This study proposes a convolutional neural network (CNN) model for accurate and reliable automated detection of intracranial aneurysms using 3D time-of-flight magnetic resonance angiography (TOF-MRA). The model utilizes 3D patches extracted along vessel skeletons, semantic segmentation with a 3D U-Net, and an auxiliary classifier to address data imbalance. The method achieved high accuracy (0.910 internally, 0.883 externally) and demonstrates potential for clinical application.
Publisher
Scientific Reports
Published On
Jul 11, 2023
Authors
Sungwon Ham, Jiyeon Seo, Jihye Yun, Yun Jung Bae, Tackeun Kim, Leonard Sunwoo, Sooyoung Yoo, Seung Chai, Jeong-Whun Kim, Namkug Kim
Tags
convolutional neural network
intracranial aneurysms
3D TOF-MRA
semantic segmentation
data imbalance
clinical application
machine learning
Related Publications
Explore these studies to deepen your understanding of the subject.
