Department of Automation, Key Laboratory of System Control and Information Processing of Ministry of Education, Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education, and Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai Jiao Tong University, Shanghai, China
Auto-generated profile, based on publicly available information
Huixin Zhang
Department of Automation, Key Laboratory of System Control and Information Processing of Ministry of Education, Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education, and Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai Jiao Tong University, Shanghai, China
The semantic segmentation of point clouds is an important part of the environment perception for robots. However, it is difficult to directly adopt the traditional 3-D convolution kernel to extract features from raw 3-D point clouds because of the unstructured property of point clouds. In this article, a spherical interpolated convolution operator is proposed to replace the traditional grid-shaped 3-D convolution operator. In addition, this article analyzes the defect of point cloud interpolation methods based on the distance as the interpolation weight and proposes the self-learned distance–feature density by combining the distance and the feature correlation. The proposed method makes the feature extraction of the spherical interpolated convolution network more rational and effective. The effectiveness of the proposed network is demonstrated on the 3-D semantic segmentation task of point clouds. Experiments show that the proposed method achieves good performance on the ScanNet dataset and Paris-Lille-3D dataset. The comparison experiments with the traditional grid-shaped 3-D convolution operator demonstrated that the newly proposed feature extraction operator improves the accuracy of the network and reduces the parameters of the network. The source codes will be released on https://github.com/IRMVLab/SIConv .
To explore the small-world properties of brain functional networks in patients with obstructive sleep apnea-hypopnea syndrome (OSAHS) to aid diagnosis. A total of 29 OSAHS patients and 26 matched healthy volunteers were scanned with blood oxygen level-dependent functional magnetic resonance imaging (BOLD-fMRI) separately, and the whole brain was divided into 90 districts via automated anatomical labeling. The matrix Z was then built through a Fisher Z transformation. Two-sample t tests were applied to evaluate the changes in small-world properties in OSAHS patients compared to the control group. The properties included Eglobal, Elocal, and small-world parameters Lp, Cp, γ, λ, and σ. Both groups satisfied the small-world properties (σ > 1) within the sparsity range of 0.1-0.2. However, compared with the control group, the OSAHS group performed significantly lower in Cp, Elocal, and Eglobal (p < 0.05) and higher in Lp (p < 0.05). The γ, σ, and λ values were not significantly different between the two groups. Both healthy and OSAHS patients exhibited small-world properties in functional networks, but a subset of these small-world properties in OSAHS patients performed differently. These changes will not only provide a new perspective for pathophysiological mechanisms of OSAHS but will also help in understanding the disease in terms of whole-brain functional networks.