Search

Your search keyword '"Ju, Rui"' showing total 579 results
Start Over Author "Ju, Rui"
579 results on '"Ju, Rui"'

1. Efficient GANs for Document Image Binarization Based on DWT and Normalization

Author

Ju, Rui-Yang, Wong, KokSheik, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

For document image binarization task, generative adversarial networks (GANs) can generate images where shadows and noise are effectively removed, which allow for text information extraction. The current state-of-the-art (SOTA) method proposes a three-stage network architecture that utilizes six GANs. Despite its excellent model performance, the SOTA network architecture requires long training and inference times. To overcome this problem, this work introduces an efficient GAN method based on the three-stage network architecture that incorporates the Discrete Wavelet Transformation and normalization to reduce the input image size, which in turns, decrease both training and inference times. In addition, this work presents novel generators, discriminators, and loss functions to improve the model's performance. Experimental results show that the proposed method reduces the training time by 10% and the inference time by 26% when compared to the SOTA method while maintaining the model performance at 73.79 of Avg-Score. Our implementation code is available on GitHub at https://github.com/RuiyangJu/Efficient_Document_Image_Binarization.

Published
2024

2. Global Context Modeling in YOLOv8 for Pediatric Wrist Fracture Detection

Author

Ju, Rui-Yang, Chien, Chun-Tse, Lin, Chia-Min, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Children often suffer wrist injuries in daily life, while fracture injuring radiologists usually need to analyze and interpret X-ray images before surgical treatment by surgeons. The development of deep learning has enabled neural network models to work as computer-assisted diagnosis (CAD) tools to help doctors and experts in diagnosis. Since the YOLOv8 models have obtained the satisfactory success in object detection tasks, it has been applied to fracture detection. The Global Context (GC) block effectively models the global context in a lightweight way, and incorporating it into YOLOv8 can greatly improve the model performance. This paper proposes the YOLOv8+GC model for fracture detection, which is an improved version of the YOLOv8 model with the GC block. Experimental results demonstrate that compared to the original YOLOv8 model, the proposed YOLOv8-GC model increases the mean average precision calculated at intersection over union threshold of 0.5 (mAP 50) from 63.58% to 66.32% on the GRAZPEDWRI-DX dataset, achieving the state-of-the-art (SOTA) level. The implementation code for this work is available on GitHub at https://github.com/RuiyangJu/YOLOv8_Global_Context_Fracture_Detection.

Published
2024

3. FAD-SAR: A Novel Fishing Activity Detection System via Synthetic Aperture Radar Images Based on Deep Learning Method

Author

Bai, Yanbing, Li, Siao, Ju, Rui-Yang, Yang, Zihao, Yu, Jinze, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Illegal, unreported, and unregulated (IUU) fishing activities seriously affect various aspects of human life. However, traditional methods for detecting and monitoring IUU fishing activities at sea have limitations. Although synthetic aperture radar (SAR) can complement existing vessel detection systems, extracting useful information from SAR images using traditional methods remains a challenge, especially in IUU fishing. This paper proposes a deep learning based fishing activity detection system, which is implemented on the xView3 dataset using six classical object detection models: SSD, RetinaNet, FSAF, FCOS, Faster R-CNN, and Cascade R-CNN. In addition, this work employs different enhancement techniques to improve the performance of the Faster R-CNN model. The experimental results demonstrate that training the Faster R-CNN model using the Online Hard Example Mining (OHEM) strategy increases the Avg-F1 value from 0.212 to 0.216.

Published
2024

4. Flood Data Analysis on SpaceNet 8 Using Apache Sedona

Author

Bai, Yanbing, Yang, Zihao, Yu, Jinze, Ju, Rui-Yang, Yang, Bin, Mas, Erick, and Koshimura, Shunichi

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

With the escalating frequency of floods posing persistent threats to human life and property, satellite remote sensing has emerged as an indispensable tool for monitoring flood hazards. SpaceNet8 offers a unique opportunity to leverage cutting-edge artificial intelligence technologies to assess these hazards. A significant contribution of this research is its application of Apache Sedona, an advanced platform specifically designed for the efficient and distributed processing of large-scale geospatial data. This platform aims to enhance the efficiency of error analysis, a critical aspect of improving flood damage detection accuracy. Based on Apache Sedona, we introduce a novel approach that addresses the challenges associated with inaccuracies in flood damage detection. This approach involves the retrieval of cases from historical flood events, the adaptation of these cases to current scenarios, and the revision of the model based on clustering algorithms to refine its performance. Through the replication of both the SpaceNet8 baseline and its top-performing models, we embark on a comprehensive error analysis. This analysis reveals several main sources of inaccuracies. To address these issues, we employ data visual interpretation and histogram equalization techniques, resulting in significant improvements in model metrics. After these enhancements, our indicators show a notable improvement, with precision up by 5%, F1 score by 2.6%, and IoU by 4.5%. This work highlights the importance of advanced geospatial data processing tools, such as Apache Sedona. By improving the accuracy and efficiency of flood detection, this research contributes to safeguarding public safety and strengthening infrastructure resilience in flood-prone areas, making it a valuable addition to the field of remote sensing and disaster management.

