Search

Your search keyword '"Yu, Chong"' showing total 2,786 results
Start Over Author "Yu, Chong" Publication Year Range Last 10 years
2,786 results on '"Yu, Chong"'

2. Communication-Efficient Hybrid Federated Learning for E-health with Horizontal and Vertical Data Partitioning

Author

Yu, Chong, Shen, Shuaiqi, Wang, Shiqiang, Zhang, Kuan, and Zhao, Hai

Subjects
Computer Science - Machine Learning, Computer Science - Distributed, Parallel, and Cluster Computing
Abstract

E-health allows smart devices and medical institutions to collaboratively collect patients' data, which is trained by Artificial Intelligence (AI) technologies to help doctors make diagnosis. By allowing multiple devices to train models collaboratively, federated learning is a promising solution to address the communication and privacy issues in e-health. However, applying federated learning in e-health faces many challenges. First, medical data is both horizontally and vertically partitioned. Since single Horizontal Federated Learning (HFL) or Vertical Federated Learning (VFL) techniques cannot deal with both types of data partitioning, directly applying them may consume excessive communication cost due to transmitting a part of raw data when requiring high modeling accuracy. Second, a naive combination of HFL and VFL has limitations including low training efficiency, unsound convergence analysis, and lack of parameter tuning strategies. In this paper, we provide a thorough study on an effective integration of HFL and VFL, to achieve communication efficiency and overcome the above limitations when data is both horizontally and vertically partitioned. Specifically, we propose a hybrid federated learning framework with one intermediate result exchange and two aggregation phases. Based on this framework, we develop a Hybrid Stochastic Gradient Descent (HSGD) algorithm to train models. Then, we theoretically analyze the convergence upper bound of the proposed algorithm. Using the convergence results, we design adaptive strategies to adjust the training parameters and shrink the size of transmitted data. Experimental results validate that the proposed HSGD algorithm can achieve the desired accuracy while reducing communication cost, and they also verify the effectiveness of the adaptive strategies.

Published
2024

3. Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression

Author

Ye, Hancheng, Yu, Chong, Ye, Peng, Xia, Renqiu, Tang, Yansong, Lu, Jiwen, Chen, Tao, and Zhang, Bo

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Recent Vision Transformer Compression (VTC) works mainly follow a two-stage scheme, where the importance score of each model unit is first evaluated or preset in each submodule, followed by the sparsity score evaluation according to the target sparsity constraint. Such a separate evaluation process induces the gap between importance and sparsity score distributions, thus causing high search costs for VTC. In this work, for the first time, we investigate how to integrate the evaluations of importance and sparsity scores into a single stage, searching the optimal subnets in an efficient manner. Specifically, we present OFB, a cost-efficient approach that simultaneously evaluates both importance and sparsity scores, termed Once for Both (OFB), for VTC. First, a bi-mask scheme is developed by entangling the importance score and the differentiable sparsity score to jointly determine the pruning potential (prunability) of each unit. Such a bi-mask search strategy is further used together with a proposed adaptive one-hot loss to realize the progressive-and-efficient search for the most important subnet. Finally, Progressive Masked Image Modeling (PMIM) is proposed to regularize the feature space to be more representative during the search process, which may be degraded by the dimension reduction. Extensive experiments demonstrate that OFB can achieve superior compression performance over state-of-the-art searching-based and pruning-based methods under various Vision Transformer architectures, meanwhile promoting search efficiency significantly, e.g., costing one GPU search day for the compression of DeiT-S on ImageNet-1K., Comment: Accepted by CVPR 2024. Our code will be available at www.github.com/HankYe/Once-for-Both

Published
2024

4. Enhanced Sparsification via Stimulative Training

Author

Tang, Shengji, Lin, Weihao, Ye, Hancheng, Ye, Peng, Yu, Chong, Li, Baopu, and Chen, Tao

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Sparsification-based pruning has been an important category in model compression. Existing methods commonly set sparsity-inducing penalty terms to suppress the importance of dropped weights, which is regarded as the suppressed sparsification paradigm. However, this paradigm inactivates the dropped parts of networks causing capacity damage before pruning, thereby leading to performance degradation. To alleviate this issue, we first study and reveal the relative sparsity effect in emerging stimulative training and then propose a structured pruning framework, named STP, based on an enhanced sparsification paradigm which maintains the magnitude of dropped weights and enhances the expressivity of kept weights by self-distillation. Besides, to find an optimal architecture for the pruned network, we propose a multi-dimension architecture space and a knowledge distillation-guided exploration strategy. To reduce the huge capacity gap of distillation, we propose a subnet mutating expansion technique. Extensive experiments on various benchmarks indicate the effectiveness of STP. Specifically, without fine-tuning, our method consistently achieves superior performance at different budgets, especially under extremely aggressive pruning scenarios, e.g., remaining 95.11% Top-1 accuracy (72.43% in 76.15%) while reducing 85% FLOPs for ResNet-50 on ImageNet. Codes will be released soon., Comment: 26 pages

Published
2024

5. MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer

Author

Cao, Jianjian, Ye, Peng, Li, Shengze, Yu, Chong, Tang, Yansong, Lu, Jiwen, and Chen, Tao

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Vision-Language Transformers (VLTs) have shown great success recently, but are meanwhile accompanied by heavy computation costs, where a major reason can be attributed to the large number of visual and language tokens. Existing token pruning research for compressing VLTs mainly follows a single-modality-based scheme yet ignores the critical role of aligning different modalities for guiding the token pruning process, causing the important tokens for one modality to be falsely pruned in another modality branch. Meanwhile, existing VLT pruning works also lack the flexibility to dynamically compress each layer based on different input samples. To this end, we propose a novel framework named Multimodal Alignment-Guided Dynamic Token Pruning (MADTP) for accelerating various VLTs. Specifically, we first introduce a well-designed Multi-modality Alignment Guidance (MAG) module that can align features of the same semantic concept from different modalities, to ensure the pruned tokens are less important for all modalities. We further design a novel Dynamic Token Pruning (DTP) module, which can adaptively adjust the token compression ratio in each layer based on different input instances. Extensive experiments on various benchmarks demonstrate that MADTP significantly reduces the computational complexity of kinds of multimodal models while preserving competitive performance. Notably, when applied to the BLIP model in the NLVR2 dataset, MADTP can reduce the GFLOPs by 80% with less than 4% performance degradation., Comment: 19 pages, 9 figures, Published in CVPR2024

Published
2024

6. A novel approach to determine the critical survival threshold of cellular oxygen within spheroids via integrating live/dead cell imaging with oxygen modeling

Author

Shang, Kuang-Ming, Kato, Hiroyuki, Gonzalez, Nelson, Kandeel, Fouad, Tai, Yu-Chong, and Komatsu, Hirotake

