Related Experiment Video
Updated: Jun 25, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
42.4K
A deep learning algorithm for automated adrenal gland segmentation on non-contrast CT images
Fanxing Meng1, Tuo Zhang1, Yukun Pan1
1Department of Radiology, Central China Subcenter of National Center for Cardiovascular Diseases, Fuwai Central-China Cardiovascular Hospital, Central China Fuwai Hospital of Zhengzhou University, Zhengzhou, 450046, China.
BMC Medical Imaging
|May 1, 2025
Summary
A new deep learning model accurately segments adrenal glands on CT scans. This technology enables large-scale studies on adrenal gland volume changes with age, revealing a pattern of initial increase followed by a decrease.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence
Background:
- Adrenal glands are vital retroperitoneal organs, but standardized CT measurements are lacking.
- Clinical practice needs reliable reference standards for adrenal gland dimensions.
Purpose of the Study:
- To develop a deep learning (DL) model for automated adrenal gland segmentation on non-contrast CT.
- To investigate age-related volume changes in normal adrenal glands using DL segmentation.
Main Methods:
- Utilized nnU-Net for DL model training on 1301 CT scans with radiologist-defined ground truth.
- Assessed model performance using Dice Similarity Coefficient (DSC) and compared with inter-observer variability.
- Applied the validated model to segment adrenal glands in a large dataset of 2000 normal CT examinations.
Main Results:
- The DL model achieved high segmentation accuracy with median DSC scores of 0.899 (left) and 0.904 (right) on the test set.
- Automated segmentation performance was comparable to manual segmentation by radiologists (P=0.541).
- Analysis of 2000 normal adrenal glands revealed a biphasic age-related volume change: initial increase followed by a decrease.
Conclusions:
- The developed DL model provides accurate adrenal gland segmentation on CT.
- This automated approach facilitates large-scale analysis of adrenal gland volume variations across different age groups.