Published
2024

5. YOLOv9 for Fracture Detection in Pediatric Wrist Trauma X-ray Images

Author

Chien, Chun-Tse, Ju, Rui-Yang, Chou, Kuang-Yi, and Chiang, Jen-Shiun

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

The introduction of YOLOv9, the latest version of the You Only Look Once (YOLO) series, has led to its widespread adoption across various scenarios. This paper is the first to apply the YOLOv9 algorithm model to the fracture detection task as computer-assisted diagnosis (CAD) to help radiologists and surgeons to interpret X-ray images. Specifically, this paper trained the model on the GRAZPEDWRI-DX dataset and extended the training set using data augmentation techniques to improve the model performance. Experimental results demonstrate that compared to the mAP 50-95 of the current state-of-the-art (SOTA) model, the YOLOv9 model increased the value from 42.16% to 43.73%, with an improvement of 3.7%. The implementation code is publicly available at https://github.com/RuiyangJu/YOLOv9-Fracture-Detection., Comment: Accepted by Electronics Letters

Published
2024

6. YOLOv8-AM: YOLOv8 with Attention Mechanisms for Pediatric Wrist Fracture Detection

Author

Chien, Chun-Tse, Ju, Rui-Yang, Chou, Kuang-Yi, Xieerke, Enkaer, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Wrist trauma and even fractures occur frequently in daily life, particularly among children who account for a significant proportion of fracture cases. Before performing surgery, surgeons often request patients to undergo X-ray imaging first and prepare for it based on the analysis of the radiologist. With the development of neural networks, You Only Look Once (YOLO) series models have been widely used in fracture detection as computer-assisted diagnosis (CAD). In 2023, Ultralytics presented the latest version of the YOLO models, which has been employed for detecting fractures across various parts of the body. Attention mechanism is one of the hottest methods to improve the model performance. This research work proposes YOLOv8-AM, which incorporates the attention mechanism into the original YOLOv8 architecture. Specifically, we respectively employ four attention modules, Convolutional Block Attention Module (CBAM), Global Attention Mechanism (GAM), Efficient Channel Attention (ECA), and Shuffle Attention (SA), to design the improved models and train them on GRAZPEDWRI-DX dataset. Experimental results demonstrate that the mean Average Precision at IoU 50 (mAP 50) of the YOLOv8-AM model based on ResBlock + CBAM (ResCBAM) increased from 63.6% to 65.8%, which achieves the state-of-the-art (SOTA) performance. Conversely, YOLOv8-AM model incorporating GAM obtains the mAP 50 value of 64.2%, which is not a satisfactory enhancement. Therefore, we combine ResBlock and GAM, introducing ResGAM to design another new YOLOv8-AM model, whose mAP 50 value is increased to 65.0%. The implementation code for this study is available on GitHub at https://github.com/RuiyangJu/Fracture_Detection_Improved_YOLOv8.

Published
2024

8. CCDWT-GAN: Generative Adversarial Networks Based on Color Channel Using Discrete Wavelet Transform for Document Image Binarization

Author

Ju, Rui-Yang, Lin, Yu-Shian, Chiang, Jen-Shiun, Chen, Chih-Chia, Chen, Wei-Han, and Chien, Chun-Tse

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

To efficiently extract textual information from color degraded document images is a significant research area. The prolonged imperfect preservation of ancient documents has led to various types of degradation, such as page staining, paper yellowing, and ink bleeding. These types of degradation badly impact the image processing for features extraction. This paper introduces a novelty method employing generative adversarial networks based on color channel using discrete wavelet transform (CCDWT-GAN). The proposed method involves three stages: image preprocessing, image enhancement, and image binarization. In the initial step, we apply discrete wavelet transform (DWT) to retain the low-low (LL) subband image, thereby enhancing image quality. Subsequently, we divide the original input image into four single-channel colors (red, green, blue, and gray) to separately train adversarial networks. For the extraction of global and local features, we utilize the output image from the image enhancement stage and the entire input image to train adversarial networks independently, and then combine these two results as the final output. To validate the positive impact of the image enhancement and binarization stages on model performance, we conduct an ablation study. This work compares the performance of the proposed method with other state-of-the-art (SOTA) methods on DIBCO and H-DIBCO ((Handwritten) Document Image Binarization Competition) datasets. The experimental results demonstrate that CCDWT-GAN achieves a top two performance on multiple benchmark datasets. Notably, on DIBCO 2013 and 2016 dataset, our method achieves F-measure (FM) values of 95.24 and 91.46, respectively., Comment: accepted by PRICAI 2023

Published
2023

9. Fracture Detection in Pediatric Wrist Trauma X-ray Images Using YOLOv8 Algorithm

Author

Ju, Rui-Yang and Cai, Weiming

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Hospital emergency departments frequently receive lots of bone fracture cases, with pediatric wrist trauma fracture accounting for the majority of them. Before pediatric surgeons perform surgery, they need to ask patients how the fracture occurred and analyze the fracture situation by interpreting X-ray images. The interpretation of X-ray images often requires a combination of techniques from radiologists and surgeons, which requires time-consuming specialized training. With the rise of deep learning in the field of computer vision, network models applying for fracture detection has become an important research topic. In this paper, we use data augmentation to improve the model performance of YOLOv8 algorithm (the latest version of You Only Look Once) on a pediatric wrist trauma X-ray dataset (GRAZPEDWRI-DX), which is a public dataset. The experimental results show that our model has reached the state-of-the-art (SOTA) mean average precision (mAP 50). Specifically, mAP 50 of our model is 0.638, which is significantly higher than the 0.634 and 0.636 of the improved YOLOv7 and original YOLOv8 models. To enable surgeons to use our model for fracture detection on pediatric wrist trauma X-ray images, we have designed the application "Fracture Detection Using YOLOv8 App" to assist surgeons in diagnosing fractures, reducing the probability of error analysis, and providing more useful information for surgery., Comment: Scientific Reports

Published
2023

10. Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy

Author

Ju, Rui-Yang, Chen, Chih-Chia, Chiang, Jen-Shiun, Lin, Yu-Shian, Chen, Wei-Han, and Chien, Chun-Tse

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The Transformer-based method has demonstrated remarkable performance for image super-resolution in comparison to the method based on the convolutional neural networks (CNNs). However, using the self-attention mechanism like SwinIR (Image Restoration Using Swin Transformer) to extract feature information from images needs a significant amount of computational resources, which limits its application on low computing power platforms. To improve the model feature reuse, this research work proposes the Interval Dense Connection Strategy, which connects different blocks according to the newly designed algorithm. We apply this strategy to SwinIR and present a new model, which named SwinOIR (Object Image Restoration Using Swin Transformer). For image super-resolution, an ablation study is conducted to demonstrate the positive effect of the Interval Dense Connection Strategy on the model performance. Furthermore, we evaluate our model on various popular benchmark datasets, and compare it with other state-of-the-art (SOTA) lightweight models. For example, SwinOIR obtains a PSNR of 26.62 dB for x4 upscaling image super-resolution on Urban100 dataset, which is 0.15 dB higher than the SOTA model SwinIR. For real-life application, this work applies the lastest version of You Only Look Once (YOLOv8) model and the proposed model to perform object detection and real-life image super-resolution on low-quality images. This implementation code is publicly available at https://github.com/Rubbbbbbbbby/SwinOIR.