Subjects
Biochemistry and Cell Biology, Medical Physiology, Biomedical and Clinical Sciences, Biological Sciences, Transplantation, Cell survival, Computational simulation, Hypoxia, Pancreatic islets, Viability assay, Physiology, Biochemistry and cell biology, Medical physiology
Abstract

Hypoxia plays a crucial role in cell physiology. Defining the oxygen level that induces cell death within 3D tissues is vital for understanding tissue hypoxia; however, obtaining accurate measurements has been technically challenging. In this study, we introduce a non-invasive, high-throughput methodology to quantify critical survival partial oxygen pressure (pO₂) with high spatial resolution within spheroids by employing a combination of controlled hypoxic conditions, semi-automated live/dead cell imaging, and computational oxygen modeling. The oxygen-permeable, micro-pyramid patterned culture plates created a precisely controlled oxygen condition around the individual spheroid. Live/dead cell imaging provided the geometric information of the live/dead boundary within spheroids. Finally, computational oxygen modeling calculated the pO₂ at the live/dead boundary within spheroids. As proof of concept, we determined the critical survival pO₂ in two types of spheroids: isolated primary pancreatic islets and tumor-derived pseudo-islets (2.43 ± 0.08 vs. 0.84 ± 0.04 mmHg), indicating higher hypoxia tolerance in pseudo-islets due to their tumorigenic origin. We also applied this method for evaluating graft survival in cell transplantations for diabetes therapy, where hypoxia is a critical barrier to successful transplantation outcomes; thus, designing oxygenation strategies is required. Based on the elucidated critical survival pO₂, 100% viability could be maintained in a typically sized primary islet under the tissue pO₂ above 14.5 mmHg. This work presents a valuable tool that is potentially instrumental for fundamental hypoxia research. It offers insights into physiological responses to hypoxia among different cell types and may refine translational research in cell therapies.

Published
2024

7. Efficient Architecture Search via Bi-level Data Pruning

Author

Tu, Chongjun, Ye, Peng, Lin, Weihao, Ye, Hancheng, Yu, Chong, Chen, Tao, Li, Baopu, and Ouyang, Wanli

Subjects
Computer Science - Computer Vision and Pattern Recognition, 68T05(Primary)
Abstract

Improving the efficiency of Neural Architecture Search (NAS) is a challenging but significant task that has received much attention. Previous works mainly adopted the Differentiable Architecture Search (DARTS) and improved its search strategies or modules to enhance search efficiency. Recently, some methods have started considering data reduction for speedup, but they are not tightly coupled with the architecture search process, resulting in sub-optimal performance. To this end, this work pioneers an exploration into the critical role of dataset characteristics for DARTS bi-level optimization, and then proposes a novel Bi-level Data Pruning (BDP) paradigm that targets the weights and architecture levels of DARTS to enhance efficiency from a data perspective. Specifically, we introduce a new progressive data pruning strategy that utilizes supernet prediction dynamics as the metric, to gradually prune unsuitable samples for DARTS during the search. An effective automatic class balance constraint is also integrated into BDP, to suppress potential class imbalances resulting from data-efficient algorithms. Comprehensive evaluations on the NAS-Bench-201 search space, DARTS search space, and MobileNet-like search space validate that BDP reduces search costs by over 50% while achieving superior performance when applied to baseline DARTS. Besides, we demonstrate that BDP can harmoniously integrate with advanced DARTS variants, like PC-DARTS and \b{eta}-DARTS, offering an approximately 2 times speedup with minimal performance compromises., Comment: 11 pages

Published
2023

10. SpVOS: Efficient Video Object Segmentation with Triple Sparse Convolution

Author

Lin, Weihao, Chen, Tao, and Yu, Chong

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Semi-supervised video object segmentation (Semi-VOS), which requires only annotating the first frame of a video to segment future frames, has received increased attention recently. Among existing pipelines, the memory-matching-based one is becoming the main research stream, as it can fully utilize the temporal sequence information to obtain high-quality segmentation results. Even though this type of method has achieved promising performance, the overall framework still suffers from heavy computation overhead, mainly caused by the per-frame dense convolution operations between high-resolution feature maps and each kernel filter. Therefore, we propose a sparse baseline of VOS named SpVOS in this work, which develops a novel triple sparse convolution to reduce the computation costs of the overall VOS framework. The designed triple gate, taking full consideration of both spatial and temporal redundancy between adjacent video frames, adaptively makes a triple decision to decide how to apply the sparse convolution on each pixel to control the computation overhead of each layer, while maintaining sufficient discrimination capability to distinguish similar objects and avoid error accumulation. A mixed sparse training strategy, coupled with a designed objective considering the sparsity constraint, is also developed to balance the VOS segmentation performance and computation costs. Experiments are conducted on two mainstream VOS datasets, including DAVIS and Youtube-VOS. Results show that, the proposed SpVOS achieves superior performance over other state-of-the-art sparse methods, and even maintains comparable performance, e.g., an 83.04% (79.29%) overall score on the DAVIS-2017 (Youtube-VOS) validation set, with the typical non-sparse VOS baseline (82.88% for DAVIS-2017 and 80.36% for Youtube-VOS) while saving up to 42% FLOPs, showing its application potential for resource-constrained scenarios., Comment: 15 pages, 6 figures

Published
2023
Full Text
View/download PDF

11. Boosting Residual Networks with Group Knowledge

Author

Tang, Shengji, Ye, Peng, Li, Baopu, Lin, Weihao, Chen, Tao, He, Tong, Yu, Chong, and Ouyang, Wanli

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Recent research understands the residual networks from a new perspective of the implicit ensemble model. From this view, previous methods such as stochastic depth and stimulative training have further improved the performance of the residual network by sampling and training of its subnets. However, they both use the same supervision for all subnets of different capacities and neglect the valuable knowledge generated by subnets during training. In this manuscript, we mitigate the significant knowledge distillation gap caused by using the same kind of supervision and advocate leveraging the subnets to provide diverse knowledge. Based on this motivation, we propose a group knowledge based training framework for boosting the performance of residual networks. Specifically, we implicitly divide all subnets into hierarchical groups by subnet-in-subnet sampling, aggregate the knowledge of different subnets in each group during training, and exploit upper-level group knowledge to supervise lower-level subnet groups. Meanwhile, We also develop a subnet sampling strategy that naturally samples larger subnets, which are found to be more helpful than smaller subnets in boosting performance for hierarchical groups. Compared with typical subnet training and other methods, our method achieves the best efficiency and performance trade-offs on multiple datasets and network structures. The code is at https://github.com/tsj-001/AAAI24-GKT., Comment: Accepted by AAAI2024

Published
2023

14. Adversarial Amendment is the Only Force Capable of Transforming an Enemy into a Friend

Author

Yu, Chong, Chen, Tao, and Gan, Zhongxue

Subjects
Computer Science - Artificial Intelligence, Computer Science - Cryptography and Security, Computer Science - Computer Vision and Pattern Recognition
Abstract

