Large-scale identification of patients with cerebral aneurysms using natural language processing
Related Experiment Video
Updated: Mar 10, 2026
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
Published on: December 15, 2023
Victor M Castro1, Dmitriy Dligach1, Sean Finan1
1From Research Information Systems and Computing (V.M.C., V.G., S.M.), Partners Healthcare; Boston Children's Hospital Informatics Program (D.D., S.F., G.S.); Harvard Medical School (D.D., S.Y., A.C., M.A.-E.-B., N.A.S., S.M., S.T.W., R.D.); Department of Medicine (S.Y., S.T.W.), Department of Neurosurgery (A.C., M.A.-E.-B., R.D.), Division of Rheumatology, Immunology and Allergy (N.A.S.), and Channing Division of Network Medicine (S.T.W., R.D.), Brigham and Women's Hospital, Boston, MA; Center for Statistical Science (S.Y.), Tsinghua University, Beijing, China; Department of Neurology (S.M.), Massachusetts General Hospital; and Biostatistics (T.C.), Harvard School of Public Health, Boston, MA.
Objective:
To use natural language processing (NLP) in conjunction with the electronic medical record (EMR) to accurately identify patients with cerebral aneurysms and their matched controls.
Methods:
ICD-9 and Current Procedural Terminology codes were used to obtain an initial data mart of potential aneurysm patients from the EMR. NLP was then used to train a classification algorithm with .632 bootstrap cross-validation used for correction of overfitting bias. The classification rule was then applied to the full data mart. Additional validation was performed on 300 patients classified as having aneurysms. Controls were obtained by matching age, sex, race, and healthcare use.
Results:
We identified 55,675 patients of 4.2 million patients with ICD-9 and Current Procedural Terminology codes consistent with cerebral aneurysms. Of those, 16,823 patients had the term aneurysm occur near relevant anatomic terms. After training, a final algorithm consisting of 8 coded and 14 NLP variables was selected, yielding an overall area under the receiver-operating characteristic curve of 0.95. After the final algorithm was applied, 5,589 patients were classified as having aneurysms, and 54,952 controls were matched to those patients. The positive predictive value based on a validation cohort of 300 patients was 0.86.
Conclusions:
We harnessed the power of the EMR by applying NLP to obtain a large cohort of patients with intracranial aneurysms and their matched controls. Such algorithms can be generalized to other diseases for epidemiologic and genetic studies.