Published
2023

11. CCDWT-GAN: Generative Adversarial Networks Based on Color Channel Using Discrete Wavelet Transform for Document Image Binarization

Author

Ju, Rui-Yang, Lin, Yu-Shian, Chiang, Jen-Shiun, Chen, Chih-Chia, Chen, Wei-Han, Chien, Chun-Tse, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Liu, Fenrong, editor, Sadanandan, Arun Anand, editor, Pham, Duc Nghia, editor, Mursanto, Petrus, editor, and Lukose, Dickson, editor

Published
2024
Full Text
View/download PDF

12. Three-stage binarization of color document images based on discrete wavelet transform and generative adversarial networks

Author

Ju, Rui-Yang, Lin, Yu-Shian, Jin, Yanlin, Chen, Chih-Chia, Chien, Chun-Tse, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The efficient segmentation of text information from the background in degraded color document images is an important challenge in the preservation of ancient manuscripts. The imperfect preservation of ancient manuscripts has led to various types of degradation over time, such as staining, yellowing, and ink seepage, badly affecting document image binarization results. This work proposes a three-stage method to generate binarization image results for degraded colour document images using generative adversarial networks (GANs). Stage-1 involves applying discrete wavelet transform and retaining the low-low subband images for document image enhancement. In Stage-2, the original input image is split into red, green, and blue (RGB) three single-channel images and one grayscale image, and each image is trained with independent GANs to extract color foreground information. In Stage-3, the output images of Stage-2 and the resized input images are used to train independent GANs to generate document binarization results, enabling the combination of global and local features. The experimental results show that the Avg-Score of the proposed method is 77.64, 77.95, 79.05, 76.38, 75.34, and 77.00 on the (H)-DIBCO 2011, 2013, 2014, 2016, 2017, and 2018 datasets, which achieves the state-of-the-art level. The implementation code for this work is available at https://github.com/abcpp12383/ThreeStageBinarization.

Published
2022

13. Connection Reduction of DenseNet for Image Recognition

Author

Ju, Rui-Yang, Chiang, Jen-Shiun, Chen, Chih-Chia, and Lin, Yu-Shian

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Convolutional Neural Networks (CNN) increase depth by stacking convolutional layers, and deeper network models perform better in image recognition. Empirical research shows that simply stacking convolutional layers does not make the network train better, and skip connection (residual learning) can improve network model performance. For the image classification task, models with global densely connected architectures perform well in large datasets like ImageNet, but are not suitable for small datasets such as CIFAR-10 and SVHN. Different from dense connections, we propose two new algorithms to connect layers. Baseline is a densely connected network, and the networks connected by the two new algorithms are named ShortNet1 and ShortNet2 respectively. The experimental results of image classification on CIFAR-10 and SVHN show that ShortNet1 has a 5% lower test error rate and 25% faster inference time than Baseline. ShortNet2 speeds up inference time by 40% with less loss in test accuracy. Code and pre-trained models are available at https://github.com/RuiyangJu/Connection_Reduction.

Published
2022

15. Culture of glioblastoma U87 stem-like cells and identification of its metabolic phenotype and tumorigenic ability

Author

QIU Jiaxing, LIU Yuhan, GUO Hongjiang, ZHANG Diya, WANG Yucheng, JU Rui, GUO Lei

Subjects
glioblastoma, stem-like cell, neuro-sphere, stemness, Medicine
Abstract

Objective To cultivate glioblastoma U87 stem-like cells (SLCs) and to detect the level of stemness biomarkers, mitochondrial respiratory capacity and the capacity of in vivo tumorigenesis. Methods B-27, growth factors EGF and bFGF was added into DMEM/F-12 culture in serum-free stem cell culture medium for U87 SLCs. Suspended culture of U87 SLCs was suspended using the neuro-sphere formation assay, while adherent culture of U87 SLCs was achieved by coating Matrigel matrix on the culture surface. The mRNA and protein level of stemness biomarkers in culture were detected using real-time quantitative PCR and Western blot. The proportion of CD133+ cells in culture was detected by flow cytometry. The changes of cell oxygen consumption rate were detected by Seahorse cell metabolism analysis. Cell tumorigenesis ability was verified by subcutaneous tumor transplantation in animals. Results U87 SLCs in stem cell culture medium would grow into typical sphere morphology within one week, and the spheres would continue to grow as the culture process prolongs. At the appropriate concentration of adhesive, U87 SLCs adhered to and grow well in stem cell culture medium. The mRNA transcription of stemness biomarkers such as CD133, nestin, OLIG2, CD44, CD15, and integrin α6(ITGA6) was significantly increased as found in both culture methods, and the protein levels of CD133 and nestin were also increased under both methods(P＜0.05). U87 SLCs showed higher mitochondrial reserve respiratory capacity (P＜0.05). U87 SLCs could form larger subcutaneous tumors with fewer inoculated cells (P＜0.05), and grew faster in vivo with stronger tumorigenic ability. Conclusions U87 SLCs have typical stemness characteristics and may function as tumor cell model with higher stemness properties.