Adversarial attack is commonly regarded as a huge threat to neural networks because of misleading behavior. This paper presents an opposite perspective: adversarial attacks can be harnessed to improve neural models if amended correctly. Unlike traditional adversarial defense or adversarial training schemes that aim to improve the adversarial robustness, the proposed adversarial amendment (AdvAmd) method aims to improve the original accuracy level of neural models on benign samples. We thoroughly analyze the distribution mismatch between the benign and adversarial samples. This distribution mismatch and the mutual learning mechanism with the same learning ratio applied in prior art defense strategies is the main cause leading the accuracy degradation for benign samples. The proposed AdvAmd is demonstrated to steadily heal the accuracy degradation and even leads to a certain accuracy boost of common neural models on benign classification, object detection, and segmentation tasks. The efficacy of the AdvAmd is contributed by three key components: mediate samples (to reduce the influence of distribution mismatch with a fine-grained amendment), auxiliary batch norm (to solve the mutual learning mechanism and the smoother judgment surface), and AdvAmd loss (to adjust the learning ratios according to different attack vulnerabilities) through quantitative and ablation experiments., Comment: Accepted to IJCAI 2023, 10 pages, 5 figures

Published
2023

15. Boost Vision Transformer with GPU-Friendly Sparsity and Quantization

Author

Yu, Chong, Chen, Tao, Gan, Zhongxue, and Fan, Jiayuan

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Computer Science - Performance
Abstract

The transformer extends its success from the language to the vision domain. Because of the stacked self-attention and cross-attention blocks, the acceleration deployment of vision transformer on GPU hardware is challenging and also rarely studied. This paper thoroughly designs a compression scheme to maximally utilize the GPU-friendly 2:4 fine-grained structured sparsity and quantization. Specially, an original large model with dense weight parameters is first pruned into a sparse one by 2:4 structured pruning, which considers the GPU's acceleration of 2:4 structured sparse pattern with FP16 data type, then the floating-point sparse model is further quantized into a fixed-point one by sparse-distillation-aware quantization aware training, which considers GPU can provide an extra speedup of 2:4 sparse calculation with integer tensors. A mixed-strategy knowledge distillation is used during the pruning and quantization process. The proposed compression scheme is flexible to support supervised and unsupervised learning styles. Experiment results show GPUSQ-ViT scheme achieves state-of-the-art compression by reducing vision transformer models 6.4-12.7 times on model size and 30.3-62 times on FLOPs with negligible accuracy degradation on ImageNet classification, COCO detection and ADE20K segmentation benchmarking tasks. Moreover, GPUSQ-ViT can boost actual deployment performance by 1.39-1.79 times and 3.22-3.43 times of latency and throughput on A100 GPU, and 1.57-1.69 times and 2.11-2.51 times improvement of latency and throughput on AGX Orin., Comment: Accepted to CVPR 2023, 11 pages, 6 figures

Published
2023

16. New evidence: Metformin unsuitable as routine adjuvant for breast cancer: a drug-target mendelian randomization analysis

Author

Jing-Xuan Xu, Qi-Long Zhu, Yu-Miao Bi, and Yu-Chong Peng

Subjects
Breast cancer, Metformin, Causal relationship, Drug-target mendelian randomization, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Abstract Purpose The potential efficacy of metformin in breast cancer (BC) has been hotly discussed but never conclusive. This genetics-based study aimed to evaluate the relationships between metformin targets and BC risk. Methods Metformin targets from DrugBank and genome-wide association study (GWAS) data from IEU OpenGWAS and FinnGen were used to investigate the breast cancer (BC)-metformin causal link with various Mendelian Randomization (MR) methods (e.g., inverse-variance-weighting). The genetic association between type 2 diabetes (T2D) and the drug target of metformin was also analyzed as a positive control. Sensitivity and pleiotropic tests ensured reliability. Results The primary targets of metformin are PRKAB1, ETFDH and GPD1L. We found a causal association between PRKAB1 and T2D (odds ratio [OR] 0.959, P = 0.002), but no causal relationship was observed between metformin targets and overall BC risk (PRKAB1: OR 0.990, P = 0.530; ETFDH: OR 0.986, P = 0.592; GPD1L: OR 1.002, P = 0.806). A noteworthy causal relationship was observed between ETFDH and estrogen receptor (ER)-positive BC (OR 0.867, P = 0.018), and between GPD1L and human epidermal growth factor receptor 2 (HER2)-negative BC (OR 0.966, P = 0.040). Other group analyses did not yield positive results. Conclusion The star target of metformin, PRKAB1, does not exhibit a substantial causal association with the risk of BC. Conversely, metformin, acting as an inhibitor of ETFDH and GPD1L, may potentially elevate the likelihood of developing ER-positive BC and HER2-negative BC. Consequently, it is not advisable to employ metformin as a standard supplementary therapy for BC patients without T2D.

Published
2024
Full Text
View/download PDF

19. Aldo-keto reductase family 1 member A1 (AKR1A1) exerts a protective function in alcohol-associated liver disease by reducing 4-HNE accumulation and p53 activation

Author

Ying-Wei Lan, Wan-Ru Chen, Gary Ro-Lin Chang, Ying-Cheng Chen, Kowit-Yu Chong, Kai-Cheng Chuang, Yung-Tsung Kao, Ming-Shan Chen, and Chuan-Mu Chen

Subjects
Alcohol-associated liver disease (ALD), Aldo-keto reductase family 1 member A1 (AKR1A1), 4-HNE, p53, Lipid accumulation, Liver fibrosis, Biotechnology, TP248.13-248.65, Biology (General), QH301-705.5, Biochemistry, QD415-436
Abstract

