Deep neural network-based semantic segmentation generally requires large-scale cost extensive annotations for training to obtain better performance. To avoid pixel-wise segmentation annotations that ...are needed for most methods, recently some researchers attempted to use object-level labels (e.g., bounding boxes) or image-level labels (e.g., image categories). In this paper, we propose a novel recursive coarse-to-fine semantic segmentation framework based on only image-level category labels. For each image, an initial coarse mask is first generated by a convolutional neural network-based unsupervised foreground segmentation model and then is enhanced by a graph model. The enhanced coarse mask is fed to a fully convolutional neural network to be recursively refined. Unlike the existing image-level label-based semantic segmentation methods, which require labeling of all categories for images that contain multiple types of objects, our framework only needs one label for each image and can handle images that contain multi-category objects. Only trained on ImageNet, our framework achieves comparable performance on the PASCAL VOC dataset with other image-level label-based state-of-the-art methods of semantic segmentation. Furthermore, our framework can be easily extended to foreground object segmentation task and achieves comparable performance with the state-of-the-art supervised methods on the Internet object dataset.
Since its recognition in December 2019, covid-19 has rapidly spread globally causing a pandemic. Pre-existing comorbidities such as hypertension, diabetes, and cardiovascular disease are associated ...with a greater severity and higher fatality rate of covid-19. Furthermore, COVID-19 contributes to cardiovascular complications, including acute myocardial injury as a result of acute coronary syndrome, myocarditis, stress-cardiomyopathy, arrhythmias, cardiogenic shock, and cardiac arrest. The cardiovascular interactions of COVID-19 have similarities to that of severe acute respiratory syndrome, Middle East respiratory syndrome and influenza. Specific cardiovascular considerations are also necessary in supportive treatment with anticoagulation, the continued use of renin-angiotensin-aldosterone system inhibitors, arrhythmia monitoring, immunosuppression or modulation, and mechanical circulatory support.
High-performance flexible strain sensors are playing an increasingly important role in wearable electronics, such as human motion detection and health monitoring, with broad application prospects. ...This study developed a flexible resistance strain sensor with a porous structure composed of carbon black and multi-walled carbon nanotubes. A simple and low-cost spraying method for the surface of a porous polydimethylsiloxane substrate was used to form a layer of synergized conductive networks built by carbon black and multi-walled carbon nanotubes. By combining the advantages of the synergetic effects of mixed carbon black and carbon nanotubes and their porous polydimethylsiloxane structure, the performance of the sensor was improved. The results show that the sensor has a high sensitivity (GF) (up to 61.82), a wide strain range (0%-130%), a good linearity, and a high stability. Based on the excellent performance of the sensor, the flexible strain designed sensor was installed successfully on different joints of the human body, allowing for the monitoring of human movement and human respiratory changes. These results indicate that the sensor has promising potential for applications in human motion monitoring and physiological activity monitoring.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Human health is adversely affected by potentially toxic elements (PTEs) in the topsoil, entering the bodies via inhalation, ingestion, and dermal contact. To visualize human health risks, we ...investigated five PTEs (Cd, As, Pb, Hg, and Cr) in 72 farmland topsoil samples from a town in Chongqing City, southwest China. Based on the human health risk assessment model, sequential indicator simulation (SIS) and the positive matrix factorization model (PMF) were used to construct the spatial health risks and to analyze the sources of PTEs; finally, health risks were combined with the source by ArcGIS. Based on our results, the use of SIS is feasible for the prediction of the spatial distribution of PTEs. Among the risks, the non-cancer risk of As for children most likely exceeded the accepted level in some areas, making As a priority pollutant. Although the health risks of soil Cd were acceptable in the region, the spatial probability distribution of Cd> 0.3 mg/kg represents a threat as Cd enters the human food chain. Even if the industrial discharge was the lowest individual contributor (29.33%), due to the impact of industrial discharge, the total non-cancer risk with a high probability (>0.85) for children still exceeded the accepted level in the northwestern area, which should be regarded as the priority pollution source. The combined method was useful to reduce efforts in environmental management, thus providing a basis for soil remediation and pollution source control.
