medRxiv : the preprint server for health sciences最新文献

{"title":"Precision Estimates of Longitudinal Brain Aging Capture Unexpected Individual Differences in One Year.","authors":"Maxwell L Elliott, Jingnan L Du, Jared A Nielsen, Lindsay C Hanford, Pia Kivisakk, Steven E Arnold, Bradford C Dickerson, Ross W Mair, Mark C Eldaief, Randy L Buckner","doi":"10.1101/2025.02.21.25322553","DOIUrl":"10.1101/2025.02.21.25322553","url":null,"abstract":"<p><p>Longitudinal studies are required to measure individual differences in human brain aging, but they are difficult to estimate over short intervals because of measurement error. Using cluster scanning, an approach that reduces error by densely repeating rapid structural scans, we assessed brain aging in individuals across three longitudinal timepoints spaced across one year. Cluster scanning substantially improved the precision of individualized estimates, revealing previously undetectable individual differences in brain change. In just one year, expected differences in the rates of brain aging between younger and older individuals were evident, as were differences between cognitively unimpaired and impaired individuals. Each person's brain change trajectory was compared to modeled normative expectations from a large cohort of age-matched UK Biobank participants. Cognitively unimpaired older individuals variably revealed relative brain maintenance, unexpectedly rapid decline, and asymmetrical changes. These atypical brain aging trajectories were found across structures and verified in independent within-individual test and retest data. Cluster scanning promises to advance our understanding of the marked heterogeneity in brain aging by affording better short-term tracking of individual variability in structural change.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11888524/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143588668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dissecting the genetic complexity of myalgic encephalomyelitis/chronic fatigue syndrome via deep learning-powered genome analysis.","authors":"Sai Zhang, Fereshteh Jahanbani, Varuna Chander, Martin Kjellberg, Menghui Liu, Katherine Glass, David Iu, Faraz Ahmed, Han Li, Rajan Douglas Maynard, Tristan Chou, Johnathan Cooper-Knock, Martin Jinye Zhang, Durga Thota, Michael Zeineh, Jennifer Grenier, Andrew Grimson, Maureen Hanson, Michael Snyder","doi":"10.1101/2025.04.15.25325899","DOIUrl":"10.1101/2025.04.15.25325899","url":null,"abstract":"<p><p>Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex, heterogeneous, and systemic disease defined by a suite of symptoms, including unexplained persistent fatigue, post-exertional malaise (PEM), cognitive impairment, myalgia, orthostatic intolerance, and unrefreshing sleep. The disease mechanism of ME/CFS is unknown, with no effective curative treatments. In this study, we present a multi-site ME/CFS whole-genome analysis, which is powered by a novel deep learning framework, HEAL2. We show that HEAL2 not only has predictive value for ME/CFS based on personal rare variants, but also links genetic risk to various ME/CFS-associated symptoms. Model interpretation of HEAL2 identifies 115 ME/CFS-risk genes that exhibit significant intolerance to loss-of-function (LoF) mutations. Transcriptome and network analyses highlight the functional importance of these genes across a wide range of tissues and cell types, including the central nervous system (CNS) and immune cells. Patient-derived multi-omics data implicate reduced expression of ME/CFS risk genes within ME/CFS patients, including in the plasma proteome, and the transcriptomes of B and T cells, especially cytotoxic CD4 T cells, supporting their disease relevance. Pan-phenotype analysis of ME/CFS genes further reveals the genetic correlation between ME/CFS and other complex diseases and traits, including depression and long COVID-19. Overall, HEAL2 provides a candidate genetic-based diagnostic tool for ME/CFS, and our findings contribute to a comprehensive understanding of the genetic, molecular, and cellular basis of ME/CFS, yielding novel insights into therapeutic targets. Our deep learning model also offers a potent, broadly applicable framework for parallel rare variant analysis and genetic prediction for other complex diseases and traits.