Abstract Background The development of alcohol-associated liver disease (ALD) is influenced by the amount and duration of alcohol consumption. The resulting liver damage can range from reversible stages, such as steatosis, steatohepatitis and alcoholic fibrosis, to the advanced and irreversible stage of cirrhosis. Aldo-keto reductase family 1 member A1 (AKR1A1) is a member of the aldo-keto reductase family that catalyzes the reduction of aldehyde groups to their corresponding alcohols in an NADPH-dependent manner. AKR1A1 was found to be downregulated in patients diagnosed with ALD. This study aims to interpret the protective effects of AKR1A1 on the development of ALD. Methods A 5% alcohol-fed (AF) Akr1a1 knockout (Akr1a1 −/−) mouse model and an AML12 hepatocyte model were used. The effects of AKR1A1 on liver function, inflammation, oxidative stress, lipid accumulation, and fibrosis were assessed by ELISA, western blotting, RT‒PCR, and a variety of histological staining methods in AF-induced wild-type (WT) and Akr1a1 −/− mice compared to control liquid diet-fed (PF) WT and Akr1a1 −/− mice. Results The results demonstrated that AF-WT mice expressed higher levels of AKR1A1 than WT mice fed a control diet, and they did not show any noticeable liver steatosis. However, AF-Akr1a1 −/− mice displayed a lower survival rate and more severe liver injury than AF-WT mice, as demonstrated by increased proinflammatory cytokines, oxidative stress, lipid accumulation, fibrosis, and reduced antioxidant enzymes in their livers. Additionally, elevated levels of 4-HNE and p53 phosphorylation were observed in AF-Akr1a1 −/− mice, suggesting that the loss of AKR1A1 led to increased 4-HNE accumulation and subsequent activation of p53, which contributed to the progression of ALD. Furthermore, in AML12 hepatocytes, Akr1a1 knockdown aggravated oxidative stress and steatosis induced by palmitic acid/oleic acid (P/O) inflammation induced by lipopolysaccharide (LPS), and fibrosis induced by TGF-β1. Conclusions This loss-of-function study suggests that AKR1A1 plays a liver-protective role during chronic alcohol consumption by reducing the accumulation of 4-HNE and inhibiting 4-HNE-mediated p53 activation.

Published
2024
Full Text
View/download PDF

21. Physiomimetic Fluidic Culture Platform on Microwell-Patterned Porous Collagen Scaffold for Human Pancreatic Islets

Author

Hiroyuki Kato, Huajian Chen, Kuang-Ming Shang, Kenji Izumi, Naoya Koba, Takanori Tsuchiya, Naoki Kawazoe, Janine Quijano, Keiko Omori, Chris Orr, Meirigeng Qi, Hsun Teresa Ku, Fouad Kandeel, Yu-Chong Tai, Guoping Chen, and Hirotake Komatsu

Subjects
Medicine
Abstract

Pancreatic islet transplantation is one of the clinical options for certain types of diabetes. However, difficulty in maintaining islets prior to transplantation limits the clinical expansion of islet transplantations. Our study introduces a dynamic culture platform developed specifically for primary human islets by mimicking the physiological microenvironment, including tissue fluidics and extracellular matrix support. We engineered the dynamic culture system by incorporating our distinctive microwell-patterned porous collagen scaffolds for loading isolated human islets, enabling vertical medium flow through the scaffolds. The dynamic culture system featured four 12 mm diameter islet culture chambers, each capable of accommodating 500 islet equivalents (IEQ) per chamber. This configuration calculates > five-fold higher seeding density than the conventional islet culture in flasks prior to the clinical transplantations (442 vs 86 IEQ/cm 2 ). We tested our culture platform with three separate batches of human islets isolated from deceased donors for an extended period of 2 weeks, exceeding the limits of conventional culture methods for preserving islet quality. Static cultures served as controls. The computational simulation revealed that the dynamic culture reduced the islet volume exposed to the lethal hypoxia (< 10 mmHg) to ~1/3 of the static culture. Dynamic culture ameliorated the morphological islet degradation in long-term culture and maintained islet viability, with reduced expressions of hypoxia markers. Furthermore, dynamic culture maintained the islet metabolism and insulin-secreting function over static culture in a long-term culture. Collectively, the physiological microenvironment-mimetic culture platform supported the viability and quality of isolated human islets at high-seeding density. Such a platform has a high potential for broad applications in cell therapies and tissue engineering, including extended islet culture prior to clinical islet transplantations and extended culture of stem cell-derived islets for maturation.

Published
2024
Full Text
View/download PDF

22. Micropyramid-patterned, oxygen-permeable bottomed dish for high density culture of pancreatic islets

Author

Myrick, Ryan J, Shang, Kuang-Ming, Betts, Jonathan F, Gonzalez, Nelson, Rawson, Jeffrey, Izumi, Kenji, Koba, Naoya, Tsuchiya, Takanori, Kato, Hiroyuki, Omori, Keiko, Kandeel, Fouad, Mullen, Yoko, Tai, Yu-Chong, Botvinick, Elliot, and Komatsu, Hirotake

Subjects
Engineering, Biomedical Engineering, Digestive Diseases, Autoimmune Disease, Transplantation, Diabetes, Metabolic and endocrine, Mice, Animals, Oxygen, Diabetes Mellitus, Experimental, Islets of Langerhans, Islets of Langerhans Transplantation, Hypoxia, spheroid culture, micropatterned dish, pancreatic islets, 3D spheroid, high seeding density culture, Medical Biotechnology, Other Technology, Medical biotechnology, Biomedical engineering
Abstract

The need for maintaining cell-spheroid viability and function within high-density cultures is unmet for various clinical and experimental applications, including cell therapies. One immediate application is for transplantation of pancreatic islets, a clinically recognized treatment option to cure type 1 diabetes; islets are isolated from a donor for subsequent culture prior to transplantation. However, high seeding conditions cause unsolicited fusion of multiple spheroids, thereby limiting oxygen diffusion to induce hypoxic cell death. Here we introduce a culture dish incorporating a micropyramid-patterned surface to prevent the unsolicited fusion and oxygen-permeable bottom for optimal oxygen environment. A 400µm-thick, oxygen-permeable polydimethylsiloxane sheet topped with micropyramid pattern of 400µm-base and 200µm-height was fabricated to apply to the 24-well plate format. The micropyramid pattern separated the individual pancreatic islets to prevent the fusion of multiple islets. This platform supported the high oxygen demand of islets at high seeding density at 260 islet equivalents cm-2, a 2-3-fold higher seeding density compared to the conventional islet culture used in a preparation for the clinical islet transplantations, demonstrating improved islet morphology, metabolism and function in a 4 d-culture. Transplantation of these islets into immunodeficient diabetic mice exhibited significantly improved engraftment to achieve euglycemia compared to islets cultured in the conventional culture wells. Collectively, this simple design modification allows for high-density cultures of three-dimensional cell spheroids to improve the viability and function for an array of investigational and clinical replacement tissues.

