Identification of an ANCA-associated vasculitis cohort using deep learning and electronic health records
Open accessPublished version of a preprint
Abstract
Background ANCA-associated vasculitis (AAV) is a rare but serious disease. Traditional case-identification methods using claims data can be time-intensive and may miss important subgroups. We hypothesized that a deep learning model analyzing electronic health records (EHR) can more accurately identify AAV cases.Methods We examined the Mass General Brigham (MGB) repository of clinical documentation from 12/1/1979 to 5/11/2021, using expert-curated keywords and ICD codes to identify a large cohort of potential AAV cases. Three labeled datasets (I, II, III) were created, each containing note sections. We trained and evaluated a range of machine learning and deep learning algorithms for note-level classification, using metrics like positive predictive value (PPV), sensitivity, F-score, area under the receiver operating characteristic curve (AUROC), and area under the precision and recall curve (AUPRC). The hierarchical attention network (HAN) was further evaluated for its ability to classify AAV cases at the patient-level, compared with rule-based algorithms in 2000 randomly chosen samples.Results Datasets I, II, and III comprised 6000, 3008, and 7500 note sections, respectively. HAN achieved the highest AUROC in all three datasets, with scores of 0.983, 0.991, and 0.991. The deep learning approach also had among the highest PPVs across the three datasets (0.941, 0.954, and 0.800, respectively). In a test cohort of 2000 cases, the HAN model achieved a PPV of 0.262 and an estimated sensitivity of 0.975. Compared to the best rule-based algorithm, HAN identified six additional AAV cases, representing 13% of the total.Conclusion The deep learning model effectively classifies clinical note sections for AAV diagnosis. Its application to EHR notes can potentially uncover additional cases missed by traditional rule-based methods.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
A hierarchical attention network model successfully predicted anti-neutrophil cytoplasmic antibody-associated vasculitis with AUC >0.9 from clinical records.
“whereas a hierarchical attention network model successfully predicted anti-neutrophil cytoplasmic antibody—associated vasculitis with AUC >0.9 from clinical records (Wang et al, 2025).”
From Artificial intelligence–enabled precision medicine for inflammatory skin diseases (2026) · cites this paper as Wang et al, 2025
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · see citing works
Watts, R. A., Lane, S., Hanslik, T., Hauser, T., Hellmich, B., Koldingsnes, W., Mahr, A., Segelmark, M., Tervaert, J. W. C., & Scott, D. G. I. (2006) · Annals of the Rheumatic Diseases · doi:10.1136/ard.2006.054593 · Cited by 1,357 · see citing works
Kitching, A. R., Anders, H., Basu, N., Brouwer, E., Gordon, J., Jayne, D. R. W., Kullman, J. A., Lyons, P., Merkel, P. A., Savage, C. O. S., et al. (2020) · Nature Reviews Disease Primers · doi:10.1038/s41572-020-0204-y · Cited by 969 · see citing works
BioWordVec, improving biomedical word embeddings with subword information and MeSH
Zhang, Y., Chen, Q., Yang, Z., Lin, H., & Lu, Z. (2019) · Scientific Data · doi:10.1038/s41597-019-0055-0 · Cited by 455 · see citing works
Mortality in ANCA-associated vasculitis: ameta-analysis of observational studies
Tan, J., Dehghan, N., Chen, W., Xie, H. W., Esdaile, J. M., & Aviña-Zubieta, J. A. (2017) · Annals of the Rheumatic Diseases · doi:10.1136/annrheumdis-2016-210942 · Cited by 182 · see citing works
Shao, Y., Zeng, Q. T., Chen, K. K., Shutes‐David, A., Thielke, S. M., & Tsuang, D. W. (2019) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-019-0846-4 · Cited by 83 · see citing works
Zhou, L., Plasek, J. M., Mahoney, L. M., Karipineni, N., CHANG, F. Y., Xuemin, Y., Chang, F., Dimaggio, D., Goldman, D. S., & Rocha, R. A. (2011) · PubMed Central · Cited by 72 · see citing works
Self-citationDu, X., Novoa-Laurentiev, J., Plasek, J. M., Chuang, Y.-W., Wang, L., Marshall, G. A., Mueller, S. K., Chang, F., Datta, S., Paek, H., et al. (2024) · eBioMedicine · doi:10.1016/j.ebiom.2024.105401 · Cited by 69 · see citing works
Self-citationDiagnostic delays in vasculitis and factors associated with time to diagnosis
