We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Automatic extraction and regional segmentation of brain and cerebrospinal fluid and AI-CAD.
- Authors
Shigeki Yamada; Hirotaka Ito; Hironori Matsumasa; Satoshi Ii; Shusaku Maeda; Naoki Takeishi; Tomohiro Otani; Motoki Tanikawa; Yoshiyuki Watanabe; Shigeo Wada; Marie Oshima; Mitsuhito Mase
- Abstract
Brain Subregion Analysis app can automatically divide the intracranial space into 21 brain subregions and 5 cerebrospinal fluid spaces using deep learning method within 1 min. We investigated the agerelated changes in the regional brain and cerebrospinal spaces. The cortical gray matter decreased linearly with age after 20s, while subcortical gray matter such as limbic system was maintained, and cerebral white matter increased slightly until the 40s, then rapidly decreased. Conversely, subarachnoid spaces increased linearly and ventricles increased after 60s. Second, we successfully developed an automatic segmentation of the intracranial cerebrospinal space, total ventricles, high-convexity part of the subarachnoid space, and Sylvian fissure and basal cistern using the 3D U-Net model. In addition, we successfully developed a highly accurate automatic detection of disproportionately enlarged subarachnoidspace hydrocephalus (DESH) including ventricular dilatation, tightened sulci in the high convexities, and Sylvian fissure dilatation, which were characteristic imaging findings in idiopathic normal-pressure hydrocephalus (iNPH).
- Publication
Transactions of Japanese Society for Medical & Biological Engineering, 2023, Vol 61, p319
- ISSN
1347-443X
- Publication type
Article