Published
2023

26. Genome-wide association study of lung adenocarcinoma in East Asia and comparison with a European population

Author

Shi, Jianxin, Shiraishi, Kouya, Choi, Jiyeon, Matsuo, Keitaro, Chen, Tzu-Yu, Dai, Juncheng, Hung, Rayjean J., Chen, Kexin, Shu, Xiao-Ou, Kim, Young Tae, Landi, Maria Teresa, Lin, Dongxin, Zheng, Wei, Yin, Zhihua, Zhou, Baosen, Song, Bao, Wang, Jiucun, Seow, Wei Jie, Song, Lei, Chang, I-Shou, Hu, Wei, Chien, Li-Hsin, Cai, Qiuyin, Hong, Yun-Chul, Kim, Hee Nam, Wu, Yi-Long, Wong, Maria Pik, Richardson, Brian Douglas, Funderburk, Karen M., Li, Shilan, Zhang, Tongwu, Breeze, Charles, Wang, Zhaoming, Blechter, Batel, Bassig, Bryan A., Kim, Jin Hee, Albanes, Demetrius, Wong, Jason Y. Y., Shin, Min-Ho, Chung, Lap Ping, Yang, Yang, An, She-Juan, Zheng, Hong, Yatabe, Yasushi, Zhang, Xu-Chao, Kim, Young-Chul, Caporaso, Neil E., Chang, Jiang, Ho, James Chung Man, Kubo, Michiaki, Daigo, Yataro, Song, Minsun, Momozawa, Yukihide, Kamatani, Yoichiro, Kobayashi, Masashi, Okubo, Kenichi, Honda, Takayuki, Hosgood, Dean H., Kunitoh, Hideo, Patel, Harsh, Watanabe, Shun-ichi, Miyagi, Yohei, Nakayama, Haruhiko, Matsumoto, Shingo, Horinouchi, Hidehito, Tsuboi, Masahiro, Hamamoto, Ryuji, Goto, Koichi, Ohe, Yuichiro, Takahashi, Atsushi, Goto, Akiteru, Minamiya, Yoshihiro, Hara, Megumi, Nishida, Yuichiro, Takeuchi, Kenji, Wakai, Kenji, Matsuda, Koichi, Murakami, Yoshinori, Shimizu, Kimihiro, Suzuki, Hiroyuki, Saito, Motonobu, Ohtaki, Yoichi, Tanaka, Kazumi, Wu, Tangchun, Wei, Fusheng, Dai, Hongji, Machiela, Mitchell J., Su, Jian, Kim, Yeul Hong, Oh, In-Jae, Lee, Victor Ho Fun, Chang, Gee-Chen, Tsai, Ying-Huang, Chen, Kuan-Yu, Huang, Ming-Shyan, Su, Wu-Chou, Chen, Yuh-Min, Seow, Adeline, Park, Jae Yong, Kweon, Sun-Seog, Chen, Kun-Chieh, Gao, Yu-Tang, Qian, Biyun, Wu, Chen, Lu, Daru, Liu, Jianjun, Schwartz, Ann G., Houlston, Richard, Spitz, Margaret R., Gorlov, Ivan P., Wu, Xifeng, Yang, Ping, Lam, Stephen, Tardon, Adonina, Chen, Chu, Bojesen, Stig E., Johansson, Mattias, Risch, Angela, Bickeböller, Heike, Ji, Bu-Tian, Wichmann, H-Erich, Christiani, David C., Rennert, Gadi, Arnold, Susanne, Brennan, Paul, McKay, James, Field, John K., Shete, Sanjay S., Le Marchand, Loic, Liu, Geoffrey, Andrew, Angeline, Kiemeney, Lambertus A., Zienolddiny-Narui, Shan, Grankvist, Kjell, Johansson, Mikael, Cox, Angela, Taylor, Fiona, Yuan, Jian-Min, Lazarus, Philip, Schabath, Matthew B., Aldrich, Melinda C., Jeon, Hyo-Sung, Jiang, Shih Sheng, Sung, Jae Sook, Chen, Chung-Hsing, Hsiao, Chin-Fu, Jung, Yoo Jin, Guo, Huan, Hu, Zhibin, Burdett, Laurie, Yeager, Meredith, Hutchinson, Amy, Hicks, Belynda, Liu, Jia, Zhu, Bin, Berndt, Sonja I., Wu, Wei, Wang, Junwen, Li, Yuqing, Choi, Jin Eun, Park, Kyong Hwa, Sung, Sook Whan, Liu, Li, Kang, Chang Hyun, Wang, Wen-Chang, Xu, Jun, Guan, Peng, Tan, Wen, Yu, Chong-Jen, Yang, Gong, Sihoe, Alan Dart Loon, Chen, Ying, Choi, Yi Young, Kim, Jun Suk, Yoon, Ho-Il, Park, In Kyu, Xu, Ping, He, Qincheng, Wang, Chih-Liang, Hung, Hsiao-Han, Vermeulen, Roel C. H., Cheng, Iona, Wu, Junjie, Lim, Wei-Yen, Tsai, Fang-Yu, Chan, John K. C., Li, Jihua, Chen, Hongyan, Lin, Hsien-Chih, Jin, Li, Liu, Jie, Sawada, Norie, Yamaji, Taiki, Wyatt, Kathleen, Li, Shengchao A., Ma, Hongxia, Zhu, Meng, Wang, Zhehai, Cheng, Sensen, Li, Xuelian, Ren, Yangwu, Chao, Ann, Iwasaki, Motoki, Zhu, Junjie, Jiang, Gening, Fei, Ke, Wu, Guoping, Chen, Chih-Yi, Chen, Chien-Jen, Yang, Pan-Chyr, Yu, Jinming, Stevens, Victoria L., Fraumeni, Jr, Joseph F., Chatterjee, Nilanjan, Gorlova, Olga Y., Hsiung, Chao Agnes, Amos, Christopher I., Shen, Hongbing, Chanock, Stephen J., Rothman, Nathaniel, Kohno, Takashi, and Lan, Qing

Published
2023
Full Text
View/download PDF

29. Reliability and validity of the translated Chinese version of comprehensive assessment of acceptance and commitment therapy processes (CompACT-C) in breast cancer survivors

Author

Wenqian Zhao, Yuen Yu Chong, Shan Yang, Dilihumaer Kuerban, Wei Zhang, Xiao Wang, Xiaomei Li, and Wai Tong Chien

Subjects
Breast cancer, Psychological flexibility, CompACT, Instrument validation, Science (General), Q1-390, Social sciences (General), H1-99
Abstract

Purpose: This study aimed to translate and validate a Chinese version of the Comprehensive assessment of Acceptance and Commitment Therapy processes (CompACT-C), a 23-item questionnaire for measuring an individual's level of psychological flexibility among Chinese breast cancer survivors for utilisation of Acceptance and commitment therapy (ACT)-based interventions in breast cancer patients. Methods: Six translators translated the original English version into Chinese according to Brislin's Translation Model. Psychometric properties tests were conducted on the CompACT-C, including internal consistency, test–retest reliability (two-week interval), face validity, content validity, convergent validity with experiential avoidance, and construct validity (confirmatory factor analysis). Results: 308 Chinese breast cancer survivors with mastectomies were conveniently recruited. The translated scale showed satisfactory test–retest reliability (two-week interval, r = 0.53–0.72, p

Published
2024
Full Text
View/download PDF

34. Radiotherapy combined with nano-biomaterials for cancer radio-immunotherapy

Author

Qingrong Dong, Tingyu Xue, Haili Yan, Fang Liu, Ruixue Liu, Kun Zhang, Yu Chong, Jiangfeng Du, and Hui Zhang