Published
2024
Full Text
View/download PDF

16. Three-scene medication process simulation and its application in teaching of nursing pharmacology

Author

JU Rui, WANG Yucheng, ZHU Lei, YU Xiaoli, DONG Yan, GUO Lei

Subjects
undergraduate nurse students, nursing pharmacology, three-scene medication process simulation, knowledge connection, nursing communication, Medicine
Abstract

Objective To explore the effectiveness of a refined three-scene medication simulation teaching in nursing pharmacology course for nursing undergraduates. Methods Undergraduates of four-year curriculum in School of Nursing, Peking Union Medical College enrolled in 2020 and 2021 were selected as objects in control group and experimental group, respectively. Students in control group learned medication knowledge in a patient case simulation scene, while a refined three-scene medication process (before, during and after medication) simulation was applied in experimental group. The learning outcome was evaluated in forms of both test of key knowledge points and a questionnaire survey at the end of the semester. Results The scores of nursing pharmacology in experimental group were significantly higher than in control group (P＜0.001). The rates of right answer for knowledge associated with the refined class in experimental group were also significantly higher than in control group (P＜0.001). However, for knowledge that has not been adjusted for instructional design, there is no significant difference in answer accuracy between two groups. The survey by questionnaire showed that the students believed the three-scene medication simulation as good learning method and may support to strengthen professional quality of clinical performance. Conclusions Compared with the unrefined simulation class, the three-scene (before, during and after medication) class further simulated the real work flow, connected the knowledge points more reasonably, and enhanced the knowledge mastery and professional responsibility awareness.

Published
2023
Full Text
View/download PDF

17. Application and evaluation of micro-lecture combined with case-based learning in the teaching of pharmacotherapeutics for undergraduate nursing students

Author

WANG Yucheng, JU Rui, GUO Lei

Subjects
undergraduate nursing student, pharmacotherapeutics, micro-lecture, case-based learning, Medicine
Abstract

Objective To explore the application effect of micro-lecture combined with case-based learning (CBL) in the teaching of pharmacotherapeutics for undergraduate nursing students. Methods Two classes of undergraduate nursing students from School of Nursing, Peking Union Medical College enrolled in 2021 were selected as the research subjects. They were taught by micro-lecture combined with CBL (experimental group) or traditional lecture-based learning (LBL) (control group) respectively; The students' in-class quiz and final exam scores were collected. The questionnaires were distributed to evaluate the teaching effect and learning experience through analyzing the scores and questionnaire results. Results The in-class quiz scores of the experimental group were 77.52±17.68, and those of the control group were 66.34±21.35; The final exam scores of the experimental group were 64.34±26.17, and those of the control group were 52.33±25.55. The scores of the experimental group in both tests were significantly higher than those of the control group, and the differences were statistically significant (P＜0.01). The questionnaire findings indicated that the experimental group improved on the theoretical understanding, knowledge retention, medication practice proficiency, interest in learning, concentration in class, and interaction willingness in comparison to the control group. Conclusions Micro-lecture combined with CBL can enhance the learning effect of pharmacotherapeutics for undergraduate nursing students, and optimize the learning experience. It is worth to be further improved and promoted as a feasible teaching method.

Published
2023
Full Text
View/download PDF

18. Efficient Convolutional Neural Networks on Raspberry Pi for Image Classification

Author

Ju, Rui-Yang, Lin, Ting-Yu, Jian, Jia-Hao, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

With the good performance of deep learning algorithms in the field of computer vision (CV), the convolutional neural network (CNN) architecture has become a main backbone of the computer vision task. With the widespread use of mobile devices, neural network models based on platforms with low computing power are gradually being paid attention. However, due to the limitation of computing power, deep learning algorithms are usually not available on mobile devices. This paper proposes a lightweight convolutional neural network, TripleNet, which can operate easily on Raspberry Pi. Adopted from the concept of block connections in ThreshNet, the newly proposed network model compresses and accelerates the network model, reduces the amount of parameters of the network, and shortens the inference time of each image while ensuring the accuracy. Our proposed TripleNet and other state-of-the-art (SOTA) neural networks perform image classification experiments with the CIFAR-10 and SVHN datasets on Raspberry Pi. The experimental results show that, compared with GhostNet, MobileNet, ThreshNet, EfficientNet, and HarDNet, the inference time of TripleNet per image is shortened by 15%, 16%, 17%, 24%, and 30%, respectively. The detail codes of this work are available at https://github.com/RuiyangJu/TripleNet.

Published
2022

19. Aggregated Pyramid Vision Transformer: Split-transform-merge Strategy for Image Recognition without Convolutions

Author

Ju, Rui-Yang, Lin, Ting-Yu, Chiang, Jen-Shiun, Jian, Jia-Hao, Lin, Yu-Shian, and Huang, Liu-Rui-Yi

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

With the achievements of Transformer in the field of natural language processing, the encoder-decoder and the attention mechanism in Transformer have been applied to computer vision. Recently, in multiple tasks of computer vision (image classification, object detection, semantic segmentation, etc.), state-of-the-art convolutional neural networks have introduced some concepts of Transformer. This proves that Transformer has a good prospect in the field of image recognition. After Vision Transformer was proposed, more and more works began to use self-attention to completely replace the convolutional layer. This work is based on Vision Transformer, combined with the pyramid architecture, using Split-transform-merge to propose the group encoder and name the network architecture Aggregated Pyramid Vision Transformer (APVT). We perform image classification tasks on the CIFAR-10 dataset and object detection tasks on the COCO 2017 dataset. Compared with other network architectures that use Transformer as the backbone, APVT has excellent results while reducing the computational cost. We hope this improved strategy can provide a reference for future Transformer research in computer vision., Comment: 6 pages, 5 figures

Published
2022

20. ThreshNet: An Efficient DenseNet Using Threshold Mechanism to Reduce Connections

Author

Ju, Rui-Yang, Lin, Ting-Yu, Jian, Jia-Hao, Chiang, Jen-Shiun, and Yang, Wei-Bin

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

With the continuous development of neural networks for computer vision tasks, more and more network architectures have achieved outstanding success. As one of the most advanced neural network architectures, DenseNet shortcuts all feature maps to solve the model depth problem. Although this network architecture has excellent accuracy with low parameters, it requires an excessive inference time. To solve this problem, HarDNet reduces the connections between the feature maps, making the remaining connections resemble harmonic waves. However, this compression method may result in a decrease in the model accuracy and an increase in the parameters and model size. This network architecture may reduce the memory access time, but its overall performance can still be improved. Therefore, we propose a new network architecture, ThreshNet, using a threshold mechanism to further optimize the connection method. Different numbers of connections for different convolution layers are discarded to accelerate the inference of the network. The proposed network has been evaluated with image classification using CIFAR 10 and SVHN datasets under platforms of NVIDIA RTX 3050 and Raspberry Pi 4. The experimental results show that, compared with HarDNet68, GhostNet, MobileNetV2, ShuffleNet, and EfficientNet, the inference time of the proposed ThreshNet79 is 5%, 9%, 10%, 18%, and 20% faster, respectively. The number of parameters of ThreshNet95 is 55% less than that of HarDNet85. The new model compression and model acceleration methods can speed up the inference time, enabling network models to operate on mobile devices., Comment: IEEE Access

