WeiQi Wei
Professor
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (7)
Year:
2025
Topic modeling utilizes unsupervised machine learning to detect underlying themes within texts and has been deployed routinely to analyze social media for insights into healthcare issues. However, the inherent messiness of social media hinders the full realization of this technique’s potential. As such, we hypothesized that restricting medical concepts in social media texts to specific related semantic types and applying topic modeling to these concepts could be a feasible approach to overcome the challenge of traditional topic modeling for social media texts. Therefore, we developed a semantic-type-based topic modeling pipeline to discover self-reported health-related topics. This pipeline integrated semantic type information and Systematized Medical Nomenclature for Medicine (SNOMED) precoordinated expressions into a traditional topic modeling approach to enhance effectiveness in clustering meaningful, distinct topics. Using social media texts regarding statins for illustration, we evaluated the efficacy of this new approach and validated a newly identified topic using real-world clinical data. Based on expert evaluations, this approach resulted in more novel, distinguishable, and meaningful health-related topics compared to traditional topic modeling. In addition, our electronic health record validation for a newly identified topic in two real-world clinical databases indicated that statin users had a higher prevalence of depression or anxiety compared to matched non-users. Our results indicate that this new topic modeling pipeline can improve the extraction of themes from noisy online discussions, thereby contributing to deeper insights for healthcare research.
Year:
2025
We report the findings of a genome-wide association study (GWAS) meta-analysis of endometriosis across 14 biobanks worldwide, including 32% non-European patient participants, as part of the Global Biobank Meta-Analysis Initiative (GBMI). Out of 58 total loci (29 previously unreported), the largest meta-analysis accounted for 46 (20 previously unreported). We detected the first genome-wide significant loci (2p13.3 and 20q13.2) uniquely driven by the African-ancestry meta-analysis. Our imaging- and surgery-confirmed phenotypes yielded six additional previously unreported loci. Leveraging our large and diverse study population, we observed SNP heritability estimates of 9-13% for all ancestry groups, and 13 loci had at least one variant in the credible set after fine-mapping. Investigating the complex array of endometriosis comorbidities and risk factors revealed 135 genetically correlated phenotypes and 95 with evidence of vertical pleiotropy, including triglycerides and anxiety disorders. We prioritized 35 disease-relevant cellular contexts from the endometrial cell atlas and found 322 examples of differentially expressed genes in cells from donors with endometriosis. Further high-throughput multi-omic analyses implicated a total of 282 genes in endometriosis pathogenesis. Our diverse, comprehensive GWASs, with downstream analyses spanning molecular to phenotypic scales, provide detailed evidence for aspects of endometriosis including the role of immune cell types, Wnt signaling, and cellular proliferation. These interconnected pathways and risk factors underscore the complex, multi-faceted etiology of endometriosis, suggesting multiple targets for precise and effective therapeutic interventions.
Year:
2025
Long COVID affects a substantial proportion of the over 778 million individuals infected with SARS-CoV-2, yet predictive models remain limited in scope. While existing efforts, such as the National COVID Cohort Collaborative (N3C), have leveraged electronic health record (EHR) data for risk prediction, accumulating evidence points to additional contributions from social, behavioral, and genetic factors. Using a diverse cohort of SARS-CoV-2-infected individuals (n>17,200) from the NIH All of Us Research Program, we investigated whether integrating EHR data with survey-based and genomic information improves model performance. Our multi-scale approach outperformed EHR-only models original AUROC 0.736 (95% CI: 0.730, 0.741), achieving an AUROC of 0.748 (0.741,0.755). Among the top predictors, active-duty service status, self-reported fatigue, and chr19:4719431:G:A_A were among the most informative survey and genetic features. These findings highlight the importance of incorporating multi-scale data to improve risk stratification and inform personalized interventions for long COVID.
Year:
2025
Current studies regarding the secondary use of electronic health records (EHR) predominantly rely on domain expertise and existing medical knowledge. Though significant efforts have been devoted to investigating the application of machine learning algorithms in the EHR, efficient and powerful representation of patients is needed to unleash the potential of discovering new medical patterns underlying the EHR. Here, we present an unsupervised method for embedding high-dimensional EHR data at the patient level, aimed at characterizing patient heterogeneity in complex diseases and identifying new disease patterns associated with clinical outcome disparities. Inspired by the architecture of modern language models—specifically transformers with attention mechanisms, we use patient diagnosis and procedure codes as vocabularies and treat each patient as a sentence to perform the patient embedding. We applied this approach to 34,851 unique medical codes across 1,046,649 longitudinal patient events, including 102,739 patients from the electronic Medical Records and GEnomics (eMERGE) Network. The resulting patient vectors demonstrated excellent performance in predicting future disease events (median AUROC = 0.87 within one year) and bulk phenotyping (median AUROC = 0.84). We then illustrated the utility of these patient vectors in revealing heterogeneous comorbidity patterns, exemplified by disease subtypes in colorectal cancer and systemic lupus erythematosus, and capturing distinct longitudinal disease trajectories. External validation using EHR data from the University of Washington confirmed robust model performance, with median AUROCs of 0.83 and 0.84 for bulk phenotyping tasks and disease onset prediction, respectively. Importantly, the model reproduced the clustering results of disease subtypes identified in the eMERGE cohort and uncovered variations in overall mortality among these subtypes. Together, these results underscore the potential of representation learning in EHRs to enhance patient characterization and associated clinical outcomes, thereby advancing disease forecasting and facilitating personalized medicine.
