A vision transformer for emphysema classification using CT images

2021 ◽  
Author(s):  
Yanan Wu ◽  
Shouliang Qi ◽  
Yu Sun ◽  
Shuyue Xia ◽  
Yudong Yao ◽  
...  
Keyword(s):  
Lung Parenchyma ◽  
Ct Images ◽  
Normal Lung ◽  
Massive Data ◽  
Large Patch ◽  
The Public ◽  
Evaluation Approach ◽  
Average Accuracy ◽  
Public Dataset ◽  
Aided Diagnosis

Abstract Objective: Emphysema is characterized by the destruction and permanent enlargement of the alveoli in the lung. According to visual CT appearance, emphysema can be divided into three subtypes: centrilobular emphysema (CLE), panlobular emphysema (PLE), and paraseptal emphysema (PSE). Automating emphysema classification can help precisely determine the patterns of lung destruction and provide a quantitative evaluation. Approach: We propose a vision transformer (ViT) model to classify the emphysema subtypes via CT images. First, large patches (61×61) are cropped from CT images which contain the area of normal lung parenchyma (NLP), CLE, PLE, and PSE. After resizing, the large patch is divided into small patches and these small patches are converted to a sequence of patch embeddings by flattening and linear embedding. A class embedding is concatenated to the patch embedding, and the positional embedding is added to the resulting embeddings described above. Then, the obtained embedding is fed into the transformer encoder blocks to generate the final representation. Finally, the learnable class embedding is fed to a softmax layer to classify the emphysema. Main results: To overcome the lack of massive data, the transformer encoder blocks (pre-trained on ImageNet) are transferred and fine-tuned in our ViT model. The average accuracy of the pre-trained ViT model achieves 95.95% in our lab’s own dataset which is higher than that of AlexNet, Inception-V3, MobileNet-V2, ResNet34, and ResNet50. Meanwhile, the pre-trained ViT model outperforms the ViT model without the pre-training. The accuracy of our pre-trained ViT model is higher than or comparable to that by available methods for the public dataset. Significance: The results demonstrated that the proposed ViT model can accurately classify the subtypes of emphysema using CT images. The ViT model can help make an effective computer-aided diagnosis of emphysema, and the ViT method can be extended to other medical applications.

scholarly journals Combining a convolutional neural network with autoencoders to predict the survival chance of COVID-19 patients

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Fahime Khozeimeh ◽  
Danial Sharifrazi ◽  
Navid Hoseini Izadi ◽  
Javad Hassannataj Joloudari ◽  
Afshin Shoeibi ◽  
...  
Keyword(s):  
Clinical Data ◽  
Data Augmentation ◽  
Clinical Information ◽  
Ct Images ◽  
Classification Performance ◽  
Survival Chance ◽  
Average Accuracy ◽  
Novel Method ◽  
Aided Diagnosis ◽  
Augmentation Procedure

AbstractCOVID-19 has caused many deaths worldwide. The automation of the diagnosis of this virus is highly desired. Convolutional neural networks (CNNs) have shown outstanding classification performance on image datasets. To date, it appears that COVID computer-aided diagnosis systems based on CNNs and clinical information have not yet been analysed or explored. We propose a novel method, named the CNN-AE, to predict the survival chance of COVID-19 patients using a CNN trained with clinical information. Notably, the required resources to prepare CT images are expensive and limited compared to those required to collect clinical data, such as blood pressure, liver disease, etc. We evaluated our method using a publicly available clinical dataset that we collected. The dataset properties were carefully analysed to extract important features and compute the correlations of features. A data augmentation procedure based on autoencoders (AEs) was proposed to balance the dataset. The experimental results revealed that the average accuracy of the CNN-AE (96.05%) was higher than that of the CNN (92.49%). To demonstrate the generality of our augmentation method, we trained some existing mortality risk prediction methods on our dataset (with and without data augmentation) and compared their performances. We also evaluated our method using another dataset for further generality verification. To show that clinical data can be used for COVID-19 survival chance prediction, the CNN-AE was compared with multiple pre-trained deep models that were tuned based on CT images.

scholarly journals ASF1B May Regulate the Tumor Microenvironment and Epithelial-mesenchymal Transition in Malignant Mesothelioma to Induce the Differentiation of Sarcomatoid Phenotype as a Prognosis Target

2021 ◽  
Author(s):  
Tianqi Chen ◽  
Yingjie Jia ◽  
Xiaojiang Li ◽  
Fanming Kong ◽  
Yafei Yin ◽  
...  
Keyword(s):  
Tumor Microenvironment ◽  
Malignant Mesothelioma ◽  
Immune Cell ◽  
Epithelial Mesenchymal Transition ◽  
Predictive Biomarker ◽  
Normal Lung ◽  
Mesenchymal Transition ◽  
The Public ◽  
Survival Gene ◽  
Public Dataset

