A Remote Sensing Image Semantic Segmentation Method by Combining Deformable Convolution with Conditional Random Fields

Currently,deep convolutional neural networks have made great progress in the field of semantic segmentation.Because of the fixed convolution kernel geometry,standard convolution neural networks have been limited the ability to simulate geometric transformations.Therefore,a deformable convolution is introduced to enhance the adaptability of convolutional networks to spatial transformation.Considering that the deep convolutional neural networks cannot adequately segment the local objects at the output layer due to using the pooling layers in neural network architecture.To overcome this shortcoming,the rough prediction segmentation results of the neural network output layer will be processed by fully connected conditional random fields to improve the ability of image segmentation.The proposed method can easily be trained by end-to-end using standard backpropagation algorithms.Finally,the proposed method is tested on the ISPRS dataset.The results show that the proposed method can effectively overcome the influence of the complex structure of the segmentation object and obtain state-of-the-art accuracy on the ISPRS Vaihingen 2D semantic labeling dataset.

Design of curriculum for specialised midwive training based on investigation of needs of midwives in Beijing

Objective:This study aims to provide a basis for the curriculum provision,training form and training time of midwifery managers in Beijing and understand the needs of the midwives' participation in specialist certification training.Methods:Ten midwifery managers and midwives in Beijing were selected for a group interview.Based on the interview results,the researchers designed a questionnaire.As the midwives were from different hospital levels in Beijing,stratified random sampling method was adopted.A total of 137 people were surveyed through a web questionnaire.Results:In total,99.25% of the respondents believe that midwifery specialist training needs to be carried out in Beijing.Moreover,55% of them believe that half-time training is reasonable.The respondents believe that the proportion of theoretical and practice training should be 1∶2.In total,91.79% and 85.07% of the respondents believe that contents of midwifery research and midwifery management need to be increased,respectively.Conclusion:Midwives in Beijing need midwifery specialist training.Half-time certification training complies with the clinical needs.Midwifery specialist certification enhances the professional identity of midwives.

Power Grid Fault Diagnosis Based on Deep Pyramid Convolutional Neural Network

Existing power grid fault diagnosis methods relyon manual experience to design diagnosis models, lack theability to extract fault knowledge, and are difficult to adaptto complex and changeable engineering sites. Considering thissituation, this paper proposes a power grid fault diagnosismethod based on a deep pyramid convolutional neural networkfor the alarm information set. This approach uses the deepfeature extraction ability of the network to extract fault featureknowledge from alarm information texts and achieve end-to-endfault classification and fault device identification. First, a deeppyramid convolutional neural network model for extracting theoverall characteristics of fault events is constructed to identifyfault types. Second, a deep pyramidal convolutional neuralnetwork model for alarm information text is constructed, thetext description characteristics associated with alarm informationtexts are extracted, the key information corresponding to faultsin the alarm information set is identified, and suspicious faultydevices are selected. Then, a fault device identification strategythat integrates fault-type and time sequence priorities is proposedto identify faulty devices. Finally, the actual fault cases and thefault cases generated by the simulation are studied, and theresults verify the effectiveness and practicability of the methodpresented in this paper.

Factors that contribute to poor adherence to statin therapy in coronary heart disease patients from Chongqing and measures to improve their therapeutic outcomes

To assess the efficacy and short-term outcomes of adherence to statin therapy among coronary heart disease(CHD)patients following their hospital discharge,we enrolled 615 CHD patients who were prescribed statins from The First Affiliated Hospital of Chongqing Medical University in China between February 1st and October 31st of 2013.Statin adherence was evaluated by identifying the proportion of patients who remained adherent or became non-adherent to statin therapy over 4e8 months post-discharge from the hospital.The composite outcomes included all-cause mortality and re-hospitalization with cardiovascular disease.We found that 15.9%patients were non-adherent to their statin therapies and that coronary artery stenosis<75%(OR Z 3.433,95%CI:2.191e5.380,p<0.001)and adverse effects(OR Z 2.542,95%CI:1.327e4.869,p Z 0.005)both clearly contributed to poor adherence.The primary self-reported reasons for non-adherence included a lack of knowledge about the benefits of statin therapy(36.7%),the treatment being halted at the advice of their doctor(19.4%),and the difficulty in obtaining statins(12.2%).Non-adherence to statin therapy was significantly associated with an increased risk of cardiovascular events(OR Z 1.741,95%CI:1.035e2.929,p Z 0.037).In conclusion,CHD patients with moderate stenosis or adverse effects tended to have poor statin adherence,and this was significantly associated with increased cardiovascular events.We should strengthen education of the importance of statin therapy for both patients and doctors and facilitate the ability of patients to obtain their statin medication.