Browsing by Author "Chen Y"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- ItemA multi-label classification model for full slice brain computerised tomography image(BioMed Central Ltd, 2020-11-18) Li J; Fu G; Chen Y; Li P; Liu B; Pei Y; Feng HBACKGROUND: Screening of the brain computerised tomography (CT) images is a primary method currently used for initial detection of patients with brain trauma or other conditions. In recent years, deep learning technique has shown remarkable advantages in the clinical practice. Researchers have attempted to use deep learning methods to detect brain diseases from CT images. Methods often used to detect diseases choose images with visible lesions from full-slice brain CT scans, which need to be labelled by doctors. This is an inaccurate method because doctors detect brain disease from a full sequence scan of CT images and one patient may have multiple concurrent conditions in practice. The method cannot take into account the dependencies between the slices and the causal relationships among various brain diseases. Moreover, labelling images slice by slice spends much time and expense. Detecting multiple diseases from full slice brain CT images is, therefore, an important research subject with practical implications. RESULTS: In this paper, we propose a model called the slice dependencies learning model (SDLM). It learns image features from a series of variable length brain CT images and slice dependencies between different slices in a set of images to predict abnormalities. The model is necessary to only label the disease reflected in the full-slice brain scan. We use the CQ500 dataset to evaluate our proposed model, which contains 1194 full sets of CT scans from a total of 491 subjects. Each set of data from one subject contains scans with one to eight different slice thicknesses and various diseases that are captured in a range of 30 to 396 slices in a set. The evaluation results present that the precision is 67.57%, the recall is 61.04%, the F1 score is 0.6412, and the areas under the receiver operating characteristic curves (AUCs) is 0.8934. CONCLUSION: The proposed model is a new architecture that uses a full-slice brain CT scan for multi-label classification, unlike the traditional methods which only classify the brain images at the slice level. It has great potential for application to multi-label detection problems, especially with regard to the brain CT images.
- ItemDoes information asymmetry lead to higher debt financing? Evidence from China during the NTS Reform period(Emerald Publishing Limited, 2018-07-16) Qu W; Wongchoti U; Wu F; Chen YPurpose The purpose of this paper is to test an implication of the pecking order theory to explain capital structure decisions among Chinese listed companies during the 2005-2007 NTS Reform transition period. Design/methodology/approach The authors utilize direct proxies for information asymmetry based on microstructure models including Probability of the arrival of informed trades (PIN), Adverse selection component of the bid-ask spread (λ), Illiquidity ratio (ILLIQ) and liquidity ratio, and Information asymmetry index (InfoAsy) to examine their relation with firms’ debt financing. Findings Consistent with the prediction of Pecking Order Theory, the authors find that companies for which stock investors are challenged with more severe informational disadvantages are associated with higher degree of leverage use. Originality/value The study provides a more direct test on the positive relation between information asymmetry and financial leverage of Chinese firms. In contrast to previous findings by Chen (2004), the results suggest that capital structure choices among Chinese firms progressively conform to conventional finance theories (e.g. Pecking Order Theory) with the decline of non-tradable shares.
- ItemEnhanced removal of arsenic and cadmium from contaminated soils using a soluble humic substance coupled with chemical reductant.(2023-03-01) Wei J; Tu C; Xia F; Yang L; Chen Q; Chen Y; Deng S; Yuan G; Wang H; Jeyakumar P; Bhatnagar ASoil washing is an efficient, economical, and green remediation technology for removing several heavy metal (loid)s from contaminated industrial sites. The extraction of green and efficient washing agents from low-cost feedback is crucially important. In this study, a soluble humic substance (HS) extracted from leonardite was first tested to wash soils (red soil, fluvo-aquic soil, and black soil) heavily contaminated with arsenic (As) and cadmium (Cd). A D-optimal mixture design was investigated to optimize the washing parameters. The optimum removal efficiencies of As and Cd by single HS washing were found to be 52.58%-60.20% and 58.52%-86.69%, respectively. Furthermore, a two-step sequential washing with chemical reductant NH2OH•HCl coupled with HS (NH2OH•HCl + HS) was performed to improve the removal efficiency of As and Cd. The two-step sequential washing significantly enhanced the removal of As and Cd to 75.25%-81.53% and 64.53%-97.64%, which makes the residual As and Cd in soil below the risk control standards for construction land. The two-step sequential washing also effectively controlled the mobility and bioavailability of residual As and Cd. However, the activities of soil catalase and urease significantly decreased after the NH2OH•HCl + HS washing. Follow-up measures such as soil neutralization could be applied to relieve and restore the soil enzyme activity. In general, the two-step sequential soil washing with NH2OH•HCl + HS is a fast and efficient method for simultaneously removing high content of As and Cd from contaminated soils.