School of Environment and Energy, Key Laboratory of Pollution Control and Ecosystem Restoration in Industry Clusters (Ministry of Education), Guangzhou, South China University of Technology, Guangzhou, Guangdong, 510006, PR China
Auto-generated profile, based on publicly available information
Tianbao Liu
School of Environment and Energy, Key Laboratory of Pollution Control and Ecosystem Restoration in Industry Clusters (Ministry of Education), Guangzhou, South China University of Technology, Guangzhou, Guangdong, 510006, PR China
A feasible and expandable two-step strategy for the preparation of reduced graphene oxide (rGO) supported NiO nanocrystals (rGO-NiO) was developed using graphene oxide (GO), nickel sulfate and ammonia as raw materials. The rGO-NiO composites were obtained with the benefit of electrostatic attraction between GO and Ni2+ (or Ni(OH)2) in a mild solution phase precipitation process and a sequent simultaneous thermal decomposition of Ni(OH)2 and reduction of GO in Ar. The influence of calcination temperature on the morphology, structure and electrochemical performance of the composites was investigated. The results show that the calcination temperature has less impact on the morphology of the composites. Homogenous cubic NiO nanocrystals (<10 nm) were well coated onto the surface of rGO both at 250 °C and 300 °C. However, the composites obtained at 300 °C (rGO-NiO 300) possess better crystallinity of NiO and a higher reduction degree of GO but lower specific surface area as well as pore volume compared with the composites achieved at 250 °C (rGO-NiO 250). rGO-NiO 250 is superior to rGO-NiO 300 in specific capacitance at all current densities (1044 vs. 422 F g−1 at 1 A g−1 and 600 vs. 360 F g−1 at 20 A g−1), but shows inferior rate capability (57.4% vs. 85.3% from 1 to 20 A g−1) and cycling stability (83% vs. 137% after 2000 cycles).
Automated multi-class segmentation of vertebrae and intervertebral discs for volumetric magnetic resonance image(MRI) can help the diagnosis and the treatment of many spinal diseases. However, existing methods based on fully convolutional network mostly stacked local convolution and pooling operations, and thus failed to capture the long-range dependencies for spine segmentation. In this paper, we propose a novel and computation-efficient residual U-Net with strip-pooling attention mechanism (SPA-ResUNet) for the multi-class segmentation of vertebrae and intervertebral discs. Due to the effectiveness of skip connections in U-Net, the strip pooling attention module is adopted in the skip connections to learn the long-range dependencies for improving the segmentation performance of the top structures and sacral crest of spine. Our SPA-ResUNet was implemented on the MRSpineSeg Challenge dataset. Experiments show that our SPA-ResUNet achieves impressive performance with mean Dice similarity coefficients of 88.14% and 87.88% in all 19 spinal structures on two test datasets, respectively. The visual segmentation results also show that our method achieves superior segmentation in top structures and sacral crest of spine. The proposed SPA-ResUNet has high potential in aiding the diagnosis and the treatment of spinal diseases.