Related Experiment Video
Updated: Nov 7, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Fully Automated Segmentation of Head CT Neuroanatomy Using Deep Learning
Jason C Cai1, Zeynettin Akkus1, Kenneth A Philbrick1
1Departments of Radiology (J.C.C., K.A.P., S.H., P.R., G.M.C., D.C.V., Q.H., B.J.E.) and Cardiovascular Science (Z.A.), Mayo Clinic Rochester, 200 First St. SW, RO_PB_02_RIL, Rochester, MN 55905; Department of Radiology, Khon Kaen University, Khon Kaen, Thailand (A.B.); Department of Health Sciences Research, Mayo Clinic Florida, Jacksonville, Fla (A.D.W.); and Department of Internal Medicine, Ascension St. John Hospital, Detroit, Mich (A.Z.).
Purpose:
To develop a deep learning model that segments intracranial structures on head CT scans.
Materials And Methods:
In this retrospective study, a primary dataset containing 62 normal noncontrast head CT scans from 62 patients (mean age, 73 years; age range, 27-95 years) acquired between August and December 2018 was used for model development. Eleven intracranial structures were manually annotated on the axial oblique series. The dataset was split into 40 scans for training, 10 for validation, and 12 for testing. After initial training, eight model configurations were evaluated on the validation dataset and the highest performing model was evaluated on the test dataset. Interobserver variability was reported using multirater consensus labels obtained from the test dataset. To ensure that the model learned generalizable features, it was further evaluated on two secondary datasets containing 12 volumes with idiopathic normal pressure hydrocephalus (iNPH) and 30 normal volumes from a publicly available source. Statistical significance was determined using categorical linear regression with P < .05.
Results:
Overall Dice coefficient on the primary test dataset was 0.84 ± 0.05 (standard deviation). Performance ranged from 0.96 ± 0.01 (brainstem and cerebrum) to 0.74 ± 0.06 (internal capsule). Dice coefficients were comparable to expert annotations and exceeded those of existing segmentation methods. The model remained robust on external CT scans and scans demonstrating ventricular enlargement. The use of within-network normalization and class weighting facilitated learning of underrepresented classes.
Conclusion:
Automated segmentation of CT neuroanatomy is feasible with a high degree of accuracy. The model generalized to external CT scans as well as scans demonstrating iNPH.Supplemental material is available for this article.© RSNA, 2020.