Subjects
Radiotherapy, Immunotherapy, Nano-biomaterials, Radio-immunotherapy, Tumor, Biotechnology, TP248.13-248.65, Medical technology, R855-855.5
Abstract

Abstract Radiotherapy (RT) plays an important role in tumor therapy due to its noninvasiveness and wide adaptation. In recent years, radiation therapy has been discovered to induce an anti-tumor immune response, which arouses widespread concern among scientists and clinicians. In this review, we highlight recent advances in the applications of nano-biomaterials for radiotherapy-activated immunotherapy. We first discuss the combination of different radiosensitizing nano-biomaterials and immune checkpoint inhibitors to enhance tumor immune response and improve radiotherapy efficacy. Subsequently, various nano-biomaterials-enabled tumor oxygenation strategies are introduced to alleviate the hypoxic tumor environment and amplify the immunomodulatory effect. With the aid of nano-vaccines and adjuvants, radiotherapy refreshes the host's immune system. Additionally, ionizing radiation responsive nano-biomaterials raise innate immunity-mediated anti-tumor immunity. At last, we summarize the rapid development of immune modulatable nano-biomaterials and discuss the key challenge in the development of nano-biomaterials for tumor radio-immunotherapy. Understanding the nano-biomaterials-assisted radio-immunotherapy will maximize the benefits of clinical radiotherapy and immunotherapy and facilitate the development of new combinational therapy modality. Graphical Abstract

Published
2023
Full Text
View/download PDF

35. An Ex Vivo Study of Outward Electrical Impedance Tomography (OEIT) for Intravascular Imaging

Author

Luo, Yuan, Huang, Dong, Huang, Zi-Yu, Hsiai, Tzung K, and Tai, Yu-Chong

Subjects
Engineering, Biomedical Engineering, Heart Disease, Atherosclerosis, Biomedical Imaging, Heart Disease - Coronary Heart Disease, Bioengineering, Cardiovascular, 4.2 Evaluation of markers and technologies, Detection, screening and diagnosis, Animals, Electric Impedance, Phantoms, Imaging, Swine, Tomography, Tomography, X-Ray Computed, Imaging, Electrical impedance tomography, Conductivity, Lesions, Electrodes, Solid modeling, electrical impedance tomography, intravascular imaging, intravascular navigation, Artificial Intelligence and Image Processing, Electrical and Electronic Engineering, Biomedical engineering, Electronics, sensors and digital hardware, Computer vision and multimedia computation
Abstract

ObjectiveAtherosclerosis is a chronic immuno-inflammatory condition emerging in arteries and considered the cause of a myriad of cardiovascular diseases. Atherosclerotic lesion characterization through invasive imaging modalities is essential in disease evaluation and determining intervention strategy. Recently, electrical properties of the lesions have been utilized in assessing its vulnerability mainly owing to its capability to differentiate lipid content existing in the lesion, albeit with limited detection resolution. Electrical impedance tomography is the natural extension of conventional spectrometric measurement by incorporating larger number of interrogating electrodes and advanced algorithm to achieve imaging of target objects and thus provides significantly richer information. It is within this context that we develop Outward Electrical Impedance Tomography (OEIT), aimed at intravascular imaging for atherosclerotic lesion characterization.MethodsWe utilized flexible electronics to establish the 32-electrode OEIT device with outward facing configuration suitable for imaging of vessels. We conducted comprehensive studies through simulation model and ex vivo setup to demonstrate the functionality of OEIT.ResultsQuantitative characterization for OEIT regarding its proximity sensing and conductivity differentiation was achieved using well-controlled experimental conditions. Imaging capability for OEIT was further verified with phantom setup using porcine aorta to emulate in vivo environment.ConclusionWe have successfully demonstrated a novel tool for intravascular imaging, OEIT, with unique advantages for atherosclerosis detection.SignificanceThis study demonstrates for the first time a novel electrical tomography-based platform for intravascular imaging, and we believe it paves the way for further adaptation of OEIT for intravascular detection in more translational settings and offers great potential as an alternative imaging tool for medical diagnosis.

Published
2022

36. Automatic segmentation of vertebral features on ultrasound spine images using Stacked Hourglass Network

Author

Zeng, Hong-Ye, Ge, Song-Han, Gao, Yu-Chong, Zhou, De-Sen, Zhou, Kang, He, Xu-Ming, Lou, Edmond, and Zheng, Rui

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

Objective: The spinous process angle (SPA) is one of the essential parameters to denote three-dimensional (3-D) deformity of spine. We propose an automatic segmentation method based on Stacked Hourglass Network (SHN) to detect the spinous processes (SP) on ultrasound (US) spine images and to measure the SPAs of clinical scoliotic subjects. Methods: The network was trained to detect vertebral SP and laminae as five landmarks on 1200 ultrasound transverse images and validated on 100 images. All the processed transverse images with highlighted SP and laminae were reconstructed into a 3D image volume, and the SPAs were measured on the projected coronal images. The trained network was tested on 400 images by calculating the percentage of correct keypoints (PCK); and the SPA measurements were evaluated on 50 scoliotic subjects by comparing the results from US images and radiographs. Results: The trained network achieved a high average PCK (86.8%) on the test datasets, particularly the PCK of SP detection was 90.3%. The SPAs measured from US and radiographic methods showed good correlation (r>0.85), and the mean absolute differences (MAD) between two modalities were 3.3{\deg}, which was less than the clinical acceptance error (5{\deg}). Conclusion: The vertebral features can be accurately segmented on US spine images using SHN, and the measurement results of SPA from US data was comparable to the gold standard from radiography., Comment: 9 pages,5 figures

Published
2021

43. Low-dose CT screening among never-smokers with or without a family history of lung cancer in Taiwan: a prospective cohort study

