Academic Journal
Natural Language Processing Algorithm Accurately Classifies Diverticulitis-Related Complications and Predicts Long-Term Outcomes.
| Τίτλος: | Natural Language Processing Algorithm Accurately Classifies Diverticulitis-Related Complications and Predicts Long-Term Outcomes. |
|---|---|
| Συγγραφείς: | Ma W; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Wu Y; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Challa PK; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Sikavi D; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Downie JM; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Nguyen LH; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Raghu VK; Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Simon TG; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Khalili H; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Kambadakone AR; Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Ananthakrishnan AN; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts., Strate LL; Division of Gastroenterology and Hepatology, University of Wisconsin-Madison School of Medicine and Public Health, Madison, Wisconsin., Chan AT; Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts; Broad Institute of MIT and Harvard, Cambridge, Boston, Massachusetts; Department of Immunology and Infectious Diseases, Harvard T.H. Chan School of Public Health, Boston, Massachusetts. Electronic address: achan@mgh.harvard.edu. |
| Πηγή: | Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association [Clin Gastroenterol Hepatol] 2026 Oct; Vol. 24 (10), pp. 2860-2869. Date of Electronic Publication: 2026 Mar 23. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: W.B. Saunders for the American Gastroenterological Association Country of Publication: United States NLM ID: 101160775 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1542-7714 (Electronic) Linking ISSN: 15423565 NLM ISO Abbreviation: Clin Gastroenterol Hepatol Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: Philadelphia, PA : W.B. Saunders for the American Gastroenterological Association, 2003- |
| Ιατρικοί όροι (MeSH): | Diverticulitis*/complications , Diverticulitis*/classification , Diverticulitis*/diagnosis , Natural Language Processing* , Algorithms*, Humans ; Male ; Female ; Tomography, X-Ray Computed ; Aged ; Middle Aged ; Recurrence ; Aged, 80 and over |
| Περίληψη: | Background & Aims: Diagnostic codes lack the precision to identify specific complications of diverticulitis, limiting their utility in large-scale, real-world data. We developed a natural language processing (NLP) algorithm to classify diverticulitis and associated features using computed tomography (CT) reports. Methods: Using data from Mass General Brigham Research Patient Data Registry (1979-2024), we identified patients with a diagnosis code for diverticular disease (International Classification of Diseases [ICD]-9: 562; ICD-10: K57) and a prior abdominopelvic CT report. We developed and validated our NLP algorithm to detect diverticulitis and associated features. We subsequently investigated the associations between NLP-defined severity at first diagnosis (ie, uncomplicated, mild, severe, or chronic complications) and risk of severe diverticulitis recurrence using a Cox proportional hazards regression model. We assessed the predictive value of NLP-detected features using random forest models. Results: The NLP algorithm achieved positive and negative predictive values of 82.8% to 99.9%, outperforming both ICD codes and a generalist large language model. Among 16,349 patients with NLP-detected diverticulitis, 3192 developed severe