您好，欢迎访问中国农业科学院 机构知识库！

全部

  • 全部
  • 标题
  • 学者
  • 机构
  • 关键词

Class Imbalance Correction for Improved Universal Lesion Detection and Tagging in CT

文献类型： 会议论文

第一作者： Peter D. Erickson

作者： Peter D. Erickson 1 ; Tejas Sudharshan Mathai 1 ; Ronald M. Summers 1 ;

作者机构： 1.Imaging Biomarkers and Computer-Aided Diagnosis Laboratory, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health

关键词： CT;Universal lesion detection;Class imbalance;Deep learning;DeepLesion

会议名称： International Workshop on Medical Image Learning with Limited and Noisy Data

主办单位：

页码： 177-186

摘要： Radiologists routinely detect and size lesions in CT to stage cancer and assess tumor burden. To potentially aid their efforts, multiple lesion detection algorithms have been developed with a large public dataset called DeepLesion (32,735 lesions, 32,120 CT slices, 10,594 studies, 4,427 patients, 8 body part labels). However, this dataset contains missing measurements and lesion tags, and exhibits a severe imbalance in the number of lesions per label category. In this work, we utilize a limited subset of DeepLesion (6%, 1331 lesions, 1309 slices) containing lesion annotations and body part label tags to train a VFNet model to detect lesions and tag them. We address the class imbalance by conducting three experiments: 1) Balancing data by the body part labels, 2) Balancing data by the number of lesions per patient, and 3) Balancing data by the lesion size. In contrast to a randomly sampled (unbalanced) data subset, our results indicated that balancing the body part labels always increased sensitivity for lesions ≥ 1 cm for classes with low data quantities (Bone: 80% vs. 46%, Kidney: 77% vs. 61%, Soft Tissue: 70% vs. 60%, Pelvis: 83% vs. 76%). Similar trends were seen for three other models tested (FasterRCNN, RetinaNet, FoveaBox). Balancing data by lesion size also helped the VFNet model improve recalls for all classes in contrast to an unbalanced dataset. We also provide a structured reporting guideline for a "Lesions" subsection to be entered into the "Findings" section of a radiology report. To our knowledge, we are the first to report the class imbalance in DeepLesion, and have taken data-driven steps to address it in the context of joint lesion detection and tagging.

分类号： tp391.41-53

  • 相关文献

[1]Universal Lesion Detection and Classification Using Limited Data and Weakly-Supervised Self-training. Varun Naga,Tejas Sudharshan Mathai,Angshuman Paul,Ronald M. Summers. 2022

[2]实时荧光定量PCR在植物害虫研究中的应用. 王忠跃,孔繁芳,武艳霞. 2011

[3]CT扫描分析秸秆隔层孔隙特征及其对土壤水入渗的影响. 张宏媛,逄焕成,卢闯,刘娜,张晓丽,李玉义. 2019

[4]适合马铃薯块茎计算机断层扫描原位观察的栽培基质筛选. 谭杰,刘建刚,刘菊,简银巧,李广存,金黎平,徐建飞. 2023

[5]Learning Functional Embedding of Genes Governed by Pair-wised Labels. Cao, Jingjun,Wu, Zhengli,Ye, Wenting,Wang, Haohan. 2017

[6]Segmenting the Thoracic, Abdominal and Pelvic Musculature on CT Scans Combining Atlas-based Model and Active Contour Model. Weidong Zhang,Jiamin Liu,Jianhua Yao,Ronald M. Summers. 2013

[7]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

[8]COMPUTER-AIDED DETECTION OF COLITIS ON COMPUTED TOMOGRAPHY USING A VISUAL CODEBOOK. Weidong Zhang,Jianfei Liu,Shijun Wang,Jianhua Yao,Ronald M. Summers,Zhuoshi Wei. 2013

[9]High Throughput Lung and Lobar Segmentation by 2D and 3D CNN on Chest CT with Diffuse Lung Disease. Xiaoyong Wang,Pangyu Teng,Pechin Lo,Ashley Banola,Grace Kim,Fereidoun Abtin,Jonathan Goldin,Matthew Brown. 2018

[10]Electronic Stool Subtraction Using Quadratic Regression, Morphological Operations, and Distance Transforms. Michael Carston,Armando Manduca,C. Daniel Johnson,SPIE-The International Society for Optical Engineering. 2006

[11]Semantic Image Synthesis for Abdominal CT. Yan Zhuang,Benjamin Hou,Tejas Sudharshan Mathai,Pritam Mukherjee,Boah Kim,Ronald M. Summers. 2024

[12]Design Optimization of Neuroendoscopic Continuum Instruments for Third Ventriculostomy and Tumor Biopsy. Kyle W. Eastwood,Thomas Looi,Hani E. Naguib,James M. Drake. 2015

[13]CT colonography of the unprepared colon: An evaluation of electronic stool subtraction. Michael J. Carston,Robert J. Wentz,Armando Manduca,C. Daniel Johnson. 2005

[14]Design Optimization of Neuroendoscopic Continuum Instruments for Third Ventriculostomy and Tumor Biopsy. Eastwood, Kyle W.,Looi, Thomas,Naguib, Hani E.,Drake, James M.. 2015

[15]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

[16]Sleep Apnea Severity Estimation from Respiratory Related Movements Using Deep Learning. Maziar Hafezi,Nasim Montazeri,Kaiyin Zhu,Hisham Alshaer,Azadeh Yadollahi,Babak Taati. 2019

[17]An Intelligent Network Planning Algorithm for Emergency Communication with Deep Learning. Zongzhe Nie,Wei Zhu,Ruopeng Yang,Yu Wang. 2019

[18]Research on the Key Techniques of Semantic Mining of Information Digest in the Field of Agricultural Major Crops Based on Deep Learning. Ping Qian,Hao G. J. M. Gong,Yunpeng Cui. 2019

[19]DEEP LEARNING WITH ORTHOGONAL VOLUMETRIC HED SEGMENTATION AND 3D SURFACE RECONSTRUCTION MODEL OF PROSTATE MRI. Ruida Cheng,Nathan Lay,Francesca Mertan,Baris Turkbey,Holger R. Roth,Le Lu,William Gandler,Evan S. McCreedy,Thomas Pohida,Peter Choyke,Matthew J. McAuliffe,Ronald M. Summers. 2017

[20]Mapping Rice Area Using Sentinel-1 SAR Data and Deep Learning. Guozhuang Shen,Chenwei Nie. 2023

作者其他论文 更多>>
置顶
购物车
反馈

© 京ICP备09089781号-29 主办单位：中国农业科学院

学术资源： 中国农业科学院 农业专业知识服务系统 农科联盟资源共建共享平台 中国农业科技文献与信息服务平台 中国农业期刊集成服务平台