Theoretical investigation of five-coordinated manganese(III) oxophlorin with different axial ligands at various spin states using different DFT methods

The Mn(III)-oxophlorin complexes with imidazole, pyridine and t-butylcyanide as axial ligands have been studied using B3LYP, Bv86p, and M06-2X methods. All of the possible optimized geometries are specified, while the M06-2X is employed. Results obtained show that the isomers of Mn(III)-oxophlorin with imidazole or pyridine are the most stable at quintet state, compared to singlet and triplet spin states. Besides, there are two and four [Formula: see text]-electrons on manganese in each of these complexes at triplet and quintet states, respectively. Also, Mn(III)-oxophlorin with t-butylcyanide as axial ligand is only stable at singlet state. Non-specific solvent effects show that dispersion and London forces have the basic role in stability of complexes in a solvent. Note that latter interactions can occur in medium with dielectric constant ([Formula: see text]) of [Formula: see text]8, such as [Formula: see text] for position of oxophlorin in heme oxygenase enzyme. NBO analysis show that there is no degeneracy between d orbitals of Mn in the five-coordinated Mn(III)-oxophlorin at singlet and triplet spin states, but two d orbitals of manganese are degenerated in latter complexes at quintet state. Such degeneracy of d orbitals is observed in a complex with square pyramid structure. Then five-coordinated Mn(III)-oxophlorin with imidazole or pyridine is the most stable at quintet spin state, because of its geometry corresponding to square pyramid configuration of atoms. Also, nonbounding interaction between Mn and the ring of oxophlorin or Mn and ligand are more effective in Mn(III)-oxophlorin with imidazole as axial ligand, compared to pyridine and t-butylcyanide.

Improved YOLO v5 with balanced feature pyramid and attention module for traffic sign detection

With the development of automatic driving technology, traffic sign detection has become a very important task. However, it is a challenging task because of the complex traffic sign scene and the small size of the target. In recent years, a number of convolutional neural network (CNN) based object detection methods have brought great progress to traffic sign detection. Considering the still high false detection rate, as well as the high time overhead and computational overhead, the effect is not satisfactory. Therefore, we employ lightweight network model YOLO v5 (You Only Look Once) as our work foundation. In this paper, we propose an improved YOLO v5 method by using balances feature pyramid structure and global context block to enhance the ability of feature fusion and feature extraction. To verify our proposed method, we have conducted a lot of comparative experiments on the challenging dataset Tsinghua-Tencent-100K (TT100K). The experimental results demonstrate that the [email protected] and [email protected]:0.95 are improved by 1.9% and 2.1%, respectively.

The Different Classification of Hospitals Impact on the Equity of Medical Outcomes of Patients in China

Background: In the context of Healthy China Strategy, the traditional idea that large hospitals provide better medical care is still deeply rooted in people's minds. The characteristics of the medical inverted pyramid structure—higher-level medical institutions are overwhelmed, while lower-level medical institutions are deserted—have existed for a long time. Because of the unreasonable allocation of resources, it is difficult to meet the healthcare needs of citizens and to establish a tiered delivery system. The objective of this paper is to study the effect of different classification of hospitals (COH) on the equity of Medical Outcomes of patients.Methods: The data used was from Urban Employee Basic Medical Insurance (UEBMI) enrolment records of Chengdu. We conducted a retrospective study and used Nested Multinomial Logit Model (NMNL) to estimate the effect of COH on the equity of Medical Outcomes.Results: COH had a significant effect on the equity of Medical Outcomes, but the effectiveness and direction of the hospital level on different outcomes were not consistent. The reimbursement rate, medical expenditures, gender, age, disease type and others factors were associated with the effect(p<0.01); length of stay has a limited effect on health outcomes. It was not the case that the longer the hospital stay was, the higher the quality of care would be. When COH was distinguished, there were significant differences in the effect of different levels of hospitals on the equity of Medical Outcomes. Horizontally speaking, hospitals of the same level had different effects on different the equity of Medical Outcomes(p<0.01). From a longitudinal perspective, different levels of hospitals had different effects on the equity of Medical Outcomes(p<0.01). It was not the case that the higher the level of hospital, the better the medical outcome. When hospital levels and disease types were distinguished, the effect of hospitals of different levels on the medical outcome of different disease types was significantly different.Conclusions: COH made a difference in the equity of Medical Outcomes. Hospitals of different levels should be reasonably selected according to disease types to achieve the optimal medical outcome. Therefore, China should promote the construction of a tiered delivery system.

Object Detection in Autonomous Driving Scenarios Based on an Improved Faster-RCNN

Object detection is one of the key algorithms in automatic driving systems. Aiming at addressing the problem of false detection and the missed detection of both small and occluded objects in automatic driving scenarios, an improved Faster-RCNN object detection algorithm is proposed. First, deformable convolution and a spatial attention mechanism are used to improve the ResNet-50 backbone network to enhance the feature extraction of small objects; then, an improved feature pyramid structure is introduced to reduce the loss of features in the fusion process. Three cascade detectors are introduced to solve the problem of IOU (Intersection-Over-Union) threshold mismatch, and side-aware boundary localization is applied for frame regression. Finally, Soft-NMS (Soft Non-maximum Suppression) is used to remove bounding boxes to obtain the best results. The experimental results show that the improved Faster-RCNN can better detect small objects and occluded objects, and its accuracy is 7.7% and 4.1% respectively higher than that of the baseline in the eight categories selected from the COCO2017 and BDD100k data sets.

Evaluation Technology of Classroom Students’ Learning State Based on Deep Learning

Facial features are an effective representation of students’ fatigue state, and the eye is more closely related to fatigue state. However, there are three main problems in the existing research: (1) the positioning of the eye is vulnerable to the external environment; (2) the ocular features need to be artificially defined and extracted for state judgment; and (3) although the student fatigue state detection based on convolutional neural network has a high accuracy, it is difficult to apply in the terminal side in real time. In view of the above problems, a method of student fatigue state judgment is proposed which combines face detection and lightweight depth learning technology. First, the AdaBoost algorithm is used to detect the human face from the input images, and the images marked with human face regions are saved to the local folder, which is used as the sample dataset of the open-close judgment part. Second, a novel reconstructed pyramid structure is proposed to improve the MobileNetV2-SSD to improve the accuracy of target detection. Then, the feature enhancement suppression mechanism based on SE-Net module is introduced to effectively improve the feature expression ability. The final experimental results show that, compared with the current commonly used target detection network, the proposed method has better classification ability for eye state and is improved in real-time performance and accuracy.

DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation

Pyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of DenseNet, this paper presents DCPNet, a densely connected pyramid network that fuses multi-scale features from multiple stages of the pyramid structure. DCPNet not only performs feature fusion between the adjacent stages, but also non-adjacent stages. To fuse these features, we design a simple and effective dense connection module (DCM). In addition, we offer a new consideration of the common upscale operation in our approach. We believe DCPNet offers a more efficient way to fuse features from multiple scales in a pyramid-like network. We perform extensive experiments using both outdoor and indoor benchmark datasets (i.e., the KITTI and the NYU Depth V2 datasets) and DCPNet achieves the state-of-the-art results.