Author

Chan, Chang-Chuan, Chan, Si-Wa, Chang, I-Shou, Chang, Jer-Hwa, Chao, Kun-San, Chen, Chi-Jen, Chen, Huei-Wen, Chiang, Chun-Ju, Chiou, Hung-Yi, Chou, Mei-Chun, Chung, Chi-Li, Chung, Ta-Jung, Guo, Yue Leon, Hsiao, Chin-Fu, Huang, Chien-Sheng, Ko, Sheung-Fat, Lee, Mei-Hsuan, Li, Yao-Jen, Liao, Yu-San, Lu, Yueh-Hsun, Ou, Hsin-You, Wu, Ping-An, Yang, Hwai-I, Yang, Shi-Yi, Yang, Szu-Chun, Chang, Gee-Chen, Chiu, Chao-Hua, Yu, Chong-Jen, Chang, Yeun-Chung, Chang, Ya-Hsuan, Hsu, Kuo-Hsuan, Wu, Yu-Chung, Chen, Chih-Yi, Hsu, Hsian-He, Wu, Ming-Ting, Yang, Cheng-Ta, Chong, Inn-Wen, Lin, Yu-Ching, Hsia, Te-Chun, Lin, Meng-Chih, Su, Wu-Chou, Lin, Chih-Bin, Lee, Kang-Yun, Wei, Yu-Feng, Lan, Gong-Yau, Chan, Wing P, Wang, Kao-Lun, Wu, Mei-Han, Tsai, Hao-Hung, Chian, Chih-Feng, Lai, Ruay-Sheng, Shih, Jin-Yuan, Wang, Chi-Liang, Hsu, Jui-Sheng, Chen, Kun-Chieh, Chen, Chun-Ku, Hsia, Jiun-Yi, Peng, Chung-Kan, Tang, En-Kuei, Hsu, Chia-Lin, Chou, Teh-Ying, Shen, Wei-Chih, Tsai, Ying-Huang, Tsai, Chun-Ming, Chen, Yuh-Min, Lee, Yu-Chin, Chen, Hsuan-Yu, Yu, Sung-Liang, Chen, Chien-Jen, Wan, Yung-Liang, Hsiung, Chao Agnes, and Yang, Pan-Chyr

Published
2024
Full Text
View/download PDF

45. Machine learning‐directed electrical impedance tomography to predict metabolically vulnerable plaques

Author

Justin Chen, Shaolei Wang, Kaidong Wang, Parinaz Abiri, Zi‐Yu Huang, Junyi Yin, Alejandro M. Jabalera, Brian Arianpour, Mehrdad Roustaei, Enbo Zhu, Peng Zhao, Susana Cavallero, Sandra Duarte‐Vogel, Elena Stark, Yuan Luo, Peyman Benharash, Yu‐Chong Tai, Qingyu Cui, and Tzung K. Hsiai

Subjects
atherosclerosis, electrochemical impedance spectroscopy, machine learning, nanomaterials, oxidized low‐density lipoprotein, Chemical engineering, TP155-156, Biotechnology, TP248.13-248.65, Therapeutics. Pharmacology, RM1-950
Abstract

Abstract The characterization of atherosclerotic plaques to predict their vulnerability to rupture remains a diagnostic challenge. Despite existing imaging modalities, none have proven their abilities to identify metabolically active oxidized low‐density lipoprotein (oxLDL), a marker of plaque vulnerability. To this end, we developed a machine learning‐directed electrochemical impedance spectroscopy (EIS) platform to analyze oxLDL‐rich plaques, with immunohistology serving as the ground truth. We fabricated the EIS sensor by affixing a six‐point microelectrode configuration onto a silicone balloon catheter and electroplating the surface with platinum black (PtB) to improve the charge transfer efficiency at the electrochemical interface. To demonstrate clinical translation, we deployed the EIS sensor to the coronary arteries of an explanted human heart from a patient undergoing heart transplant and interrogated the atherosclerotic lesions to reconstruct the 3D EIS profiles of oxLDL‐rich atherosclerotic plaques in both right coronary and left descending coronary arteries. To establish effective generalization of our methods, we repeated the reconstruction and training process on the common carotid arteries of an unembalmed human cadaver specimen. Our findings indicated that our DenseNet model achieves the most reliable predictions for metabolically vulnerable plaque, yielding an accuracy of 92.59% after 100 epochs of training.

Published
2024
Full Text
View/download PDF

46. A deep learning fusion network trained with clinical and high-frequency ultrasound images in the multi-classification of skin diseases in comparison with dermatologists: a prospective and multicenter studyResearch in context

Author

An-Qi Zhu, Qiao Wang, Yi-Lei Shi, Wei-Wei Ren, Xu Cao, Tian-Tian Ren, Jing Wang, Ya-Qin Zhang, Yi-Kang Sun, Xue-Wen Chen, Yong-Xian Lai, Na Ni, Yu-Chong Chen, Jing-Liang Hu, Li-Chao Mou, Yu-Jing Zhao, Ye-Qiang Liu, Li-Ping Sun, Xiao-Xiang Zhu, Hui-Xiong Xu, and Le-Hang Guo

Subjects
Skin disease, Convolutional neural network, High-frequency ultrasound, Multi-classification, Medicine (General), R5-920
Abstract

Summary: Background: Clinical appearance and high-frequency ultrasound (HFUS) are indispensable for diagnosing skin diseases by providing internal and external information. However, their complex combination brings challenges for primary care physicians and dermatologists. Thus, we developed a deep multimodal fusion network (DMFN) model combining analysis of clinical close-up and HFUS images for binary and multiclass classification in skin diseases. Methods: Between Jan 10, 2017, and Dec 31, 2020, the DMFN model was trained and validated using 1269 close-ups and 11,852 HFUS images from 1351 skin lesions. The monomodal convolutional neural network (CNN) model was trained and validated with the same close-up images for comparison. Subsequently, we did a prospective and multicenter study in China. Both CNN models were tested prospectively on 422 cases from 4 hospitals and compared with the results from human raters (general practitioners, general dermatologists, and dermatologists specialized in HFUS). The performance of binary classification (benign vs. malignant) and multiclass classification (the specific diagnoses of 17 types of skin diseases) measured by the area under the receiver operating characteristic curve (AUC) were evaluated. This study is registered with www.chictr.org.cn (ChiCTR2300074765). Findings: The performance of the DMFN model (AUC, 0.876) was superior to that of the monomodal CNN model (AUC, 0.697) in the binary classification (P = 0.0063), which was also better than that of the general practitioner (AUC, 0.651, P = 0.0025) and general dermatologists (AUC, 0.838; P = 0.0038). By integrating close-up and HFUS images, the DMFN model attained an almost identical performance in comparison to dermatologists (AUC, 0.876 vs. AUC, 0.891; P = 0.0080). For the multiclass classification, the DMFN model (AUC, 0.707) exhibited superior prediction performance compared with general dermatologists (AUC, 0.514; P = 0.0043) and dermatologists specialized in HFUS (AUC, 0.640; P = 0.0083), respectively. Compared to dermatologists specialized in HFUS, the DMFN model showed better or comparable performance in diagnosing 9 of the 17 skin diseases. Interpretation: The DMFN model combining analysis of clinical close-up and HFUS images exhibited satisfactory performance in the binary and multiclass classification compared with the dermatologists. It may be a valuable tool for general dermatologists and primary care providers. Funding: This work was supported in part by the National Natural Science Foundation of China and the Clinical research project of Shanghai Skin Disease Hospital.

