysis of several SFUDA techniques in the VPSFDA framework across 4 tasks:abdominal magnetic resonance imaging(MRI)to computed tomography(CT),abdominal CT to MRI,cardiac MRI to CT,and cardiac CT to MRI.Notably,in the abdominal MRI to CT adaptation task,the VP-OS method achieved a remarkable improvement,increasing the average DICE score from 0.658 to 0.773(P<0.01)and reducing the average surface distance(ASD)from 3.489 to 2.961(P<0.01).Similarly,the VP-LD and VP-DPL methods also showed significant improvements over their base algorithms in both abdominal and cardiac MRI to CT tasks.Conclusions:This paper proposes VP-SFDA,a novel 2-stage framework for SFUDA in medical imaging,which achieves superior performance through input-specific visual prompts and batch normalization constraint for domain adaptation,coupled with denoising methods for enhanced results.Comparative experiments on 4 medical SFUDA tasks demonstrate that VO-SFDA surpasses existing methods,with ablation studies confirming the benefits of domain-specific patterns.
文摘We sincerely appreciate the insightful comments and suggestions by Meng and colleagues[1]for our recently published paper,entitled“In-hospital mortality prediction among intensive care unit patients with acute ischemic stroke:A machine learning approach”[2].We are pleased that our emphasis on model fairness,interpretability,and comprehensive benchmarking were found valuable,and their perspectives are very inspirational on expanding and enhancing our work.
基金supported by the National Natural Science Foundation of China(62402017 and 82470774)Xuzhou Scientific Technological Projects(KC23143)+2 种基金the Clinical Cohort Construction Program of Peking University Third Hospital(No.BYSYDL2023004)Peking University Medicine plus X Pilot Program-Key Technologies R&D Project(2024YXXLHGG007)the receipt of studentship awards from the Health Data Research UK-The Alan Turing Institute Wellcome PhD Programme in Health Data Science(Grant Ref:218529/Z/19/Z).
文摘End-stage renal disease(ESRD)significantly impacts patients’quality of life and poses substantial socioeconomic burdens.Dietary interventions are crucial for managing ESRD,yet high-quality evidence and analysis specifically linking diet to mortality outcomes is scarce.Methods:We conducted a comprehensive study involving 656 peritoneal dialysis(PD)patients over 12 years,with an average follow-up every 3 months.Dietary intake was meticulously recorded using a 3-day dietary record method,integrated with detailed health records and outcomes.We employed a 2-stage model to evaluate nonlinear relationships between dietary nutrients and mortality risk,accounting for various confounding factors.Findings:Our analysis revealed that 14 out of 26 nutritional elements lack guidelines for ESRD and PD patients,with 13 showing significant associations with mortality.For example,while guidelines suggest a dietary protein intake of 1.0 to 1.2 g/kg/d,our findings indicate an optimal range of 0.88 to 1.13 g/kg/d.Similarly,the recommended dietary energy intake of 25 to 35 kcal/kg/d was refined to 26 to 42 kcal/kg/d.We identified that 69% of dietary intake-outcome relationships are nonlinear,especially in patients with poor health status.Interpretations:Our study provides detailed dietary intake thresholds that correlate with improved prognosis in ESRD patients,enhancing current guidelines.The findings highlight the importance of personalized nutritional management and underscore the nonlinear nature of nutrient–disease relationships,particularly in severely ill patients.This approach can refine dietary recommendations and improve patient care in ESRD.
基金sponsored by Tsinghua-Toyota Joint Research Institute Inter-disciplinary Program.
文摘Background:The electrocardiogram(ECG)is a valuable,noninvasive tool for monitoring heart-related conditions,providing critical insights.However,the interpretation of ECG data alongside patient information demands substantial medical expertise and resources.While deep learning methods help streamline this process,they often fall short in integrating patient data with ECG readings and do not provide the nuanced clinical suggestions and insights necessary for accurate diagnosis.Methods:Although recent advancements in multi-modal large language modeling have propelled their application scope beyond the natural language processing domain,their applicability to ECG processing remains largely unexplored,partly due to the lack of text–ECG data.To this end,we develop ECG-Language Model(ECG-LM),the first multi-modal large language model able to process natural language and understand ECG signals.The model employs a specialized ECG encoder that transforms raw ECG signals into a high-dimensional feature space,which is then aligned with the textual feature space derived from the large language model.To address the scarcity of text–ECG data,we generated text–ECG pairs by leveraging detailed ECG pattern descriptions from medical guidelines,creating a robust dataset for pre-training ECG-LM.Additionally,we fine-tune ECG-LM with public clinical conversation datasets and build an additional supervised fine-tuning dataset based on real clinical data from the hospital,aiming to provide a more comprehensive and customized user experience.Results:ECG-LM outperforms existing few-shot and zero-shot solutions in cardiovascular disease detection across all 3 tasks(diagnostic,rhythm,and form)while also demonstrating strong potential in ECG-related question answering.Conclusions:The results across various tasks demonstrate that ECG-LM effectively captures the intricate features of ECGs,showcasing its versatility in applications such as disease prediction and advanced question answering.
基金the National Institutes of Health through the following grant:RM1HG009034(to B.A.M.).
文摘Background:Multimodal large language models(LLMs)have shown potential in various health-related fields.However,many healthcare studies have raised concerns about the reliability and biases of LLMs in healthcare applications.Methods:To explore the practical application of multimodal LLMs in skin disease identification,and to evaluate sex and age biases,we tested the performance of 2 popular multimodal LLMs,ChatGPT-4 and LLaVA-1.6,across diverse sex and age groups using a subset of a large dermatoscopic dataset containing around 10,000 images and 3 skin diseases(melanoma,melanocytic nevi,and benign keratosis-like lesions).Results:In comparison to 3 deep learning models(VGG16,ResNet50,and Model Derm)based on convolutional neural network(CNN),one vision transformer model(Swin-B),we found that ChatGPT-4 and LLaVA-1.6 demonstrated overall accuracies that were 3% and 23% higher(and F1-scores that were 4% and 34% higher),respectively,than the best performing CNN-based baseline while maintaining accuracies that were 38% and 26% lower(and F1-scores that were 38% and 19% lower),respectively,than Swin-B.Meanwhile,ChatGPT-4 is generally unbiased in identifying these skin diseases across sex and age groups,while LLaVA-1.6 is generally unbiased across age groups,in contrast to Swin-B,which is biased in identifying melanocytic nevi.Conclusions:This study suggests the usefulness and fairness of LLMs in dermatological applications,aiding physicians and practitioners with diagnostic recommendations and patient screening.To further verify and evaluate the reliability and fairness of LLMs in healthcare,experiments using larger and more diverse datasets need to be performed in the future.
文摘We value Cummins et al.’s study on an XGBoost model for in-hospital mortality prediction in intensive care unit(ICU)patients with acute ischemic stroke(AIS)[1].Its focus on model fairness across demographics and in-depth feature analysis gives useful references for clinical predictive analytics.To boost the model’s reproducibility,we suggest clarifying specific inclusion criteria and data preprocessing methods for the electronic ICU(eICU)database in future external validation.