Published
2022
Full Text
View/download PDF

23. New Pruning Method Based on DenseNet Network for Image Classification

Author

Ju, Rui-Yang, Lin, Ting-Yu, and Chiang, Jen-Shiun

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Deep neural networks have made significant progress in the field of computer vision. Recent studies have shown that depth, width and shortcut connections of neural network architectures play a crucial role in their performance. One of the most advanced neural network architectures, DenseNet, has achieved excellent convergence rates through dense connections. However, it still has obvious shortcomings in the usage of amount of memory. In this paper, we introduce a new type of pruning tool, threshold, which refers to the principle of the threshold voltage in MOSFET. This work employs this method to connect blocks of different depths in different ways to reduce the usage of memory. It is denoted as ThresholdNet. We evaluate ThresholdNet and other different networks on datasets of CIFAR10. Experiments show that HarDNet is twice as fast as DenseNet, and on this basis, ThresholdNet is 10% faster and 10% lower error rate than HarDNet., Comment: 5 pages, 3 figures, TAAI 2021

Published
2021

25. Practice guideline on ovarian tissue cryopreservation and transplantation in the prevention and treatment of iatrogenic premature ovarian insufficiency

Author

Ruan, Xiangyan, Xu, Che, Huang, Hefeng, Xu, Binghe, Du, Juan, Cheng, Jiaojiao, Jin, Fengyu, Gu, Muqing, Kong, Weimin, Yin, Chenghong, Wu, Yurui, Tian, Qinjie, Cao, Yunxia, Wu, Ruifang, Xu, Liangzhi, Jin, Jing, Li, Yanglu, Dai, Yinmei, Ju, Rui, Ma, Fei, Wang, Gang, Wei, Wei, Huang, Xiaojun, Qin, Maoquan, Lin, Yuan, Sun, Yuan, Liu, Rong, Zhang, Wei, Li, Xiaodong, Zou, Lin, Hao, Min, Ye, Xiyang, Wang, Fuling, Wang, Yue, Hu, Zhuoying, Huang, Yanhong, Zhu, Tianyuan, Yang, Caihong, Wang, Jinping, Yang, Xiaomin, Ni, Rong, Wang, Liqun, Luo, Guangxia, Min, Aiping, Zhang, Siyou, Li, Peiling, Cheng, Linghui, Li, Lianfang, Jin, Quanfang, Shi, Dongmei, Li, Yan, Ren, Fangying, Cheng, Yanxiang, Niu, Jumin, Tian, Ying, and Mueck, Alfred O.

Published
2024
Full Text
View/download PDF

26. UTX condensation underlies its tumour-suppressive activity

Author

Shi, Bi, Li, Wei, Song, Yansu, Wang, Zhenjia, Ju, Rui, Ulman, Aleksandra, Hu, Jing, Palomba, Francesco, Zhao, Yanfang, Le, John Philip, Jarrard, William, Dimoff, David, Digman, Michelle A, Gratton, Enrico, Zang, Chongzhi, and Jiang, Hao

Subjects
Biochemistry and Cell Biology, Biomedical and Clinical Sciences, Biological Sciences, Genetics, 1.1 Normal biological development and functioning, Underpinning research, Cancer, Animals, Cell Differentiation, Chromatin, DNA-Binding Proteins, Embryonic Stem Cells, Genes, Tumor Suppressor, HEK293 Cells, Histone Demethylases, Humans, Intrinsically Disordered Proteins, Mice, Neoplasm Proteins, Protein Processing, Post-Translational, THP-1 Cells, General Science & Technology
Abstract

UTX (also known as KDM6A) encodes a histone H3K27 demethylase and is an important tumour suppressor that is frequently mutated in human cancers1. However, as the demethylase activity of UTX is often dispensable for mediating tumour suppression and developmental regulation2-8, the underlying molecular activity of UTX remains unknown. Here we show that phase separation of UTX underlies its chromatin-regulatory activity in tumour suppression. A core intrinsically disordered region (cIDR) of UTX forms phase-separated liquid condensates, and cIDR loss caused by the most frequent cancer mutation of UTX is mainly responsible for abolishing tumour suppression. Deletion, mutagenesis and replacement assays of the intrinsically disordered region demonstrate a critical role of UTX condensation in tumour suppression and embryonic stem cell differentiation. As shown by reconstitution in vitro and engineered systems in cells, UTX recruits the histone methyltransferase MLL4 (also known as KMT2D) to the same condensates and enriches the H3K4 methylation activity of MLL4. Moreover, UTX regulates genome-wide histone modifications and high-order chromatin interactions in a condensation-dependent manner. We also found that UTY, the Y chromosome homologue of UTX with weaker tumour-suppressive activity, forms condensates with reduced molecular dynamics. These studies demonstrate a crucial biological function of liquid condensates with proper material states in enabling the tumour-suppressive activity of a chromatin regulator.