Abstract IntroductionMalignant mesothelioma (MM) is a rare malignant tumor with a poor survival. However, few markers are verified that related to sarcomatoid differentiation or further stratification of prognosis.MethodsWe analysis the significance of ASF1B expression in MM by analyzing the public dataset from TCGA, GEO and Oncomine database. Furthermore, we investigated the biological function of ASF1B in immune microenvironment and the effect of ASF1B in survival, gene ontology enrichment, GSEA enrichment and related microRNA.ResultsASF1B expression was significantly higher in MM by GEO data compared to normal lung tissue (p<0.0001). This study provides evidence that the increased expression of ASF1B is significantly associated with poor prognosis and inhibitory immune cell infiltration in patients with MM and highlight that ASF1B could be used as a novel predictive biomarker for the prognosis of MM which is related to the differentiation of sarcomatoid phenotype.ConclusionASF1B May Regulate the Tumor Microenvironment and Epithelial-mesenchymal Transition in Malignant Mesothelioma to Induce the Differentiation of Sarcomatoid Phenotype as a Prognosis Target. Further researches need to be conducted to figure out how exactly the ASF1B affect the microenvironment.

Heterogeneous fairness algorithm based on federated learning in intelligent transportation system

2021 ◽  
pp. 1-9
Author(s):  
Yue Jiang ◽  
Gaochao Xu ◽  
Zhiyi Fang ◽  
Shinan Song ◽  
Bingbing Li
Keyword(s):  
Intelligent Transportation System ◽  
Influence Factor ◽  
Learning Algorithm ◽  
Intelligent Transportation ◽  
Transportation System ◽  
Massive Data ◽  
Infrared Sensors ◽  
Distributed Sensors ◽  
Average Accuracy ◽  
Global Fairness

With the development of the Intelligent Transportation System, various distributed sensors (including GPS, radar, infrared sensors) process massive data and make decisions for emergencies. Federated learning is a new distributed machine learning paradigm, in which system heterogeneity is the difficulty of fairness design. This paper designs a system heterogeneous fair federated learning algorithm (SHFF). SHFF introduces the equipment influence factor I into the optimization target and dynamically adjusts the equipment proportion with other performance. By changing the global fairness parameter θ, the algorithm can control fairness according to the actual needs. Experimental results show that, compared with the popular q-FedAvg algorithm, the SHFF algorithm proposed in this paper improves the average accuracy of the Worst 10% by 26% and reduces the variance by 61%.

Evaluation of a Nontraditional Approach to Fund, Finance, and Manage Metropolitan Freeways

2017 ◽  
Vol 2670 (1) ◽  
pp. 33-41 ◽  
Author(s):  
Patrick DeCorla-Souza ◽  
Morteza Farajian
Keyword(s):  
Cost Savings ◽  
Time Frame ◽  
Assessment Method ◽  
Public Welfare ◽  
Maintenance Cost ◽  
The Public ◽  
Welfare Benefits ◽  
Financial Position ◽  
Evaluation Approach ◽  
Operations And Maintenance

The purpose of this paper is twofold: ( a) to present a nontraditional, performance outcome–based public–private partnership (PPP) approach to finance and fund freeway reconstruction that relies not just on generating new revenue but also on optimizing scope and costs to achieve financial viability and ( b) to demonstrate how the approach can be evaluated for a specific project with an innovative value for money (VfM) assessment method that considers financial parameters, risk elements, and social benefits. The paper assesses the potential effects of the approach for a hypothetical project on ( a) the public agency’s financial position and ( b) public welfare. For this assessment, the effects of the project itself are assessed first by comparing conventional delivery of the project with “no build,” assuming that the project can be conventionally delivered in the same time frame as the PPP. Next, the effects of project acceleration are assessed by analyzing the effects of delaying conventional project delivery because of the public agency’s fiscal constraints. Finally, the PPP approach is compared with conventional delivery using public financing. The evaluation approach demonstrates how current VfM analysis practice may be enhanced by ( a) including a quantitative assessment of public welfare benefits and ( b) considering “no build” operations and maintenance cost savings to assess the net effect on the financial position of the agency.

An effective computer aided diagnosis model for pancreas cancer on PET/CT images

2018 ◽  
Vol 165 ◽  
pp. 205-214 ◽  
Author(s):  
Siqi Li ◽  
Huiyan Jiang ◽  
Zhiguo Wang ◽  
Guoxu Zhang ◽  
Yu-dong Yao
Keyword(s):  
Pancreas Cancer ◽  
Ct Images ◽  
Computer Aided Diagnosis ◽  
Pet Ct ◽  
Computer Aided ◽  
Diagnosis Model ◽  
Aided Diagnosis

scholarly journals Lamin A/C Mechanosensor Drives Tumor Cell Aggressiveness and Adhesion on Substrates With Tissue-Specific Elasticity

2021 ◽  
Vol 9 ◽  
Author(s):  
Enrica Urciuoli ◽  
Valentina D’Oria ◽  
Stefania Petrini ◽  
Barbara Peruzzi
Keyword(s):  
Tumor Cell ◽  
Tumor Cells ◽  
Lung Parenchyma ◽  
Normal Lung ◽  
Lamin A ◽  
Osteosarcoma Cell ◽  
Adhesion Properties ◽  
Tissue Specific ◽  
Definition Of ◽  
High Level

