Invention Grant
- Patent Title: Method for training model of deep neural network, and method and system using the model to determine whether there is abnormality in brain
-
Application No.: US17160168Application Date: 2021-01-27
-
Publication No.: US11379982B2Publication Date: 2022-07-05
- Inventor: Yu-Wei Chang , Albert Chihchieh Yang , Shih-Jen Tsai
- Applicant: NATIONAL YANG-MING UNIVERSITY
- Applicant Address: TW Taipei
- Assignee: NATIONAL YANG-MING UNIVERSITY
- Current Assignee: NATIONAL YANG-MING UNIVERSITY
- Current Assignee Address: TW Taipei
- Agency: Hamre, Schumann, Mueller & Larson, P.C.
- Priority: TW109104565 20200213
- Main IPC: G06K9/00
- IPC: G06K9/00 ; G06T7/00

Abstract:
A method for determining whether there is an abnormality in a brain includes steps of: receiving a to-be-determined (TBD) 3D brain image; implementing the calibration procedure on the TBD 3D brain image; generating a first TBD image of WM, a second TBD image of GM, and a third TBD image of CSF based on a multi-voxel pattern of the TBD 3D brain image; and inputting the first TBD image of WM into a first DNN model, the second TBD image of GM into a second DNN model, and the third TBD image of CSF into a third DNN model to determine whether there is an abnormality in the white matter, the gray matter and the cerebrospinal fluid, respectively.
Public/Granted literature
Information query