recurrence over 76,736 person-years. Compared with uncomplicated diverticulitis, the multivariable-adjusted hazard ratio (HR) for severe recurrence was 1.39 (95% confidence interval [CI], 1.14-1.69) for mild complications, 3.02 (95% CI, 2.80-3.27) for severe complications, and 5.41 (95% CI, 4.78-6.13) for chronic complications. NLP-detected features significantly improved the prediction of severe diverticulitis recurrence compared with codified variables. Conclusions: Our NLP algorithm accurately classifies diverticulitis features, facilitating the construction of large and high-quality electronic health record-based cohorts. Severity at initial diagnosis predicts risk of severe recurrence, supporting the use of artificial intelligence for risk stratification and long-term management. (Copyright © 2026 American Gastroenterological Association Institute. Published by Elsevier Inc. All rights reserved.) |
| References: | Gastroenterology. 2025 Jan;168(1):111-120.e4. (PMID: 39304088) J Biomed Inform. 2001 Oct;34(5):301-10. (PMID: 12123149) J Am Med Inform Assoc. 2025 May 1;32(5):866-875. (PMID: 40085006) BMJ. 2015 Apr 24;350:h1885. (PMID: 25911572) Am Health Drug Benefits. 2019 Jun-Jul;12(4):188-197. (PMID: 31428236) Gastrointest Endosc Clin N Am. 2025 Apr;35(2):309-317. (PMID: 40021231) Radiology. 2017 Dec;285(3):850-858. (PMID: 28837412) Stud Health Technol Inform. 2013;192:677-81. (PMID: 23920642) Nat Rev Dis Primers. 2020 Mar 26;6(1):20. (PMID: 32218442) Gastroenterology. 2019 Apr;156(5):1282-1298.e1. (PMID: 30660732) Gastroenterology. 2025 May;168(5):1000-1024. (PMID: 39920892) Cell Mol Gastroenterol Hepatol. 2020;9(1):15-32. (PMID: 31351939) AMIA Jt Summits Transl Sci Proc. 2025 Jun 10;2025:614-623. (PMID: 40502255) BMC Med Res Methodol. 2011 Jun 26;11:99. (PMID: 21703013) Dis Colon Rectum. 2025 Jun 1;68(6):726-735. (PMID: 40272787) Int J Colorectal Dis. 2017 May;32(5):611-622. (PMID: 28110383) Dis Colon Rectum. 2018 Mar;61(3):382-389. (PMID: 29420430) Clin Gastroenterol Hepatol. 2014 Aug;12(8):1257-61. (PMID: 24858706) Aliment Pharmacol Ther. 2009 Dec 1;30(11-12):1171-82. (PMID: 19681811) Am J Surg. 2018 Sep;216(3):604-609. (PMID: 29454479) PLoS One. 2023 May 17;18(5):e0283553. (PMID: 37196047) J Am Med Inform Assoc. 2024 Sep 1;31(9):1994-2001. (PMID: 38613820) Br J Surg. 2018 Jul;105(8):971-979. (PMID: 29683483) Ann Surg. 2014 Sep;260(3):423-30; discussion 430-1. (PMID: 25115418) Med Decis Making. 2012 Jan-Feb;32(1):188-97. (PMID: 21393557) Dis Colon Rectum. 2025 May 1;68(5):562-571. (PMID: 39902821) Dis Colon Rectum. 2019 Aug;62(8):1005-1012. (PMID: 30664553) Inflamm Bowel Dis. 2023 Apr 3;29(4):503-510. (PMID: 35657296) Clin Exp Gastroenterol. 2010;3:139-42. (PMID: 21694857) Int J Colorectal Dis. 2018 Apr;33(4):431-440. (PMID: 29511842) JAMA Surg. 2025 Jun 1;160(6):615-622. (PMID: 40202724) Hepatol Commun. 2024 Mar 29;8(4):. (PMID: 38551386) Gastroenterology. 2009 Mar;136(3):741-54. (PMID: 19166855) Inflamm Bowel Dis. 2013 Jun;19(7):1411-20. (PMID: 23567779) Gastroenterology. 2021 Feb;160(3):906-911.e1. (PMID: 33279517) |
| Grant Information: | K01 DK135854 United States DK NIDDK NIH HHS; R01 DK101495 United States DK NIDDK NIH HHS |
| Contributed Indexing: | Keywords: Artificial Intelligence; Diagnostics; Diverticulitis; Natural Language Processing |
| Entry Date(s): | Date Created: 20260325 Date Completed: 20260922 Latest Revision: 20260925 |
| Update Code: | 20260925 |
| PubMed Central ID: | PMC13035340 |
| DOI: | 10.1016/j.cgh.2026.03.009 |
| PMID: | 41881290 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1542-7714 |
|---|---|
| DOI: | 10.1016/j.cgh.2026.03.009 |