Published
2021

27. Stable knockout of Mcart-1 down-regulates the proliferation and oxidative phosphorylation of RAW264.7 macrophages

Author

ZHAO Yongjing, QIU Jiaxing, ZHANG Diya, GUO Hongjiang, WANG Yucheng, JU Rui, GUO Lei

Subjects
crispr/cas9, macrophage, mcart-1, metabolic phenotype, Medicine
Abstract

Objective To construct a RAW264.7 macrophage cell strain with stable knockout of Mcart-1 by using CRISPR/Cas9 gene editing technique, and to detect its biological function. Methods Cas9 and sgRNA lentivirus were used to infect RAW264.7 macrophages in two steps. Positive cells were screened with puromycin and hygromycin, and single cells were plated by flow cytometry to obtain monoclonal cells; Expression of Cas9 and Mcart-1 was detected by qPCR and Western blot, and the mutation site was confirmed by sequence analysis. Cells number counting and carboxyfluorescein diacetate succinimdyl ester (CFSE) staining were used to detect cell proliferation; ATP detection kit was used to measure total ATP content in cells; Seahorse bioenergy analyzer was used to detect cell oxygen consumption rate(OCR) and glycolysis rate. Results Successfully constructed RAW264.7 cell strain with Mcart-1 stably knocked out, denoted as Mcart-1-/--RAW264.7; A frameshift mutation occurred in Mcart-1 gene in this cell strain; Compared with the wild-type RAW264.7 cell line (RAW264.7), the OCR increased and the extracellular acidification rate(ECAR)of Mcart-1-/--RAW264.7 cells decreased; The total ATP content in Mcart-1 knocked out cells decreased(P＜0.05),mitochondrial ATP production rate decreased (P＜0.01), while glycolytic ATP production rate increased (P＜0.01); The proliferation activity of RAW264.7 cells decreased after Mcart-1 was knocked out(P＜0.001). Conclusions The proliferation activity and oxidative phosphorylation level of RAW264.7 cells were both down-regulated after the Mcart-1 was stably knocked out. This cell strain is an important tool for exploring the function of cells in the tumor microenvironment.

Published
2023
Full Text
View/download PDF

30. Emerging risks of non‐native species escapes from aquaculture: Call for policy improvements in China and other developing countries

Author

Ju, Rui‐Ting, Li, Xiao, Jiang, Jia‐Jia, Wu, Jihua, Liu, Jianguo, Strong, Donald R, and Li, Bo

Subjects
Prevention, Life on Land, Life Below Water, aquaculture, aquatic species, biological diversity, biological invasion, conservation policy, domestic species, exotic species, non-native aquatic species, Ecological Applications, Environmental Science and Management, Ecology
Abstract

Global aquaculture relies heavily on the farming of non-native aquatic species (hereafter, NAS). NAS escapes from aquaculture facilities can result in serious aquatic bio-invasions, which has been an important issue in the FAO Blue Growth Initiative. A regulatory quagmire regarding NAS farming and escapes, however, exists in most developing countries. We discuss aquaculture expansion and NAS escapes, illustrate emerging risks and propose recommendations for improved aquaculture management across developing countries and particularly for China. In China, 68 NAS are known to have successfully established feral populations in natural habitats due to recurrent leakages or escapes; among the 68 NAS, 52 represent risks to native aquatic ecosystems. In addition to affecting a country's own biodiversity and ecosystem functions, NAS escapees can also threaten the biosecurity of shared waters in neighbouring countries. Policy implications. Non-native aquatic species (NAS) escapes have already had adverse ecological effects in China and other developing countries. The importance of this problem, however, is not adequately recognized by current conservation policies in developing countries. To conserve biodiversity and to support the goal of FAO's sustainable aquaculture, developing countries should now take responsible actions to address NAS escapes through policy and management improvements. Specifically, these countries should pass comprehensive legislation, establish effective agencies and national standards and planning and enhance integrated research and education to deal with risk assessment, prevention, monitoring and control of NAS escapes. Given that China is the world's largest aquacultural producer, China can create a model for other developing countries that will increase the biosecurity and sustainability of global aquaculture.

Published
2020

31. Effects of carboxyamidotriazole-orotate on oncogenic phenotypes of mouse glioma-associated macrophage cell lines

Author

MA Rui, YANG Li-xing, CHEN Qiu-xia, QIU Jia-xing, WANG Yu-cheng, JU Rui, GUO Lei

Subjects
carboxyamidotriazole-orotate, tumor-associated macrophages, oxidative phosphorylation, hypoxia inducible factor, ppar coactivator-1β, Medicine
Abstract

Objective To explore regulation mechanism of carboxyamidotriazole-orotate (CTO) on oncogenic phenotype of mouse glioma-associated macrophages. Methods Bone marrow-derived macrophages(BMDMs) or RAW264.7 macrophages (tumor associated macrophages,TAMs) were prepared by super-culture of mouse glioma GL261 cells. The mRNA of TAMs oncogenic mediators was detected by qPCR. The oxygen consumption rate (OCR) of TAMs was measured by Seahorse bioenergy method. The protein level of hypoxia-inducible factor-1 α(HIF-1α), hypoxia-inducible factor-2 α (HIF-2α) and peroxisome proliferator-activated receptor (PPAR) co-activator factor-1 β (PGC-1β) in TAMs was detected by Western blot. Results CTO decreased M1-related mediators such as IL-1β, IL-6 and TNF-α(P＜0.001) and M2-related mediators such as IL-10, ARG-1 and TGF-β1(P＜0.001) in TAMs. CTO significantly down-regulated OCR in TAMs (P＜0.01); The protein level of HIF-1α and of HIF-2α in TAMs was down-regulated by CTO(P＜0.01),which was contrary to the effect of carboxycyanide-4-trifluormethoxyphenylhydrazone (FCCP),which in turn promoted mitochondrial oxygen consumption. CTO significantly reduced PGC-1β protein in TAMs (P＜0.001),an effect consistent with rotenone(RTN), a mitochondrial respiratory chain inhibitor. Conclusions A decreased oxidative phosphorylation down-regulates the expression of PGC-1β and destabilizes HIF-1α and HIF-2α, thus affects the production of several M1 and M2 phenotypes of carcinogenic mediators in TAMs, which may be a potential mechanism through which CTO improves the chemotherapy efficacy of glioblastoma.

