Object detection plays an important role in autonomous driving systems. LiDAR is widely used in autonomous driving vehicles and robots as a sensor for environmental perception. Recently, the development of computational power and deep learning technology makes it possible to classify and locate objects from LiDAR point cloud in a single end-to-end learnable network. However, objects are sparsely distributed in large point cloud field, and are always been partly scanned by LiDAR, which pose a challenge for accurate and rapid object positioning and classification from the raw point cloud. In this paper, we introduce a new single-shot refinement neural network for fast and accurate 3D object detection from the raw LiDAR point cloud. Firstly, we exploit self-attention mechanism in main object detection branch to enhance object feature representation. Secondly, we apply deformable convolution for learning adaptive receptive fields to fully capture the features of rotating and partially visible objects. Thirdly, an object refinement branch is introduced to produce a finer regression of objects upon the primary estimation from the main detection branch. All proposed modules have been proven to effectively improve the accuracy of object detection. Our method is evaluated on KITTI 3D detection benchmark and achieves state-of-the-art results while maintains real-time efficiency. Furthermore, real-time test in autonomous driving vehicle demonstrates that our method is robust to 16 channels LiDAR and can meet the demands of accuracy, efficiency, and visibility of object detection in various urban scenarios.
Falcipain-3, the major cysteine hemoglobinase from the human malaria parasite Plasmodium falciparum, is critical for parasite development and is considered as a promising chemotherapeutic target. In order to understand the structure–activity correlation of falcipain-3 inhibitors, a set of ligand- and receptor-based 3D-QSAR models were developed in the present work employing comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) for 247 2-pyrimidinecarbonitrile derivatives. An optimum ligand-based CoMSIA model yielded a cross validation Q2 = 0.501, non-cross validation Rncv2 = 0.821 and predictive Rpred2 = 0.750. In addition, docking analysis and molecular dynamics simulation were applied to elucidate the probable binding modes of the ligand in the falcipain-3 binding pocket. Graphic representation of the results, as contoured 3D coefficient plots, also provides a clue to the reasonable modification of molecules. (1) Bulky substituents at the 3-position, and rings B and D increase the biological activity; (2) electrostatic groups at rings B, C and D are likely helpful to increase the falcipain-3 inhibition; (3) hydrophobic groups at rings B and D are favored; (4) Gly92, Ile94 and Thr95 which formed several H-bonds and a water-bridged H-bond are crucial for falcipain-3 inhibitors. This model, we hope, will be of help in designing and predicting novel falcipain-3 inhibitors.
Bone defects are one of the toughest challenges faced by orthopedic surgeons worldwide, especially at critical sizes, which are caused by severe trauma, malignancy, or congenital disease. The ideal bone tissue-engineered scaffold for bone regeneration is the one that has good osteoconductivity, osteoinductivity, pore structure, and antibacterial properties. Metal ions have been recognized in recent years to be essential regulators of bone metabolism, and they are widely used for bone tissue engineering. In particular, zinc ions are of interest because of their ideal biocompatibility, osteogenesis-promoting properties, and antibacterial properties. Moreover, the dual role of strontium (Sr) in promoting osteogenesis and inhibiting osteolysis provides academic support for Zn-Sr co-doped scaffolds. Based on true bone ceramics (TBC), Zn-Sr-sintered scaffolds with good pore structures were prepared using immersion-calcination. The biocompatibility, cell adhesion, osteogenic properties, and antibacterial activity of Zn-Sr-sintered TBC scaffolds in bone marrow mesenchymal stem cells (BMSCs) are superior to those of control TBC scaffolds. The Zn-Sr-sintered TBC scaffold was used to repair rat cranial defects. Its good in vivo repair performance was confirmed by osseointegration and inward bone growth compared with that of the control TBC scaffold. Zn0.25Sr0.20-TBC is an ideal material for bone repair because of its good biocompatibility and favorable in vitro osteogenic properties.
The cytotoxicity of chitosan/nano-hydroxyapatite(CS/nHA)composite scaffold material was evaluated through the determination of the chondrocytes cells relative proliferation.The chondrocytes of rats were isolated and multiplied in vitro,and then the chondrocytes were seeded onto the scaffold.The effect of the CS/nHA composite scaffold on the cell adhesion,proliferation,morphological changes were observed by the phase-contrast microscopy and scanning electron microscopy(SEM).The in vivo tissue response to CS/nHA was investigated by implanting into the subcutaneous and intramuscular sites.The results prove the CS/nHA composite scaffold has good biocompatibility and it will be a suitable scaffold for the osteochondral tissue engineering.
Melanin concentrating hormone receptor 1 (MCHR1), a crucial regulator of energy homeostasis involved in the control of feeding and energy metabolism, is a promising target for treatment of obesity. In the present work, the up-to-date largest set of 181 quinoline/quinazoline derivatives as MCHR1 antagonists was subjected to both ligand- and receptor-based three-dimensional quantitative structure–activity (3D-QSAR) analysis applying comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). The optimal predictable CoMSIA model exhibited significant validity with the cross-validated correlation coefficient (Q2) = 0.509, non-cross-validated correlation coefficient (R2ncv) = 0.841 and the predicted correlation coefficient (R2pred) = 0.745. In addition, docking studies and molecular dynamics (MD) simulations were carried out for further elucidation of the binding modes of MCHR1 antagonists. MD simulations in both water and lipid bilayer systems were performed. We hope that the obtained models and information may help to provide an insight into the interaction mechanism of MCHR1 antagonists and facilitate the design and optimization of novel antagonists as anti-obesity agents.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)