12 February 2021 in IEEE Transactions on Industrial Informatics
Automatic segmentation of lung lesions from COVID-19 computed tomography (CT) images can help to establish a quantitative model for diagnosis and treatment. For this reason, this article provides a new segmentation method to meet the needs of CT images processing under COVID-19 epidemic. The main steps are as follows: First, the proposed region of interest extraction implements patch mechanism strategy to satisfy the applicability of 3-D network and remove irrelevant background. Second, 3-D network is established to extract spatial features, where 3-D attention model promotes network to enhance target area. Then, to improve the convergence of network, a combination loss function is introduced to lead gradient optimization and training direction. Finally, data augmentation and conditional random field are applied to realize data resampling and binary segmentation. This method was assessed with some comparative experiment. By comparison, the proposed method reached the highest performance. Therefore, it has potential clinical applications.
23 October 2020 in IEEE Access
Retinal blood vessels are the basis for clinical diagnosis of some diseases. Achieving automatic retinal vessel segmentation from fundus images is an important and challenging work. In this paper, a neural network architecture based on the Dense U-net using Inception module is proposed for retinal vessel segmentation. First, the skip connections in traditional U-net are replaced with Dense Block to achieve full fusion of features from shallow layers to deep layers. Then, the Inception module is applied to supersede the traditional convolution operation. Thus, vessel features corresponding to the convolution kernels of different sizes can be extracted. Finally, Generative Adversarial Networks (GAN) are adopted in the training phase. The Dense U-net using Inception module is treated as the generator of GAN, and a multilayer neural network is created as the discriminator of GAN. The generator and discriminator are trained alternately. The loss function is a combination of segmentation loss and GAN loss. So that segmentation results can be fitted to the ground truth from both pixel value and pixel distribution. The algorithm proposed in this paper is verified on the public Digital Retinal Images for Vessel Extraction (DRIVE) dataset, where the Dice rate reaches 82.15%, and the AU-ROC and AU-PR reach 0.9772 and 0.9058, respectively. Experiments show that the proposed algorithm is effective in realizing automatic retinal vessel segmentation.
1 August 2018 in Journal of Food Processing and Preservation
Chinese eaglewood leaves are dried in a constant environment testing machine, and the curves of drying and rolling deformation are obtained under different temperatures, humidity levels, and restricted space conditions. Results show that drying rate increases with drying temperature but slightly decreases with an increase in humidity or a decrease in restricted space. The edges of the leaves roll inwardly when dried without any limit but roll outwardly under a restricted space. A low restricted height indicates a small leaf roll index (LRI), which is conducive to shredding in the later stage of manufacture of eaglewood cigarettes. When the leaves are dried restrictedly, the effects of drying temperature and air humidity on the LRI are minimal, and the LRI slightly increases with an increase in temperature or a decrease in humidity. Practical Applications China Jiangsu Tobacco Industrial Company has added a certain amount of dry eaglewood leaves to tobacco and manufactured cigarettes of eaglewood leaves. The pharmacological effects of eaglewood can reduce the harm of tobacco on the human body. Fresh eaglewood leaves are dried until the moisture content is below 14% and then rehydrated and shredded during the manufacture of cut tobacco. Rolling deformation occurs during drying, and over rolling of leaves produces irregular shredding that affects the quality of cut tobacco. In this study, the drying and rolling deformation characteristics of eaglewood leaves in restricted spaces are explored to reduce the rolling deformation of the leaves during drying.
20 January 2020 in IEEE Access
Identification of the coronary artery in sequence angiograms is an important issue in the analysis and treatment of coronary angiography data, due to the ambiguities in the coronary vessels resulting from cardiac motion and the differences in the projection angles in angiograms. In this paper, we propose a novel coronary artery identification method based on curvature scale space image matching. First, points on the centerline of each coronary artery branch are convolved with Gaussian kernel functions of different scales to obtain the corresponding curvature scale space (CSS) image during the curve evolution. Then, a CSS image matching method for open curves, which uses a combination of open feature vectors and closed feature vectors, is proposed. Finally, curve matching transitivity is used to solve the problem of the inability to recognize the coronary artery when the difference in the imaging angles is too large. The experimental results demonstrate that the proposed method can recognize every coronary artery in different views, even when the difference in the imaging angles is 29.8°.
5 October 2020 in bioRxiv
Background: Auditory brainstem response (ABR) test is widely used in newborn hearing screening and hearing disease diagnosis. Identifying and marking are challenging and repetitive tasks because of complex rules of ABR characteristic waveform and interference of background noise. Methods: This study proposes an automatic method to recognize ABR characteristic waveform. First, binarization is created to mark 1024 sampling points accordingly. The selected characteristic area of ABR data is 0-8ms. The marking area is enlarged to expand feature information and reduce marking error. Second, a bi-directional long short-term memory (BiLSTM) network structure is established to improve relevance of sampling points, and an ABR sampling point classifier is obtained by training. Finally, mark points are obtained through thresholding. Results: Specific structure, related parameters, recognition effect, and noise resistance of network were explored in 614 sets of ABR clinical data, and recognition accuracy of waves I, III, and V can reach 92.91%. Discussion: Thus, the proposed method can reduce the repetitive work of doctors and meet accuracy effectively. Therefore, this method has clinical potential.