Published
2022
Full Text
View/download PDF

32. Carboxyamidotriazole-orotate inhibits proliferation and mitochondrial metabolism in human pancreatic cancer gemcitabine-resistant cell strain

Author

CHEN Qiu-xia, YANG Li-xing, MA Rui, ZHAO Yong-jing, WANG Yu-cheng, JU Rui, GUO Lei

Subjects
carboxamidetriazole-orotate, pancreatic cancer, gemcitabine-resistant, apoptosis, Medicine
Abstract

Objective To investigate the effects of carboxyamidotriazole-orotate (CTO) on proliferation, apoptosis and metabolism of human pancreatic cancer gemcitabine-resistant cell line (ASPC-1-GEM, AG cells). Methods Cell viability was detected by sulforhodamine B(SRB); apoptosis was detected by annexin V/PI staining and flow cytometry; cell division rate was examined by CFSE microscopy and flow cytometry; The oxygen consumption rate (OCR) was detected by Seahorse bio-energy assay; intracellular NAD+ and NADH contents were detected with MTT and NAD+/NADH ratio was calculated. Autophagy-related protein expression was detected by Western blot. Results Comparing with the control group, the living AG cells were significantly reduced and the proportion of apoptotic cells was increased in CTO-treated group. The drug effect was time-dose dependent. The fluorescence intensity of CFSE staining of AG cells was significantly increased in the 20 μmol/L CTO group (P＜0.05); Compared with the control group, intracellular nicotinamide adenine dinucleotidehydrogen(NADH) content was increased after incubation with different concentrations of CTO treatment(P＜0.05 or P＜0.01) and no significant change in the content of nicotinamide adenine dinucleotide (NAD+) NDA+/NADH ratio was decreased(P＜0.01). The OCR in AG cells was significantly reduced and the expression of autophagy-related proteins was significantly inhibited in the CTO-treated group (P＜0.05). Conclusions CTO inhibits the proliferation of gemcitabine-resistant pancreatic cancer cells by inducing apoptosis, impairing mitochondrial respiration, and down-regulating the expression of autophagy-related proteins in AG cells.

Published
2022
Full Text
View/download PDF

39. Carboxyamidotriazole-orotate inhibits the proliferation and mitochondrial energy metabolism of mouse glioma cell line GL261

Author

HUANG Yu-qing, GAO Hong-ting, YANG Li-xing, CHEN Qiu-xia, WANG Yu-cheng, JU Rui, GUO Lei

Subjects
carboxyamidotriazole-orotate, cell line gl261, proliferation, apoptosis, mitochondrial respiration, Medicine
Abstract

Objective To investigate the effects of carboxyamidotriazole-orotate (CTO) on the proliferation and apoptosis of mouse glioma cell line GL261. Methods The proliferation of GL261 cells was detected by live cell counting. Cell cycle of GL261 cells was detected by PI staining and flow cytometry. Apoptosis of GL261 cells was detected by annexinⅤ/PI double staining and flow cytometry; DCFH-DA staining and flow cytometry were used to detect the content of intracellular reactive oxygen species (ROS) of GL261 cells. The oxygen consumption rate (OCR) of GL261 cells was measured in Seahorse bioenergy assay. The ATP content in GL261 cells was detected by ATP detection kit. Results Compared to the control group, the proliferation of GL261 cells in CTO treatment group was significantly inhibited in a time- and dose-dependent way, the proportion of GL261 cells in S phase was significantly increased (P＜0.05) in 20 μmol/L CTO group. Compared with the control group, more apoptosis occurred in GL261 cells of CTO group, and ROS content in GL261 cells of 20 μmol/L CTO group was also significantly increased (P＜0.01). Compared with the control group, the contents of OCR and ATP in GL261 cells was signifi- cantly reduced in the CTO group(P＜0.01). Conclusions CTO inhibits proliferation of GL261 cells and promotes their apoptosis by increased production of ROS. In addition, CTO inhibits the mitochondrial respiration of GL261 cells and the production of ATP in mitochondria, thus inhibits growth of glioma cells.

Published
2021

40. Carboxyamidotriazole enhances the effects of mouse spleen lymphocytes on promoting the apoptosis of colon cancer cell lines MC38 and C26

Author

GAO Hong-ting, HUANG Yu-qing, YANG Li-xing, MA Rui, WANG Yu-cheng, JU Rui, GUO Lei

Subjects
carboxyamidotriazole, spleen lymphocytes of mice, colon cancer, apoptosis, Medicine
Abstract

Objective To investigate the direct effects of carboxyamidotriazole(CAI)on the survival and apoptosis of colon cancer cells(MC38 and C26)and its indirect effects through enhancing the anticancer effects of spleen lymphocytes in mice. Methods MC38 and C26 cells were cultured in vitro,spleen lymphocytes of mice were isolated,MC38 and C26 cells were co-cultured with spleen lymphocytes of mice,and the cells were treated with different concentrations of CAI. CCK-8 method was used to detect cell survival,flow cytometry was used to detect cell apoptosis and the proportion of CD4+CD3+ T cells and CD8+CD3+ T cells in spleen lymphocytes of mice. Results The number of living cells of MC38 and C26 cells decreased(P＜0.05,P＜0.01 and P＜0.001)and the apoptosis rate was increased(P＜0.01 and P＜0.001)after being treated with different concentrations of CAI. The number of living cells of MC38 and C26 cells was decreased(P＜0.001)upon being co-cultured with spleen lymphocytes of mice for 48 h, and the apoptosis rate was increased(P＜0.001). The number of living cells of MC38 and C26 cells in co- culture groups treated with different concentrations of CAI for 48 h further decreased(P＜0.001)and the apoptosis rate increased(P＜0.05 and P＜0.001)as compared with those co-cultured with spleen lymphocytes of mice without CAI. The proportion of CD4+CD3+ T cells and CD8+CD3+ T cells in spleen lymphocytes of mice was significantly increased(P＜0.05 and P＜0.01)after 48 h treatment with different concentrations of CAI. Conclusions CAI can inhibit the survival and promote apoptosis of MC38 and C26 cells,and enhance the effects of spleen lymphocytes in mice on inhibiting the survival and promoting apoptosis of MC38 and C26 cells through a potential mechanism of increasing the proportion of CD4+CD3+ T cells and CD8+CD3+ T cells in spleen lymphocytes of mice.

