TY - GEN
T1 - Enhance the Focal Cortical Dysplasia Lesions in a MR-negative Image
T2 - 4th International Conference on Digital Signal Processing, ICDSP 2020
AU - Feng, Cuixia
AU - Zhao, Hulin
AU - Yang, Chen
AU - Lu, Miaomiao
AU - Tian, Maoyu
AU - Wen, Junhai
N1 - Publisher Copyright:
© 2020 ACM.
PY - 2020/6/19
Y1 - 2020/6/19
N2 - Focal cortical dysplasia (FCD) is one of the malformations of cortical development and one of the common causes of intractable epilepsy. The method of image post-processing can enhance the lesion area of the magnetic resonance (MR) image, help doctors to make a better diagnosis, alleviate doctors' dependence on the experience of their supervisors, and reduce the probability of misdiagnosis. In this study, we first calculate the cortical thickness of the T1 weighted images (T1WI) by the three-dimensional (3D) Laplace method and then calculate the highlighted area of the Fluid Attenuated Inversion Recovery (FLAIR) image by the K-means method. Finally, we combine the cortical thickness feature, K-means feature, and normalized positron emission tomography (PET) image to effectively enhance the lesion area of an epileptic patient with MR-negative images. Compared with the traditional methods of junction map and extension map in the morphometric analysis program (MAP) algorithm, this method successfully detected the lesion of FCD, displayed the epileptogenic lesion area through three-dimensional visualization, and showed less false-positive area.
AB - Focal cortical dysplasia (FCD) is one of the malformations of cortical development and one of the common causes of intractable epilepsy. The method of image post-processing can enhance the lesion area of the magnetic resonance (MR) image, help doctors to make a better diagnosis, alleviate doctors' dependence on the experience of their supervisors, and reduce the probability of misdiagnosis. In this study, we first calculate the cortical thickness of the T1 weighted images (T1WI) by the three-dimensional (3D) Laplace method and then calculate the highlighted area of the Fluid Attenuated Inversion Recovery (FLAIR) image by the K-means method. Finally, we combine the cortical thickness feature, K-means feature, and normalized positron emission tomography (PET) image to effectively enhance the lesion area of an epileptic patient with MR-negative images. Compared with the traditional methods of junction map and extension map in the morphometric analysis program (MAP) algorithm, this method successfully detected the lesion of FCD, displayed the epileptogenic lesion area through three-dimensional visualization, and showed less false-positive area.
KW - Epilepsy
KW - Focal cortical dysplasia
KW - MR images
UR - http://www.scopus.com/inward/record.url?scp=85091578164&partnerID=8YFLogxK
U2 - 10.1145/3408127.3408193
DO - 10.1145/3408127.3408193
M3 - Conference contribution
AN - SCOPUS:85091578164
T3 - ACM International Conference Proceeding Series
SP - 127
EP - 130
BT - ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings
PB - Association for Computing Machinery
Y2 - 19 June 2020 through 21 June 2020
ER -