Year:
2024
Endometriosis is a complex and heterogeneous condition affecting 10% of reproductive-age women, and yet, it often goes undiagnosed for several years. Limited observed heritability (7%) of large genetic association studies may be attributable to underlying heterogeneity of disease mechanisms. Therefore, we conducted this study to investigate genetic associations across sub-phenotypes of endometriosis. We performed unsupervised clustering of 4,078 women with endometriosis based on known endometriosis risk factors, symptoms, and concomitant conditions. The clusters were characterized by examining electronic health record (EHR) data and comprehensive chart reviews. We then performed genetic association for each cluster with 39 endometriosis-associated loci (Total endometriosis cases = 12,350). We identified five sub-phenotype clusters: (1) pain comorbidities, (2) uterine disorders, (3) pregnancy complications, (4) cardiometabolic comorbidities, and (5) EHR-asymptomatic. Bonferroni significant loci included PDLIM5 for the cluster 1, GREB1 for cluster 2, WNT4 for cluster 3, RNLS for cluster 4, and ABO for cluster 5. The difference in associations between the groups suggests complex and varied genetic mechanisms of endometriosis and its symptoms. This study enhances our understanding of the clinical patterns of endometriosis sub-phenotypes, showcasing the innovative approach employed to investigate this complex disease.
Year:
2024
Background Large population-based DNA biobanks linked to electronic health records (EHRs) may provide novel opportunities to identify genetic drivers of ARDS. Research Question Can we develop an EHR-based algorithm to identify ARDS in a biobank database, and can this validate a previously reported ARDS genetic risk factor? Study Design and Methods We analyzed two parallel genotyped cohorts: a prospective biomarker cohort of critically ill adults (VALID), and a retrospective cohort of hospitalized participants enrolled in a de-identified EHR biobank (BioVU). ARDS was identified by clinician-investigator review in VALID and an EHR algorithm in BioVU (EHR-ARDS). We tested the association between the MUC5B promoter polymorphism rs35705950 with development of ARDS, and assessed if age modified this genetic association in each cohort. Results In VALID, 2,795 patients were included, age was 55 [43, 66] (median [IQR]) years, and 718 (25.7%) developed ARDS. In BioVU, 9,025 hospitalized participants were included, age was 60 [48, 70] years, and 1,056 (11.7%) developed EHR-ARDS. We observed a significant age-related interaction effect on ARDS in VALID: among older patients, rs35705950 was associated with increased ARDS risk (OR: 1.44; 95%CI 1.08-1.92; p=0.012) whereas among younger patients this effect was absent (OR: 0.84; 95%CI: 0.62-1.14; p=0.26). In BioVU, rs35705950 was associated with increased risk for EHR-ARDS among all participants (OR: 1.20; 95%CI: 1.00-1.43, p=0.043) and this did not vary by age. The polymorphism was also associated worse oxygenation in mechanically ventilated BioVU participants, but had no association with oxygenation in VALID. Interpretation The MUC5B promoter polymorphism was associated with ARDS in two cohorts of at-risk adults. Although age-related effect modification was observed only in VALID, BioVU identified a consistent association between MUC5B and ARDS risk regardless of age, and a novel association with oxygenation impairment. Our study highlights the potential for EHR biobanks to enable precision-medicine ARDS studies.
Year:
2024
Collaborators (23)
Lorraine Ware
Vanderbilt University Medical Center
Christopher Chute
Bloomberg Distinguished Professor of Health Informatics
Johns Hopkins University School of Medicine
Stacey Missmer
Professor / Associate Chair for Research
University of Michigan
Vern Kerchberger
Assistant Professor of Medicine
Vanderbilt University Medical Center
Joel McNeil
Vanderbilt University Medical Center
Yuan Luo
Chief AI Officer and Associate Professor
Northwestern University
Digna R. Velez Edwards
Vanderbilt University Medical Center
Yuji Yamamoto
Osaka University Graduate School of Medicine
Dan M Roden
Professor
Vanderbilt University Medical Center
Alan Sanders
Clinical Associate Professor of Psychiatry
NorthShore University HealthSystem
Alyson Dickson
Vanderbilt University Medical Center
Marijana Vujkovic
Research Assistant Professor
University of Pennsylvania
Matthew Krantz
Assistant Professor of Medicine
Vanderbilt University Medical Center
Jonathan Shortt
Assistant Research Professor
University of Colorado Anschutz Medical Campus
Leah Kottyan
Associate Professor
Cincinnati Children's Hospital Medical Center
Todd Edwards
Assistant Professor
Vanderbilt University
Julie A. Bastarache
Professor
Vanderbilt University Medical Center
Angela Rogers
Professor
University of Texas Southwestern Medical Center
Yukinori Okada
University of Tokyo
Iftikhar Kullo
Professor
Medical College of Wisconsin
Paul Harris
Professor
Vanderbilt University Medical Center
Hiral Master
Asst Prof
University of North Carolina at Chapel Hill
QiPing Feng
Associate Professor
Vanderbilt University Medical Center
Make your next move.
Prepare your application
Review drafts before you send them.