Mask-Guided Data Augmentation for Multiparametric MRI Generation with a Rare Hepatocellular Carcinoma
文献类型： 会议论文
第一作者： Karen Sanchez
作者： Karen Sanchez 1 ; Carlos Hinojosa 2 ; Kevin Arias 2 ; Henry Arguello 2 ; Denis Kouamé 3 ; Olivier Meyrignac 4 ; Adrian Basarab 5 ;
作者机构： 1.HDSP Research Group, Universidad Industrial de Santander, Bucaramanga, Colombia|Univ Lyon, Université Claude Bernard Lyon 1, France
2.HDSP Research Group, Universidad Industrial de Santander, Bucaramanga, Colombia
3.IRIT, University of Toulouse CNRS 5505, Université Paul Sabatier Toulouse 3, France
4.Assistance Publique des h?pitaux de Paris, Bicêtre University Hospital, Biomaps, France
5.Univ Lyon, Université Claude Bernard Lyon 1, France
关键词： Deep learning;Image segmentation;Visualization;Magnetic resonance imaging;Image edge detection;Liver;Data augmentation
会议名称： IEEE International Symposium on Biomedical Imaging
主办单位：
页码： 1-5
摘要： Data augmentation is classically used to improve the overall performance of deep learning models. It is, however, challenging in the case of medical applications, and in particular for multiparametric datasets. For example, traditional geometric transformations used in several applications to generate synthetic images can modify in a non-realistic manner the patients’ anatomy. Therefore, dedicated image generation techniques are necessary in the medical field to, for example, mimic a given pathology in a realistic manner. This paper introduces a new data augmentation architecture that generates synthetic multiparametric (T1 arterial, T1 portal, and T2) magnetic resonance images (MRI) of massive macrotrabecular subtype hepatocellular carcinoma with their corresponding tumor masks through a generative deep learning approach. The proposed architecture creates liver tumor masks and abdominal edges used as input in a Pix2Pix network for synthetic data creation. The efficiency of the method is demonstrated by training it on a limited multiparametric dataset of MRI triplets from 89 patients with liver lesions to generate 1, 000 synthetic triplets and their corresponding liver tumor masks. The resulting Frechet Inception Distance score was 86.55. The proposed approach was among the winners of the 2021 data augmentation challenge organized by the French Society of Radiology.
分类号： tp391.4
- 相关文献
[1]A multi-level convolutional LSTM model for the segmentation of left ventricle myocardium in infarcted porcine cine MR images. Dongqing Zhang,Ilknur Icke,Belma Dogdas,Sarayu Parimal,Smita Sampath,Joseph Forbes,Ansuman Bagchi,Chih-Liang Chin,Antong Chen. 2018
[2]Automatic Tool for Pulmonary Artery Hemodynamic Assessment from 4D flow MRI. Maialen Stephens,ángel Gaitán,Rahul Kumar,Jorge Nuche,Irene Marco,Juan Delgado,Jesús Ruíz-Cabello,Karen López-Linares,Javier Bóbeda,Haizea Erostarbe. 2023
[3]Learning Functional Embedding of Genes Governed by Pair-wised Labels. Cao, Jingjun,Wu, Zhengli,Ye, Wenting,Wang, Haohan. 2017
[4]CuInS2/ZnS Quantum Dots Conjugating Gd(III) Chelates for Near-Infrared Fluorescence and Magnetic Resonance Bimodal Imaging. Yang, Yongbo,Lin, Li,Jing, Lijia,Yue, Xiuli,Dai, Zhifei,Yang, Yongbo. 2017
[5]Pulsed vacuum drying of Thompson seedless grape: Effects of berry ripeness on physicochemical properties and drying characteristic. Wang, Jun,Xue, Ling-Yang,Xie, Long,Xiao, Hong-Wei,Gao, Zhen-Jiang,Mu, Wei-Song,Fang, Xiao-Ming,Mujumdar, A. S.,Yang, Xu-Hai.
[6]Image segmentation of G bands of Triticum monococcum chromosomes based on the model-based neural network. Cai, N,Hu, KH,Xiong, HT,Li, SY,Su, WF,Zhu, FS.
[7]Robust Image Segmentation Method for Cotton Leaf Under Natural Conditions Based on Immune Algorithm and PCNN Algorithm. Zhang, Jianhua,Kong, Fantao,Wu, Jianzhai,Han, Shuqing,Zhai, Zhifen. 2018
[8]Image Segmentation to HSI Model Based on Improved Particle Swarm Optimization. Zhao, Bo,Mao, Wenhua,Zhang, Xiaochao,Chen, Yajun. 2009
[9]A segmentation method for greenhouse vegetable foliar disease spots images using color information and region growing. Ma, Juncheng,Du, Keming,Zheng, Feixiang,Chu, Jinxiang,Sun, Zhongfu,Zhang, Lingxian.
[10]Abdominal multi-organ localization on contrast-enhanced CT based on maximum a posteriori probability and minimum volume overlap. Liu, Xiaofeng,Linguraru, Marius George,Yao, Jianhua,Summers, Ronald M.. 2011
[11]CT-Realistic Data Augmentation Using Generative Adversarial Network for Robust Lymph Node Segmentation. You-Bao Tang,Sooyoun Oh,Yu-Xing Tang,Jing Xiao,Ronald M. Summers. 2019
[12]Image Similarity Ranking of Focal Computed Tomography Liver Lesions Using a 2AFC Technique. Jessica Faruque,Sameer Antani,Rodney Long,Lauren Kim,George Thoma. 2016
[13]CT-Realistic Data Augmentation Using Generative Adversarial Network for Robust Lymph Node Segmentation. You-Bao Tang,Sooyoun Oh,Yu-Xing Tang,Jing Xiao,Ronald M. Summers. 2019
[14]Visualization of Protein in Peanut Using Hyperspectral Image with Chemometrics. Liu Hong-zhi. 2017
[15]GFF3sort: a novel tool to sort GFF3 files for tabix indexing. Zhu, Tao,Liang, Chengzhen,Meng, Zhigang,Guo, Sandui,Zhang, Rui. 2017
[16]Development of loop-mediated isothermal amplification for rapid detection of avian leukosis virus subgroup A. Wang, Yongqiang,Kang, Zhonghui,Gao, Yulong,Qin, Liting,Chen, Lei,Wang, Qi,Li, Jiukuan,Gao, Honglei,Qi, Xiaole,Lin, Huan,Wang, Xiaomei. 2011
[17]Study on the rice growth visualization system. Yan, Dingchun,Zhu, Yeping. 2007
[18]RESEARCH ON CROP SIMULATION MODEL BASED ON PDA. Zhang, Jianbing,Zhu, Yeping. 2009
[19]Integrated analysis of mRNA and miRNA expression profiles in livers of Yimeng black pigs with extreme phenotypes for backfat thickness. Li, Wentong,Yang, Yalan,Zhou, Rong,Li, Kui,Li, Wentong,Liu, Shuai,Li, Xiuxiu,Wang, Yingping,Tang, Hui,Li, Wentong,Yang, Yalan,Liu, Ying,Zhang, Yanmin,Zhou, Rong,Li, Kui,Li, Kui. 2017
[20]GROWTH PERFORMANCE AND ANTIOXIDANT CAPACITIES IN PIGLETS WITH DIFFERENT WEANING AGE. Li, L. A.,Yang, J. J.,Li, Y.,Jin, T. M.,Li, R. N.,Qin, S. Y.,Xie, J. J.,Du, G. M.. 2016