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12047926/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144034485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Status and Opportunities of Machine Learning Applications in Obstructive Sleep Apnea: A Narrative Review.","authors":"Matheus Lima Diniz Araujo, Trevor Winger, Samer Ghosn, Carl Saab, Jaideep Srivastava, Louis Kazaglis, Piyush Mathur, Reena Mehra","doi":"10.1101/2025.02.27.25322950","DOIUrl":"10.1101/2025.02.27.25322950","url":null,"abstract":"<p><strong>Background: </strong>Obstructive sleep apnea (OSA) is a prevalent and potentially severe sleep disorder characterized by repeated interruptions in breathing during sleep. Machine learning models have been increasingly applied in various aspects of OSA research, including diagnosis, treatment optimization, and developing biomarkers for endotypes and disease mechanisms.</p><p><strong>Objective: </strong>This narrative review evaluates the application of machine learning in OSA research, focusing on model performance, dataset characteristics, demographic representation, and validation strategies. We aim to identify trends and gaps to guide future research and improve clinical decision-making that leverages machine learning.</p><p><strong>Methods: </strong>This narrative review examines data extracted from 254 scientific publications published in the PubMed database between January 2018 and March 2023. Studies were categorized by machine learning applications, models, tasks, validation metrics, data sources, and demographics.</p><p><strong>Results: </strong>Our analysis revealed that most machine learning applications focused on OSA classification and diagnosis, utilizing various data sources such as polysomnography, electrocardiogram data, and wearable devices. We also found that study cohorts were predominantly overweight males, with an underrepresentation of women, younger obese adults, individuals over 60 years old, and diverse racial groups. Many studies had small sample sizes and limited use of robust model validation.</p><p><strong>Conclusion: </strong>Our findings highlight the need for more inclusive research approaches, starting with adequate data collection in terms of sample size and bias mitigation for better generalizability of machine learning models in OSA research. Addressing these demographic gaps and methodological opportunities is critical for ensuring more robust and equitable applications of artificial intelligence in healthcare.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11888534/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143589034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Unified Flexible Large Polysomnography Model for Sleep Staging and Mental Disorder Diagnosis.","authors":"Guifeng Deng, Mengfan Niu, Shuying Rao, Yuxi Luo, Jianjia Zhang, Junyi Xie, Zhenghe Yu, Wenjuan Liu, Junhang Zhang, Sha Zhao, Gang Pan, Xiaojing Li, Wei Deng, Wanjun Guo, Yaoyun Zhang, Tao Li, Haiteng Jiang","doi":"10.1101/2024.12.11.24318815","DOIUrl":"10.1101/2024.12.11.24318815","url":null,"abstract":"<p><p>Sleep disorders affect billions worldwide, yet clinical polysomnography (PSG) analysis remains hindered by labor-intensive manual scoring and limited generalizability of automated sleep staging tools across heterogeneous protocols. We present LPSGM, a large-scale PSG model designed to address two critical challenges in sleep medicine: cross-center generalization and adaptable diagnosis of neuropsychiatric disorders. Trained on 220,500 hours of multi-center PSG data (24,000 full-night recordings from 16 public datasets), LPSGM integrates domain-adaptive pre-training, flexible channel configurations, and a unified architecture to mitigate variability in equipment, montages, and populations during sleep staging while enabling downstream fine-tuning for mental disorder detection. In prospective validation, LPSGM achieves expert-level consensus in sleep staging (κ = 0.845 ± 0.066 vs. inter-expert κ = 0.850 ± 0.102) and matches the performance of fully supervised models on two independent private cohorts. When fine-tuned, it attains 88.01% accuracy in narcolepsy detection and 100% accuracy in identifying major depressive disorder (MDD), highlighting shared physiological biomarkers between sleep architecture and neuropsychiatric symptoms. By bridging automated sleep staging with real-world clinical deployment, LPSGM establishes a scalable, data-efficient framework for integrated sleep and mental health diagnostics. The code and pre-trained model are publicly available at https://github.com/Deng-GuiFeng/LPSGM to advance reproducibility and translational research in sleep medicine.