Jialei Wang
Jialei Wang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Guotang Jian
School of Computer Science and Technology, Zhejiang Gongshang University, Hangzhou, China
1 shared publication - He Huang
Tianjin University
1 shared publication - Hong Chen
School of Information Renmin University of China Beijing China
1 shared publication - Jianyuan Li
School of Computer and Computing Science, Zhejiang University City College, Hangzhou, China
1 shared publication - Jinhua Sheng
School of Computer Science, Hangzhou Dianzi University, 310018 Hangzhou, Zhejiang, China
1 shared publication - Lei Pan
Department of Plastic & Reconstructive Surgery, Center for Plastic & Reconstructive Surgery, Zhejiang Provincial Peoples Hospital, Hangzhou Medical College, Hangzhou, China
1 shared publication
Latest Publications
- 20 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 563-568
Functional Brain Subnetwork Analysis in Alzheimer's Disease and Mild Cognitive Impairment
In studies aimed at diagnosing Alzheimer's Disease (AD) and Mild Cognitive Impairment (MCI) using resting-state functional magnetic resonance imaging (fMRI), researchers typically construct whole-brain functional connectivity networks. We investigate whether localized regional networks, rather than the entire brain network, can have a positive impact on the classification of MCI and AD. We utilized T1-weighted data and resting-state fMRI data from 26 AD subjects and 26 age-matched individuals diagnosed with MCI. Employing the Human Connectome Project cortical parcellation, the brain cortex was divided into 360 brain regions. A whole-brain region-wise correlation matrix was established using correlation coefficients. By employing an improved maximal global cost efficiency method, the correlation matrix was binarized to create a 360-node undirected graph representing the functional brain network. Simultaneously, employing a higher-level brain network partitioning, the entire brain was divided into 12 groups, establishing 12 subnetworks. Among these, the Default, Cingulo-Opercular, Language, and Frontoparietal subnetworks demonstrated enhanced discriminative potential for distinguishing AD from MCI compared to the entire brain network. Furthermore, the Frontoparietal subnetwork of the AD group exhibited stronger connectivity than that of the MCI group, outperforming other networks. Our study findings suggest that complex network metrics of functional subnetworks could serve as imaging-based biomarkers for distinguishing AD and MCI.
- 20 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 563-568
Functional Brain Subnetwork Analysis in Alzheimer's Disease and Mild Cognitive Impairment
In studies aimed at diagnosing Alzheimer's Disease (AD) and Mild Cognitive Impairment (MCI) using resting-state functional magnetic resonance imaging (fMRI), researchers typically construct whole-brain functional connectivity networks. We investigate whether localized regional networks, rather than the entire brain network, can have a positive impact on the classification of MCI and AD. We utilized T1-weighted data and resting-state fMRI data from 26 AD subjects and 26 age-matched individuals diagnosed with MCI. Employing the Human Connectome Project cortical parcellation, the brain cortex was divided into 360 brain regions. A whole-brain region-wise correlation matrix was established using correlation coefficients. By employing an improved maximal global cost efficiency method, the correlation matrix was binarized to create a 360-node undirected graph representing the functional brain network. Simultaneously, employing a higher-level brain network partitioning, the entire brain was divided into 12 groups, establishing 12 subnetworks. Among these, the Default, Cingulo-Opercular, Language, and Frontoparietal subnetworks demonstrated enhanced discriminative potential for distinguishing AD from MCI compared to the entire brain network. Furthermore, the Frontoparietal subnetwork of the AD group exhibited stronger connectivity than that of the MCI group, outperforming other networks. Our study findings suggest that complex network metrics of functional subnetworks could serve as imaging-based biomarkers for distinguishing AD and MCI.
- 26 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (12) , 5827-5836
A Revised Approach to Orthodontic Treatment Monitoring From Oralscan Video
Research on orthodontic treatment monitoring from oralscan video is a new direction in dental digitalization. We designed an approach to reconstruct, segment, and estimate the pose of individual teeth to measure orthodontic treatment. To handle the semantic gap in heterogeneous data on the condition that they are combined linearly, we present a multimedia interaction network (MIN) to combine heterogeneous information in point cloud segmentation by extending the graph attention mechanism. Moreover, a structure-aware quadruple loss is designed to explore the relation between multiple and diverse unmatched points in point cloud registration. The performance of our approach is evaluated on multiple tooth registration datasets, and extensive experiments show that our approach improves the accuracy by a margin of 1.4% in the inlier ratio on the Aoralscan3 dataset when it is compared with prevailing approaches.
- 26 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (12) , 5827-5836
A Revised Approach to Orthodontic Treatment Monitoring From Oralscan Video
Research on orthodontic treatment monitoring from oralscan video is a new direction in dental digitalization. We designed an approach to reconstruct, segment, and estimate the pose of individual teeth to measure orthodontic treatment. To handle the semantic gap in heterogeneous data on the condition that they are combined linearly, we present a multimedia interaction network (MIN) to combine heterogeneous information in point cloud segmentation by extending the graph attention mechanism. Moreover, a structure-aware quadruple loss is designed to explore the relation between multiple and diverse unmatched points in point cloud registration. The performance of our approach is evaluated on multiple tooth registration datasets, and extensive experiments show that our approach improves the accuracy by a margin of 1.4% in the inlier ratio on the Aoralscan3 dataset when it is compared with prevailing approaches.