Sreih, A. G., Cronin, K., Shaw, D. G., Young, K., Burroughs, C., Kullman, J. A., Machireddy, K., McAlear, C. A., & Merkel, P. A. (2021) · Orphanet Journal of Rare Diseases · doi:10.1186/s13023-021-01794-5 · Cited by 63 · see citing works
Sreih, A. G., Annapureddy, N., Springer, J. M., Casey, G. C., Byram, K., Cruz, A., Estephan, M., Frangiosa, V., George, M., Liu, M., et al. (2016) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.4116 · Cited by 45 · see citing works
Yang, J., Wang, L., Phadke, N. A., Wickner, P. G., Mancini, C. M., Blumenthal, K. G., & Zhou, L. (2020) · JAMA Network Open · doi:10.1001/jamanetworkopen.2020.22836 · Cited by 42 · see citing works
Self-citationWang, L., Novoa-Laurentiev, J., Yang, J., Lo, Y.-C., Amariglio, R. E., Blacker, D., Sperling, R. A., Marshall, G. A., & Zhou, L. (2021) · JAMA Network Open · doi:10.1001/jamanetworkopen.2021.35174 · Cited by 38 · see citing works
Self-citationWallace, Z. S., Fu, X., Cook, C., Ahola, C., Williams, Z. K., Doliner, B., Hanberg, J. S., Stone, J. H., Zhang, Y., & Choi, H. K. (2023) · Arthritis & Rheumatology · doi:10.1002/art.42515 · Cited by 18 · see citing works
Self-citationHua, Y., Wang, L., Nguyen, V., Rieu‐Werden, M. L., McDowell, A., Bates, D. W., Foer, D., & Zhou, L. (2023) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2023.104507 · Cited by 18 · see citing works
Self-citationWang, L.-Q., Miloslavsky, E., Stone, J., Choi, H. K., Zhou, L., & Wallace, Z. (2020) · Seminars in Arthritis & Rheumatism · doi:10.1016/j.semarthrit.2020.10.012 · Cited by 11 · see citing works
Self-citationMettler, C., Bouam, S., Liva-Yonnet, S., Pugnet, G., & Terrier, B. (2022) · European Journal of Internal Medicine · doi:10.1016/j.ejim.2022.05.017 · Cited by 8 · see citing works
Flagged first, then by citations
Research works printed but not linked: 2
- [11]Kitching AR, Anders HJ, Basu N, Brouwer E, Gordon J, Jayne DR, ANCA-ssociated vasculitis. Nat Rev Dis Primers. 2020; 6(1): 71. 32855422
- [14]Devlin J, Chang M-W, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:181004805. 2018.
Cited by
Sorted by citations
Marvisi, C., Ricordi, C., Catellani, C., Croci, S., Muratore, F., & Salvarani, C. (2025) · Frontiers in Immunology · doi:10.3389/fimmu.2025.1616837 · Cited by 7 · see citing works
Artificial intelligence–enabled precision medicine for inflammatory skin diseases
Tang, A. S., Wei, M. L., Haemel, A., La, C., Sirota, M., & Lee, E. Y. (2026) · Journal of Investigative Dermatology · doi:10.1016/j.jid.2025.10.596 · Cited by 6 · see citing works
How We Treat ANCA-Associated Vasculitis: A Focus on the Maintenance Therapy
Roccatello, D., Fenoglio, R., Simone, E. D., & Sciascia, S. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14010208 · Cited by 4 · see citing works
Fijołek, J., & Śniady, A. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14134631 · Cited by 3 · see citing works
Colina, M., & Diversi, R. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15176864
- Authors
- Wang, L., Novoa-Laurentiev, J., Cook, C., Srivatsan, S., Hua, Y., Yang, J., Miloslavsky, E., Choi, H. K., Zhou, L., & Wallace, Z. S.
- Affiliations
- Brigham and Women's Hospital (US)
- Harvard University (US)
- Massachusetts General Hospital (US)
- Corresponding author
- Wang, Liqin
- ORCID
- Venue
- International Journal of Medical Informatics
- Publisher
- Elsevier BV
- Published
- 2025-04
- Volume
- 196
- Pages
- 105797
- ISSN
- 1386-5056 (ISSN-L)
- PMC
- PMC11213085
- OpenAlex
- W4406580675
- Cited by
- 5
- References
- 16
- Citations by year
- 2026: 2 · 2025: 3
- FWCI
- 2.00
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- International Journal of Medical Informatics (publishedVersion)
- medRxiv (acceptedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- PubMed Central (submittedVersion, cc-by-nc-nd)
- Funders
- National Institutes of Health (K23AR073334, R03AR078938, L30AR070520, K99AG075190)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Medicine › Pulmonary and Respiratory Medicine › Vasculitis and related conditions
- Health Sciences › Nursing › Issues, ethics and legal aspects › Nursing Diagnosis and Documentation
- Keywords
- ANCA-associated vasculitis, case finding, electronic health records, hierarchical attention network, deep learning, rule-based algorithm, clinical documentation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