Published
2024
Full Text
View/download PDF

47. Compact CubeSat Gamma-Ray Detector for GRID Mission

Author

Wen, Jia-Xing, Zheng, Xu-Tao, Yu, Jian-Dong, Che, Yue-Peng, Yang, Dong-Xin, Gao, Huai-Zhong, Jin, Yi-Fei, Long, Xiang-Yun, Liu, Yi-Hui, Xu, Da-Cheng, Zhang, Yu-Chong, Zeng, Ming, Tian, Yang, Feng, Hua, Zeng, Zhi, Cang, Ji-Rong, Wu, Qiong, Zhao, Zong-Qing, Zhang, Bin-Bin, An, Peng, and collaboration, GRID

Subjects
Astrophysics - Instrumentation and Methods for Astrophysics
Abstract

Gamma-Ray Integrated Detectors (GRID) mission is a student project designed to use multiple gamma-ray detectors carried by nanosatellites (CubeSats), forming a full-time all-sky gamma-ray detection network that monitors the transient gamma-ray sky in the multi-messenger astronomy era. A compact CubeSat gamma-ray detector, including its hardware and firmware, was designed and implemented for the mission. The detector employs four Gd2Al2Ga3O12 : Ce (GAGG:Ce) scintillators coupled with four silicon photomultiplier (SiPM) arrays to achieve a high gamma-ray detection efficiency between 10 keV and 2 MeV with low power and small dimensions. The first detector designed by the undergraduate student team onboard a commercial CubeSat was launched into a Sun-synchronous orbit on October 29, 2018. The detector was in a normal observation state and accumulated data for approximately one month after on-orbit functional and performance tests, which were conducted in 2019., Comment: final manuscript, 9 pages, 10 figures, published on NST

Published
2021
Full Text
View/download PDF

48. Accelerating Sparse Deep Neural Networks

Author

Mishra, Asit, Latorre, Jorge Albericio, Pool, Jeff, Stosic, Darko, Stosic, Dusan, Venkatesh, Ganesh, Yu, Chong, and Micikevicius, Paulius

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Hardware Architecture
Abstract

As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in parameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsity support in its matrix-math units, Tensor Cores. We present the design and behavior of Sparse Tensor Cores, which exploit a 2:4 (50%) sparsity pattern that leads to twice the math throughput of dense matrix units. We also describe a simple workflow for training networks that both satisfy 2:4 sparsity pattern requirements and maintain accuracy, verifying it on a wide range of common tasks and model architectures. This workflow makes it easy to prepare accurate models for efficient deployment on Sparse Tensor Cores.

Published
2021

49. Specific gut microbiome signature predicts hepatitis B virus-related hepatocellular carcinoma patients with microvascular invasion

Author

Yu-Chong Peng, Jing-Xuan Xu, Xue-Mei You, Yi-Yue Huang, Liang Ma, Le-Qun Li, and Lu-Nan Qi

Subjects
Hepatitis B virus-related hepatocellular carcinoma, microvascular invasion, gut microbiomes, microbiome functions, prediction model, Medicine
Abstract

AbstractBackground We aimed to assess differences in intestinal microflora between patients with operable hepatitis B virus-related hepatocellular carcinoma (HBV-HCC) with microvascular invasion (MVI) and those without MVI. Additionally, we investigated the potential of the microbiome as a non-invasive biomarker for patients with MVI.Methods We analyzed the preoperative gut microbiomes (GMs) of two groups, the MVI (n = 46) and non-MVI (n = 56) groups, using 16S ribosomal RNA gene sequencing data. At the operational taxonomic unit level, we employed random forest models to predict MVI risk and validated the results in independent validation cohorts [MVI group (n = 17) and non-MVI group (n = 15)].Results β diversity analysis, utilizing weighted UniFrac distances, revealed a significant difference between the MVI and non-MVI groups, as indicated by non-metric multidimensional scaling and principal coordinate analysis. We also observed a significant correlation between the characteristic intestinal microbial communities at the genus level and their main functions. Nine optimal microbial markers were identified, with an area under the curve of 79.76% between 46 MVI and 56 non-MVI samples and 79.80% in the independent verification group.Conclusion This pioneering analysis of the GM in patients with operable HBV-HCC with and without MVI opens new avenues for treating HBV-HCC with MVI. We successfully established a diagnostic model and independently verified microbial markers for patients with MVI. As preoperative targeted biomarkers, GM holds potential as a non-invasive tool for patients with HBV-HCC with MVI.

Published
2023
Full Text
View/download PDF

50. Prognostic factors and an innovative nomogram model for patients with hepatocellular carcinoma treated with postoperative adjuvant transarterial chemoembolization

Author

Jing-Xuan Xu, Shui-Ling Qin, Hao-Wen Wei, Yuan-Yuan Chen, Yu-Chong Peng, and Lu-Nan Qi

Subjects
Hepatocellular carcinoma (HCC), post-operative adjuvant transcatheter arterial chemoembolization (PA-TACE), PA-TACE-insensitivity, nomogram, individualized prediction, Medicine
Abstract

AbstractPurpose The purpose of this study was to estimate the clinical efficacy and identify the best beneficiaries of postoperative adjuvant transcatheter arterial chemoembolization (PA-TACE) in hepatocellular carcinoma (HCC).Patients and methods A total of 749 HCC patients who underwent surgical resection (380 underwent PA-TACE, 369 had resection only) with a high risk of recurrence were reviewed retrospectively. Patients receiving PA-TACE were randomly split into development and validation cohorts. Univariate and multivariate analyses were performed in the development cohort. A novel model for PA-TACE-insensitivity prediction was built based on univariate and multivariate analysis and was multi-dimensionally validated in the validation set and all samples.Results After propensity score matching (PSM), in the early-recurrence group, no significant improvement in RFS was achieved with PA-TACE compared to radical hepatic resection alone. PA-TACE insensitive patients were considered as the PA-TACE non-benefit population and were associated with six clinicopathological factors: AFP, node number, tumor capsule, Ki-67 index, MVI, and complications in the development cohort. These factors were incorporated into a nomogram model, which reliably predicted PA-TACE insensitivity, with concordance indices of 0.874 and 0.897 for the development and validation cohort, respectively. In the overall sample, PA-TACE did not significantly improve patients’ RFS and OS in the high-score group, while the low-score group had statistical significance. Recurrence pattern diversity was also found to be a factor leading to PA-TACE insensitivity.Conclusion We constructed a new PA-TACE-insensitivity prediction model with potential clinical value. The good predictive performance and availability would allow this model to effectively screen PA-TACE beneficiaries.KEY MESSAGESThe independent influencing factors of PA-TACE insensitivity in patients who received PA-TACE were analyzed to construct a predictive model and its clinical application performance was verified with multi-dimensional methods.PA-TACE treatment should be avoided for patients with high scores according to this model, while it should be cautiously recommended for patients with low scores after multiple considerations.Compared with other related models, this model has obvious advantages in versatility and effectiveness. It can effectively screen the best benefit population of PA-TACE and provide a reliable reference for the selection of precise treatment plans for patients after radical resection of hepatocellular carcinoma.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results