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11661386/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142879328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mediating effect of pro-inflammatory cytokines in the association between depression, anxiety, and cardiometabolic disorders in an ethnically diverse community-dwelling middle-aged and older US population.","authors":"Asma Hallab","doi":"10.1101/2025.04.14.25325836","DOIUrl":"10.1101/2025.04.14.25325836","url":null,"abstract":"<p><strong>Introduction: </strong>Neuroinflammation is associated with depression and anxiety risk, both of which demonstrate a bilateral relationship with cardiometabolic disorders. Systemic inflammation is also commonly described in patients with cardiometabolic disorders. It is, thus, unclear whether pro-inflammatory cytokines might mediate the relationship between depression, anxiety, and cardiometabolic disorders, particularly in advanced ages.</p><p><strong>Methods: </strong>The multiethnic ≥ 50-year-old study population is a subset of the Health and Aging Brain Study: Health Disparities (HABS-HD). Adjusted logistic and linear regression models were applied to assess associations. Non-linearity was evaluated using restricted cubic splines. Statistical mediation analysis was used to determine the role of inflammation (Tumor Necrosis Factor-alpha (TNF-alpha) and Interleukin-6 (IL-6)). Models were corrected for multiple testing using the False Discovery Rate (FDR)-method.</p><p><strong>Results: </strong>In the 2,093 included cases, depression and/or anxiety were significantly associated with 62% higher odds of Cardiovascular Disorder (CVD) (OR=1.62 [95% CI: 1.22-2.15]), 54% of type 2 diabetes (T2DM) (OR=1.54 [95% CI: 1.29-1.85]), 26% of hypertension (OR=26% [95% CI: 1.07-1.48]), and 29% of obesity (OR=1.29 [95% CI: 1.11-1.51]). Only IL-6 showed a significant mediating role in the association of depression and/or anxiety with CVD (10%, <i>p-value _FDR</i> =0.016), T2DM (13%, <i>p-value _FDR</i> <0.001), hypertension (16%, <i>p-value _FDR</i> <0.001), and obesity (23%, <i>p-value _FDR</i> <0.001).</p><p><strong>Conclusions: </strong>Depression and anxiety were significantly associated with higher odds of major cardiometabolic disorders, and IL-6 partly mediated these associations. Clinical studies are needed to replicate the findings and specifically cluster high-risk profiles.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12047933/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144060094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automating Evaluation of AI Text Generation in Healthcare with a Large Language Model (LLM)-as-a-Judge.","authors":"Emma Croxford, Yanjun Gao, Elliot First, Nicholas Pellegrino, Miranda Schnier, John Caskey, Madeline Oguss, Graham Wills, Guanhua Chen, Dmitriy Dligach, Matthew M Churpek, Anoop Mayampurath, Frank Liao, Cherodeep Goswami, Karen K Wong, Brian W Patterson, Majid Afshar","doi":"10.1101/2025.04.22.25326219","DOIUrl":"10.1101/2025.04.22.25326219","url":null,"abstract":"<p><p>Electronic Health Records (EHRs) store vast amounts of clinical information that are difficult for healthcare providers to summarize and synthesize relevant details to their practice. To reduce cognitive load on providers, generative AI with Large Language Models have emerged to automatically summarize patient records into clear, actionable insights and offload the cognitive burden for providers. However, LLM summaries need to be precise and free from errors, making evaluations on the quality of the summaries necessary. While human experts are the gold standard for evaluations, their involvement is time-consuming and costly. Therefore, we introduce and validate an automated method for evaluating real-world EHR multi-document summaries using an LLM as the evaluator, referred to as LLM-as-a-Judge. Benchmarking against the validated Provider Documentation Summarization Quality Instrument (PDSQI)-9 for human evaluation, our LLM-as-a-Judge framework demonstrated strong inter-rater reliability with human evaluators. GPT-o3-mini achieved the highest intraclass correlation coefficient of 0.818 (95% CI 0.772, 0.854), with a median score difference of 0 from human evaluators, and completes evaluations in just 22 seconds. Overall, the reasoning models excelled in inter-rater reliability, particularly in evaluations that require advanced reasoning and domain expertise, outperforming non-reasoning models, those trained on the task, and multi-agent workflows. Cross-task validation on the Problem Summarization task similarly confirmed high reliability. By automating high-quality evaluations, medical LLM-as-a-Judge offers a scalable, efficient solution to rapidly identify accurate and safe AI-generated summaries in healthcare settings.