•A combined method was proposed to divide soil PTE health risk area.•Cd and As were priority pollutants in farmland soil of Youxi town.•The cement plants pose potential health risks to the surrounding population, especially children.•A hierarchical management strategy was proposed based on probability risk map.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Background Cardiac right ventricular remodeling plays a substantial role in pathogenesis, progression, and prognosis of pulmonary hypertension. Cardiac magnetic resonance is considered an excellent ...tool for evaluation of right ventricle. However, value of right ventricular remodeling parameters derived from cardiac magnetic resonance in predicting adverse events is controversial. Methods The Pubmed (MEDLINE), Embase, Cochrane Library, Web of Science, China National Knowledge Infrastructure platform (CNKI), China Science and Technology Journal Database (VIP), and Wanfang databases were systematically searched until November 2019. Studies reporting hazard ratios (HRs) for all-cause death and composite end point of pulmonary hypertension were included. Univariate HRs were extracted from the included studies to calculate pooled HRs of each right ventricular remodeling parameter. Results Eight studies with 1120 patients examining all-cause death (female: 44%-92%, age: 40-67 years old, follow-up time: 27-48 months) and 10 studies with 604 patients examining composite end point (female: 60%-83%, age: 29-57 years old, follow-up time: 10-68 months) met the criteria. Right ventricular ejection fraction was the only parameter which could predict both all-cause death (pooled HR=0.95;
=0.014) and composite end point (pooled HR=0.95;
<0.001), although right ventricular end-diastolic volume index (pooled HR=1.01;
<0.001), right ventricular end-systolic volume index (pooled HR=1.01,
=0.045), and right ventricular mass index (pooled HR=1.03,
=0.032) only predicted composite outcome. Similar results were observed when we conducted the meta-analysis among patients with World Health Organization type I of pulmonary hypertension. Conclusions Cardiac magnetic resonance-derived right ventricular remodeling parameters have independent prognostic value for all-cause death and composite end point of patients with pulmonary hypertension. Right ventricular ejection fraction was the strongest prognostic factor among all the right ventricular remodeling parameters. Right ventricular mass index, right ventricular end-diastolic volume index, and right ventricular end-systolic volume index also demonstrated prognostic value.
With the accelerated ageing of the population in China, the health problems of elderly people have attracted much attention. Although religious belief has been shown to be a key way to improve the ...health of elderly people in various studies, little is known about the causal relationship between these variables in China. This paper explores the effect of religious belief on the health of elderly people in China, which will provide an important reference for China to achieve healthy ageing.
Balanced panel data collected between 2012 and 2016 from the China Family Panel Studies (CFPS) were used. Health was assessed using self-rated health, and religious belief was measured by whether the respondents believed in a religion. The DID+PSM method was employed to solve the endogeneity problem caused by self-selection and omitted variables. In addition, the CESD score (replacing self-rated health) and different matching methods (the method of PSM after DID method) were used to perform the robustness test.
The results show that religious belief has no significant effect on the health of elderly people. With the application of different matching methods (one-to-one matching, K-nearest neighbour matching, radius matching and kernel matching) and replacing the health indicator (the CESD score) with the above matching methods, the results are still robust.
In China, religious belief plays a limited role in promoting "healthy ageing", and it is difficult to improve the health of elderly people only via religious belief. Therefore, except for focusing on the guidance of religion with regard to healthy lifestyles, multiple measures need to be taken to improve the health of elderly people.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Acute myocardial infarction (AMI) is a serious cardiovascular disease, followed by a high readmission rate within 30-days of discharge. Accurate prediction of AMI readmission is a crucial way to ...identify the high-risk group and optimize the distribution of medical resources.
In this study, we propose a stacking-based model to predict the risk of 30-day unplanned all-cause hospital readmissions for AMI patients based on clinical data. Firstly, we conducted an under-sampling method of neighborhood cleaning rule (NCR) to alleviate the class imbalance and then utilized a feature selection method of SelectFromModel (SFM) to select effective features. Secondly, we adopted a self-adaptive approach to select base classifiers from eight candidate models according to their performances in datasets. Finally, we constructed a three-layer stacking model in which layer 1 and layer 2 were base-layer and level 3 was meta-layer. The predictions of the base-layer were used to train the meta-layer in order to make the final forecast.
The results show that the proposed model exhibits the highest AUC (0.720), which is higher than that of decision tree (0.681), support vector machine (0.707), random forest (0.701), extra trees (0.709), adaBoost (0.702), bootstrap aggregating (0.704), gradient boosting decision tree (0.710) and extreme gradient enhancement (0.713).
It is evident that our model could effectively predict the risk of 30-day all cause hospital readmissions for AMI patients and provide decision support for the administration.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK