An Tang

Université de Montréal

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Claude B. Sirlin

University of California, San Diego

Samuel Kadoury

Centre Hospitalier Universitaire Sainte-Justine

Guy Cloutier

Université de Montréal

Bich Nguyen

Université de Montréal

Gilles Soulez

Centre Hospitalier de l’Université de Montréal

Mustafa R. Bashir

Duke University Hospital

Victoria Chernyak

Memorial Sloan Kettering Cancer Center

Milena Cerny

Université de Montréal

Damien Olivié

Centre Hospitalier de l’Université de Montréal

Zhang Hongsong

Henan University of Engineering

Cooperative Institutions

Université de Montréal

242

Centre Hospitalier de l’Université de Montréal

224

McGill University

100

University of Toronto

Stanford University

The University of Texas Southwestern Medical Center

University of California, San Diego

University of California, San Francisco

University of Ottawa

New York University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

LI-RADS ancillary features on contrast-enhanced ultrasonography

ULTRASONOGRAPHY (2020)

Christoph F. Dietrich Yi Dong Yuko Kono Cosmin Caraiani Claude B. Sirlin

The Liver Imaging Reporting and Data System (LI-RADS) was created to standardize liver imaging in patients at high risk for hepatocellular carcinoma (HCC), and it uses a diagnostic algorithm to assign categories that reflect the relative probability of HCC, non-HCC malignancies, or benign focal liver lesions. In addition to major imaging features, ancillary features (AFs) are used by radiologists to refine the categorization of liver nodules. In the present document, we discuss and explain the application of AFs currently defined within the LI-RADS guidelines. We also explore possible additional AFs visible on contrast-enhanced ultrasonography (CEUS). Finally, we summarize the management of CEUS LI-RADS features, including the role of current and potential future AFs.

BI-RADS

Contrast Enhancement

Contrast-enhanced ultrasound

10.14366/usg.19052

Cite

Citations (18)

Homodyned K-Distribution Parameter Estimation in Quantitative Ultrasound: Autoencoder and Bayesian Neural Network Approaches

IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control (2024)

Ali K. Z. Tehrani Guy Cloutier An Tang Iván M. Rosado-Méndez Hassan Rivaz

Quantitative ultrasound (QUS) analyzes the ultrasound (US) backscattered data to find the properties of scatterers that correlate with the tissue microstructure. Statistics of the envelope of the backscattered radio frequency (RF) data can be utilized to estimate several QUS parameters. Different distributions have been proposed to model envelope data. The homodyned K-distribution (HK-distribution) is one of the most comprehensive distributions that can model US backscattered envelope data under diverse scattering conditions (varying scatterer number density and coherent scattering). The scatterer clustering parameter ( α ) and the ratio of the coherent to diffuse scattering power ( k ) are the parameters of this distribution that have been used extensively for tissue characterization in diagnostic US. The estimation of these two parameters (which we refer to as HK parameters) is done using optimization algorithms in which statistical features such as the envelope point-wise signal-to-noise ratio (SNR), skewness, kurtosis, and the log-based moments have been utilized as input to such algorithms. The optimization methods minimize the difference between features and their theoretical value from the HK model. We propose that the true value of these statistical features is a hyperplane that covers a small portion of the feature space. In this article, we follow two approaches to reduce the effect of sample features' error. We propose a model projection neural network based on denoising autoencoders to project the noisy features into this space based on this assumption. We also investigate if the noise distribution can be learned by the deep estimators. We compare the proposed methods with conventional methods using simulations, an experimental phantom, and data from an in vivo animal model of hepatic steatosis. The network weight and a demo code are available online at ht.tp://code.sonography.ai.

Kurtosis

Envelope (radar)

Generalized normal distribution

10.1109/tuffc.2024.3357438

Cite

Citations (4)

Potential Presymptomatic Transmission of SARS-CoV-2, Zhejiang Province, China, 2020

Emerging infectious diseases (2020)

Zhendong Tong An Tang Kefeng Li Peng Li Hongling Wang

Abstract We report a 2-family cluster of persons infected with severe acute respiratory syndrome coronavirus 2 in the city of Zhoushan, Zhejiang Province, China, during January 2020. The infections resulted from contact with an infected but potentially presymptomatic traveler from the city of Wuhan in Hubei Province.

Sars virus

2019-20 coronavirus outbreak

Betacoronavirus

10.3201/eid2605.200198

Cite

Citations (500)

Canadian Association of Radiologists White Paper on Ethical and Legal Issues Related to Artificial Intelligence in Radiology

Canadian Association of Radiologists Journal (2019)

Jacob L. Jaremko Marleine Azar Rebecca Jaremko Bromwich Andrea Lum Li Hsia Alicia Cheong

Artificial intelligence (AI) software that analyzes medical images is becoming increasingly prevalent. Unlike earlier generations of AI software, which relied on expert knowledge to identify imaging features, machine learning approaches automatically learn to recognize these features. However, the promise of accurate personalized medicine can only be fulfilled with access to large quantities of medical data from patients. This data could be used for purposes such as predicting disease, diagnosis, treatment optimization, and prognostication. Radiology is positioned to lead development and implementation of AI algorithms and to manage the associated ethical and legal challenges. This white paper from the Canadian Association of Radiologists provides a framework for study of the legal and ethical issues related to AI in medical imaging, related to patient data (privacy, confidentiality, ownership, and sharing); algorithms (levels of autonomy, liability, and jurisprudence); practice (best practices and current legal framework); and finally, opportunities in AI from the perspective of a universal health care system.

White paper

Best practice

10.1016/j.carj.2019.03.001

Cite

Citations (146)

Channel-Selective Normalization for Label-Shift Robust Test-Time Adaptation

arXiv (Cornell University) (2024)

Pedro Vianna Muawiz Chaudhary Paria Mehrbod An Tang Guy Cloutier

Deep neural networks have useful applications in many different tasks, however their performance can be severely affected by changes in the data distribution. For example, in the biomedical field, their performance can be affected by changes in the data (different machines, populations) between training and test datasets. To ensure robustness and generalization to real-world scenarios, test-time adaptation has been recently studied as an approach to adjust models to a new data distribution during inference. Test-time batch normalization is a simple and popular method that achieved compelling performance on domain shift benchmarks. It is implemented by recalculating batch normalization statistics on test batches. Prior work has focused on analysis with test data that has the same label distribution as the training data. However, in many practical applications this technique is vulnerable to label distribution shifts, sometimes producing catastrophic failure. This presents a risk in applying test time adaptation methods in deployment. We propose to tackle this challenge by only selectively adapting channels in a deep network, minimizing drastic adaptation that is sensitive to label shifts. Our selection scheme is based on two principles that we empirically motivate: (1) later layers of networks are more sensitive to label shift (2) individual features can be sensitive to specific classes. We apply the proposed technique to three classification tasks, including CIFAR10-C, Imagenet-C, and diagnosis of fatty liver, where we explore both covariate and label distribution shifts. We find that our method allows to bring the benefits of TTA while significantly reducing the risk of failure common in other methods, while being robust to choice in hyperparameters.

Normalization

10.48550/arxiv.2402.04958

Cite

Citations (0)

Microbial and isotopomer analysis of N2O production pathways in a calcareous film-mulched farmland

Biology and Fertility of Soils (2023)

Na Gao Fangfang Zhang Qifei Bo An Tang JiaRui Gao

Soil carbon

10.1007/s00374-023-01711-x

Cite

Citations (16)

発熱・血小板減少症候群の流行病学研究の進展を合併した【JST・京大機械翻訳】

Zhonghua Liuxingbingxue Zazhi (2016)

Jian‐Yue Wang Wu Hui Tong Zhendong Yan Jianbo Li Kefeng

Source

Cite

Citations (0)

Understanding LI-RADS

Magnetic Resonance Imaging Clinics of North America (2014)

Cynthia Santillan An Tang Irene Cruite Amol Shah Claude B. Sirlin

BI-RADS

10.1016/j.mric.2014.04.007

Cite

Citations (46)

Comparative transcriptomic analyzes of human lung epithelial cells infected with wild-type SARS-CoV-2 and its variant with a 12-bp missing in the E gene

Frontiers in Microbiology (2023)

Yisheng Sun Hao Sun Hanping Zhu Gao-Lei Li Xu Fang

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel coronavirus that caused a global outbreak of coronavirus disease 2019 (COVID-19) pandemic. To elucidate the mechanism of SARS-CoV-2 replication and immunogenicity, we performed a comparative transcriptome profile of mRNA and long non-coding RNAs (lncRNAs) in human lung epithelial cells infected with the SARS-CoV-2 wild-type strain (8X) and the variant with a 12-bp deletion in the E gene (F8). In total, 3,966 differentially expressed genes (DEGs) and 110 differentially expressed lncRNA (DE-lncRNA) candidates were identified. Of these, 94 DEGs and 32 DE-lncRNAs were found between samples infected with F8 and 8X. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyzes revealed that pathways such as the TNF signaling pathway and viral protein interaction with cytokine and cytokine receptor were involved. Furthermore, we constructed a lncRNA-protein-coding gene co-expression interaction network. The KEGG analysis of the co-expressed genes showed that these differentially expressed lncRNAs were enriched in pathways related to the immune response, which might explain the different replication and immunogenicity properties of the 8X and F8 strains. These results provide a useful resource for studying the pathogenesis of SARS-CoV-2 variants.

KEGG

Coronavirus

10.3389/fmicb.2022.1079764

Cite

Citations (2)

Liver segmentation: a primer for interventional radiologists

Journal of Vascular and Interventional Radiology (2016)

Akshat Gotra P. Delli Fraine Ali Bessissow Isabelle Dupuis Hassan Baydoun

Portal vein embolization

10.1016/j.jvir.2015.12.661

Cite

Citations (1)