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12045442/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144061278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accelerated Metabolomic Aging and Its Association with Social Determinants of Health in Multiple Sclerosis.","authors":"Fatemeh Siavoshi, Rezvan Noroozi, Gina Chang, Vinicius A Schoeps, Matthew D Smith, Farren B S Briggs, Jennifer S Graves, Emmanuelle Waubant, Ellen M Mowry, Peter A Calabresi, Pavan Bhargava, Kathryn C Fitzgerald","doi":"10.1101/2025.01.29.25321260","DOIUrl":"10.1101/2025.01.29.25321260","url":null,"abstract":"<p><strong>Objectives: </strong>Biological age may better capture differences in disease course among people with multiple sclerosis (PwMS) of identical chronological age. We investigated biological age acceleration through metabolomic age (mAge) in PwMS and its association with social determinants of health (SDoH) measured by area deprivation index (ADI).</p><p><strong>Methods: </strong>mAge was calculated for three cohorts: 323 PwMS and 66 healthy controls (HCs); 101 HCs and 71 DMT-naïve PwMS; and 64 HCs and 67 pediatric-onset MS/clinically isolated syndrome patients, using an aging clock derived from 11,977 healthy adults. mAge acceleration, the difference between mAge and chronological age, was compared between groups using generalized linear and mixed-effects models, and its association with ADI was assessed via linear regression.</p><p><strong>Results: </strong>Cross-sectionally, PwMS had higher age acceleration than HCs: 9.77 years in adult PwMS (95% CI:6.57-12.97, p=5.3e-09), 4.90 years in adult DMT-naïve PwMS (95% CI:0.85-9.01, p=0.02), and 6.98 years (95% CI:1.58-12.39, p=0.01) in pediatric-onset PwMS. Longitudinally, PwMS aged 1.19 mAge years per chronological year (95% CI:0.18, 2.20; p=0.02), faster than HCs. In PwMS, a 10-percentile increase in ADI was associated with a 0.63-year (95% CI:0.10-1.18; p=0.02) increase in age acceleration.</p><p><strong>Discussion: </strong>We demonstrated accelerated mAge in adult and pediatric-onset PwMS and its association with social disadvantage.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11838630/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143461287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Low TET1 Expression Levels in COPD Are Associated with Airway and Blood Neutrophilia.","authors":"Hong Ji, Xue Zhang, Angela L Linderholm, Maya Juarez, Michael Schivo, Brooks Kuhn, Richart W Harper, Amir A Zeki, Angela Haczku","doi":"10.1101/2025.04.15.25325889","DOIUrl":"10.1101/2025.04.15.25325889","url":null,"abstract":"<p><p>Epigenetic dysregulation, particularly DNA methylation variations, is implicated in the pathogenesis of chronic obstructive pulmonary disease (COPD). Ten-eleven translocation (TET) proteins (TET1, TET2, and TET3) regulate DNA methylation and gene transcription. Impaired TET1 expression was previously associated with airway inflammation and asthma. Here we investigated TET gene associations with COPD severity. We found that reduced TET1 expression in peripheral blood mononuclear cells was associated with higher sputum and blood neutrophil counts, decreased lung function and increased disease severity in patients. These findings support a potential protective role and warrant further mechanistic investigations into the actions of TET1 in COPD.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12051492/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144061201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lower Odor Identification in Subjective Cognitive Decline: A Meta-analysis.","authors":"Benoît Jobin, Coline Zigrand, Johannes Frasnelli, Benjamin Boller, Mark W Albers","doi":"10.1101/2025.04.15.25325887","DOIUrl":"10.1101/2025.04.15.25325887","url":null,"abstract":"<p><strong>Introduction: </strong>Odor identification correlates with Alzheimer's disease (AD) biomarkers, and its decline may emerge before measurable cognitive deficits-as early as the subjective cognitive decline (SCD) stage. We aimed to compare odor identification between SCD and cognitively normal (CN) stages and investigate whether cognitive differences moderate olfactory deficits.</p><p><strong>Methods: </strong>A systematic search of four databases identified studies assessing olfactory identification and cognitive screening in individuals aged 50+. A random-effects meta-analysis was performed on 11 studies (660 SCD, 574 CN).