Besides its structural properties in the nucleoskeleton, Lamin A/C is a mechanosensor protein involved in perceiving the elasticity of the extracellular matrix. In this study we provide evidence about Lamin A/C-mediated regulation of osteosarcoma cell adhesion and spreading on substrates with tissue-specific elasticities. Our working hypothesis is based on the observation that low-aggressive and bone-resident SaOS-2 osteosarcoma cells express high level of Lamin A/C in comparison to highly metastatic, preferentially to the lung, osteosarcoma 143B cells, thereby suggesting a role for Lamin A/C in tumor cell tropism. Specifically, LMNA gene over-expression in 143B cells induced a reduction in tumor cell aggressiveness in comparison to parental cells, with decreased proliferation rate and reduced migration capability. Furthermore, LMNA reintegration into 143B cells changed the adhesion properties of tumor cells, from a preferential tropism toward the 1.5 kPa PDMS substrate (resembling normal lung parenchyma) to the 28 kPa (resembling pre-mineralized bone osteoid matrix). Our study suggests that Lamin A/C expression could be involved in the organ tropism of tumor cells, thereby providing a rationale for further studies focused on the definition of cancer mechanism of metastatization.

scholarly journals MAGAN: Mask Attention Generative Adversarial Network for Liver Tumor CT Image Synthesis

2020 ◽  
Author(s):  
Yang Liu ◽  
Lu Meng ◽  
Jianping Zhong
Keyword(s):  
Liver Tumor ◽  
Medical Image ◽  
Ct Images ◽  
Computer Aided Diagnosis ◽  
Generative Adversarial Network ◽  
Liver Ct ◽  
Adversarial Network ◽  
Computer Aided ◽  
Image Dataset ◽  
Aided Diagnosis

Abstract Background: For deep learning, the size of the dataset greatly affects the final training effect. However, in the field of computer-aided diagnosis, medical image datasets are often limited and even scarce.Methods: We aim to synthesize medical images and enlarge the size of the medical image dataset. In the present study, we synthesized the liver CT images with a tumor based on the mask attention generative adversarial network (MAGAN). We masked the pixels of the liver tumor in the image as the attention map. And both the original image and attention map were loaded into the generator network to obtain the synthesized images. Then the original images, the attention map, and the synthesized images were all loaded into the discriminator network to determine if the synthesized images were real or fake. Finally, we can use the generator network to synthesize liver CT images with a tumor.Results: The experiments showed that our method outperformed the other state-of-the-art methods, and can achieve a mean peak signal-to-noise ratio (PSNR) as 64.72dB.Conclusions: All these results indicated that our method can synthesize liver CT images with tumor, and build large medical image dataset, which may facilitate the progress of medical image analysis and computer-aided diagnosis.

scholarly journals Improving tuberculosis surveillance by detecting international transmission using publicly available whole-genome sequencing data

2019 ◽  
Author(s):  
Andrea Sanchini ◽  
Christine Jandrasits ◽  
Julius Tembrockhaus ◽  
Thomas Andreas Kohl ◽  
Christian Utpatel ◽  
...  
Keyword(s):  
Whole Genome Sequencing ◽  
Genome Sequencing ◽  
Large Scale ◽  
Added Value ◽  
Whole Genome Sequencing Data ◽  
Whole Genome ◽  
International Transmission ◽  
The Public ◽  
Public Dataset ◽  
Public Repositories

AbstractIntroductionImproving the surveillance of tuberculosis (TB) is especially important for multidrug-resistant (MDR) and extensively drug-resistant (XDR)-TB. The large amount of publicly available whole-genome sequencing (WGS) data for TB gives us the chance to re-use data and to perform additional analysis at a large scale.AimWe assessed the usefulness of raw WGS data of global MDR/XDR-TB isolates available from public repositories to improve TB surveillance.MethodsWe extracted raw WGS data and the related metadata of Mycobacterium tuberculosis isolates available from the Sequence Read Archive. We compared this public dataset with WGS data and metadata of 131 MDR- and XDR-TB isolates from Germany in 2012-2013.ResultsWe aggregated a dataset that includes 1,081 MDR and 250 XDR isolates among which we identified 133 molecular clusters. In 16 clusters, the isolates were from at least two different countries. For example, cluster2 included 56 MDR/XDR isolates from Moldova, Georgia, and Germany. By comparing the WGS data from Germany and the public dataset, we found that 11 clusters contained at least one isolate from Germany and at least one isolate from another country. We could, therefore, connect TB cases despite missing epidemiological information.ConclusionWe demonstrated the added value of using WGS raw data from public repositories to contribute to TB surveillance. By comparing the German and the public dataset, we identified potential international transmission events. Thus, using this approach might support the interpretation of national surveillance results in an international context.

Sign in / Sign up

Export Citation Format

Share Document