Published
2021

47. Divergent geographic variation in above‐ versus below‐ground secondary metabolites of Reynoutria japonica.

Author

Bi, Jingwen, Bossdorf, Oliver, Liao, Zhiyong, Richards, Christina L., Parepa, Madalin, Zhao, Weihan, Berninger, Frank, Zhao, Yujie, Liu, Zekang, Feng, Xiangyan, Ju, Rui‐Ting, Li, Bo, and Wu, Jihua

Subjects
JAPANESE knotweed, METABOLITES, PLANT metabolites, PLANT adaptation, LEAF area, EMODIN
Abstract

Copyright of Journal of Ecology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Published
2024
Full Text
View/download PDF

49. Carboxyamidotriazole reduces programmed cell death 1 expression by inhibiting NFAT2 nuclear transport in murine CD8+T cells

Author

XU Meng-jiao, GAO Hong-ting, SHI Jing, JU Rui, YANG Li-xing, LI Jian-heng, GUO Lei

Subjects
carboxyamidotriazole, cd8+t cells, ca2+, activated t cell nuclear factor 2, pd-1, Medicine
Abstract

Objective To study the effect of carboxyamidotriazole (CAI) on CD8+T cells, and to explore the critical mechanism of CAI in reinforcing cytotoxicity of activated T cells. Methods CD8+T cells of mouse were isolated by immunomagnetic beads and divided into control group, CAI group, ZK756326 (Ca2+ activator) group and CAI+ZK756326 group. Flow cytometry was used to detect the level of intracellular free calcium ions in CD8+T cells; Enzyme-linked immunosorbent assay was used to detect the expression of intracellular calcineurin (CaN); Immunoflu- orescence staining was used to detect NFAT2 nuclear transportation; Chromatin immunoprecipitation-qPCR was used to detect the expression of PD-1 regulated by NFAT2. Mice spleen cytotoxic T lymphocytes (CTLs) were isolated and examined for the expression of PD-1 in CD8+T cells by flow cytometry. Results Intracellular Ca2+ concentration in CD8+T cells significantly decreased in the CAI group (P＜0.001), while intracellular Ca2+ concentration increased in the CAI+ZK756326 group(P＜0.01); CAI significantly decreased the content of calcineurin phosphatase in CD8+T cells (P＜0.001);CAI significantly inhibited NFAT2 nuclear transportation (P＜0.001) and significantly inhibited PD-1 transcription process depend on NFAT2 (P＜0.001); CAI significantly reduced the proportion of PD-1+CD8+T cells in mice spleen CTLs cells (P＜0.001). Conclusions CAI inhibits NFAT2 nuclear translocation by decreasing the level of calcium ions in the CD8+ T cells and inhibiting the expression of calcineurin, thereby reducing the expression of PD-1 in CD8+ T cells, which facilitated immunotherapy intervention.

Published
2020

50. Berberine activates macrophage autophagy and improves atherosclerotic plaque vulnerability

Author

LIU Qin, YANG Li-xing, SHI Jing, HUANG Yu-qing, GAO Hong-ting, JU Rui, GUO Lei

Subjects
berberine, atherosclerotic plaque vulnerability, autophagy, nlrp3, mif, Medicine
Abstract

Objective To explore the effect of berberine (BBR) on the vulnerability of atherosclerotic plaques caused by hyperglycemia and hyperlipidemia, and to discover the potential mechanism. Methods Male ApoE-/-mice were randomly divided into the control group, hyperglycemia group (intraperitoneal injection of streptozotocin(STZ), a dose of 45 mg/(kg·d) for 5 consecutive days), and hyperlipidemia group (high-fat diet, HFD). After two weeks, 10 mice were randomly selected from the hyperglycemia group to form a drug prevention group[gavage, BBR 50 mg/(kg·d)].Four weeks after modeling,the animals in the hyperglycemia model group were randomly divided into the model (STZ) group (gavage, solvent), BBR25, 50, and 100 (STZ) group (gavage, different doses of berberine [25, 50, and 100 mg/(kg·d)], the atorvastatin group [gavage, 2.5 mg/(kg·d)], the englitazone group [gavage, 1.25 mg/(kg·d)], and the hyperlipidemia model group were randomly divided into the model (HFD) group (gavage, solvent), BBR50 (HFD) group [gavage, berberine 50 mg/(kg·d)]. After 6 weeks of continuous treatment, the aorta of the mouse was taken for pathological and immunohistological examination to evaluate the atherosclerotic plaque lesions and their vulnerability index; Thp1 cells were stimulated to differentiate into macrophages with phorbol ester (PMA) and were used to investigate the effect of berberine on autophagy and inflammatory signaling pathways; Western blot was conducted to detect the expression of MIF, as well as NLRP3, AMPK/mTOR signal pathway and autophagy-related protein level in mouse aorta and cells.ELISA was used to detect the level of IL-1β in serum and cell supernatant. Results Compared with the control group, the atherosclerotic lesions in the hyperglycemia and hyperlipidemia groups were worsened, and the plaque vulnerability index was significantly increased. After the treatment with berberine, the plaque lesions were significantly improved, and the plaque vulnerability index decreased significantly(P＜0.05).Berberine down-regulated MIF protein expression in mouse aortic tissue homogenated and Thp1 cells, promoted AMPK activation, inhibited mTOR phosphorylation, regulated autophagy-related protein expression (LC3Ⅱ/LC3Ⅰ ratio increased, SQSTM1/P62 decreased), and down-regulated NLRP3 and IL-1β expression. Conclusions Berberine can effectively inhibit formation of atherosclerotic plaque and its vulnerability in ApoE-/-mice caused by hyperglycemia and hyperlipidemia. The mechanism may be attributed to the down-regulation of MIF protein expression and the promotion of autophagy of macrophages through the AMPK/mTOR signaling pathway to inhibit the inflammatory response.

Published
2020

Catalog

Books, media, physical & digital resources
See catalog results