</p><p><strong>Results: </strong>Individuals with SCD exhibited lower olfactory identification scores compared to CN participants (SMD = -0.67, 95%CI [-1.31, -0.03], <i>p</i> = .04). Meta-regression revealed a negative association (β = -1.79, p = .02) between cognitive and olfactory differences, indicating that greater cognitive decline was not consistently associated with greater olfactory deficits, lower odor identification scores in SCD occurred despite minimal cognitive differences across groups.</p><p><strong>Discussion: </strong>Odor identification is lower in pre-MCI individuals reporting SCD. Olfactory decline may emerge independently prior to measurable cognitive decline, supporting the role of odor identification as a screen for AD.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12047905/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144063686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Demographic and Clinicopathologic Risk Factors for Colorectal Adenoma Recurrence: A Large-Scale Surveillance Cohort Study of 59,667 Adults.","authors":"Usman Ayub Awan, Qingyuan Song, Kristen K Ciombor, Adetunji T Toriol, Jungyoon Choi, Timothy Su, Xiao-Ou Shu, Kamran Idrees, Kay M Washington, Wei Zheng, Wanqing Wen, Zhijun Yin, Xingyi Guo","doi":"10.1101/2025.03.28.25324826","DOIUrl":"https://doi.org/10.1101/2025.03.28.25324826","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Current colorectal surveillance guidelines emphasize adenoma characteristics but overlook temporal, racial, and sex-based heterogeneity in recurrence risk- an gap that limits equitable and personalized care. To evaluate the associations of demographic factors, obesity, and adenoma features with recurrence risk over time in a large longitudinal surveillance cohort.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This retrospective cohort study included 59,667 adults who underwent their first colonoscopic polypectomy between January 1990 and July 2024 at a tertiary medical center. Median follow-up was 4 years. Demographic variables included race and ethnicity (non-Hispanic White [NHW], non-Hispanic Black [NHB], Hispanic, Asian or Pacific Islander [API]), sex, obesity (BMI &gt;30), family history of colorectal cancer (CRC) or polyps, and age at adenoma onset (&lt;50 vs ≥50 years). Adenoma features included histology, size, number, and dysplasia. The primary outcome was recurrence-free survival, defined as time from initial polypectomy to histologically confirmed recurrence. Cox proportional hazards models estimated associations adjusted for confounders, with stratified analyses over 5-, 10-, and &gt;10-year follow-up intervals.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Findings: &lt;/strong&gt;Among 59,667 patients, 17,596 (29.5%) experienced recurrence within 5 years, revealing substantial temporal heterogeneity. Early recurrence was associated with male sex (adjusted hazard ratio [aHR], 1.10; 95% CI, 1.06-1.14), obesity (aHR, 1.18; 95% CI, 1.13-1.23), early-onset adenomas (aHR, 1.17; 95% CI, 1.11-1.23), and family history of CRC (aHR, 1.24; 95% CI, 1.18-1.31). Compared with NHW patients, NHB individuals had lower early recurrence risk (aHR, 0.89; 95% CI, 0.83-0.96) but higher late recurrence (&gt;10 years; aHR, 1.26; 95% CI, 1.06-1.50). API patients had a similar shift, with lower early risk (aHR, 0.80; 95% CI, 0.67- 0.96) and elevated mid-term risk (5-10 years; aHR, 1.40; 95% CI, 1.08-1.81). High-grade dysplasia (aHR 2.86; 95% CI, 2.54-3.22) and villous histology (aHR 2.55; 95% CI, 2.31-2.81showed the largest effect sizes for early recurrence. Females had stronger associations with tubulovillous histology, mixed adenomas, and large lesions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Interpretation: &lt;/strong&gt;Temporal, demographic, and histologic differences in adenoma recurrence highlight the need for surveillance strategies that incorporate population- and time-specific risk profiles to enhance colorectal cancer prevention.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Funding: &lt;/strong&gt;This work was supported by the National Cancer Institute (Grant No. R37CA227130 to Xingyi Guo).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Research in context: &lt;/strong&gt;&lt;b&gt;Evidence before this study:&lt;/b&gt; We conducted a PubMed search for publications dated before June 2024 using combinations of keywords such as \"colonoscopic polypectomy,\" \"Demographic and Clinicopathologic Risk Factors,\" \"Vannderbilt,\" and \"electronic health records.\" We found no studies that comprehensively evaluat","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12060963/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144045700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}