Literature DB >> 31016442

Deep learning for liver tumor diagnosis part I: development of a convolutional neural network classifier for multi-phasic MRI.

Charlie A Hamm1,2, Clinton J Wang1, Lynn J Savic1,2, Marc Ferrante1, Isabel Schobert1,2, Todd Schlachter1, MingDe Lin1, James S Duncan1,3, Jeffrey C Weinreb1, Julius Chapiro4, Brian Letzen1.   

Abstract

OBJECTIVES: To develop and validate a proof-of-concept convolutional neural network (CNN)-based deep learning system (DLS) that classifies common hepatic lesions on multi-phasic MRI.
METHODS: A custom CNN was engineered by iteratively optimizing the network architecture and training cases, finally consisting of three convolutional layers with associated rectified linear units, two maximum pooling layers, and two fully connected layers. Four hundred ninety-four hepatic lesions with typical imaging features from six categories were utilized, divided into training (n = 434) and test (n = 60) sets. Established augmentation techniques were used to generate 43,400 training samples. An Adam optimizer was used for training. Monte Carlo cross-validation was performed. After model engineering was finalized, classification accuracy for the final CNN was compared with two board-certified radiologists on an identical unseen test set.
RESULTS: The DLS demonstrated a 92% accuracy, a 92% sensitivity (Sn), and a 98% specificity (Sp). Test set performance in a single run of random unseen cases showed an average 90% Sn and 98% Sp. The average Sn/Sp on these same cases for radiologists was 82.5%/96.5%. Results showed a 90% Sn for classifying hepatocellular carcinoma (HCC) compared to 60%/70% for radiologists. For HCC classification, the true positive and false positive rates were 93.5% and 1.6%, respectively, with a receiver operating characteristic area under the curve of 0.992. Computation time per lesion was 5.6 ms.
CONCLUSION: This preliminary deep learning study demonstrated feasibility for classifying lesions with typical imaging features from six common hepatic lesion types, motivating future studies with larger multi-institutional datasets and more complex imaging appearances. KEY POINTS: • Deep learning demonstrates high performance in the classification of liver lesions on volumetric multi-phasic MRI, showing potential as an eventual decision-support tool for radiologists. • Demonstrating a classification runtime of a few milliseconds per lesion, a deep learning system could be incorporated into the clinical workflow in a time-efficient manner.

Entities:  

Keywords:  Artificial intelligence; Deep learning; Liver cancer

Mesh:

Year:  2019        PMID: 31016442      PMCID: PMC7251621          DOI: 10.1007/s00330-019-06205-9

Source DB:  PubMed          Journal:  Eur Radiol        ISSN: 0938-7994            Impact factor:   5.315


  20 in total

1.  Radiologic-pathologic analysis of contrast-enhanced and diffusion-weighted MR imaging in patients with HCC after TACE: diagnostic accuracy of 3D quantitative image analysis.

Authors:  Julius Chapiro; Laura D Wood; MingDe Lin; Rafael Duran; Toby Cornish; David Lesage; Vivek Charu; Rüdiger Schernthaner; Zhijun Wang; Vania Tacher; Lynn Jeanette Savic; Ihab R Kamel; Jean-François Geschwind
Journal:  Radiology       Date:  2014-07-15       Impact factor: 11.105

2.  LI-RADS (Liver Imaging Reporting and Data System): summary, discussion, and consensus of the LI-RADS Management Working Group and future directions.

Authors:  Donald G Mitchell; Jordi Bruix; Morris Sherman; Claude B Sirlin
Journal:  Hepatology       Date:  2014-12-12       Impact factor: 17.425

3.  Interreader Reliability of LI-RADS Version 2014 Algorithm and Imaging Features for Diagnosis of Hepatocellular Carcinoma: A Large International Multireader Study.

Authors:  Kathryn J Fowler; An Tang; Cynthia Santillan; Mythreyi Bhargavan-Chatfield; Jay Heiken; Reena C Jha; Jeffrey Weinreb; Hero Hussain; Donald G Mitchell; Mustafa R Bashir; Eduardo A C Costa; Guilherme M Cunha; Laura Coombs; Tanya Wolfson; Anthony C Gamst; Giuseppe Brancatelli; Benjamin Yeh; Claude B Sirlin
Journal:  Radiology       Date:  2017-11-01       Impact factor: 11.105

4.  Automated diagnosis of focal liver lesions using bidirectional empirical mode decomposition features.

Authors:  U Rajendra Acharya; Joel En Wei Koh; Yuki Hagiwara; Jen Hong Tan; Arkadiusz Gertych; Anushya Vijayananthan; Nur Adura Yaakup; Basri Johan Jeet Abdullah; Mohd Kamil Bin Mohd Fabell; Chai Hong Yeong
Journal:  Comput Biol Med       Date:  2018-01-03       Impact factor: 4.589

5.  Concordance of hypervascular liver nodule characterization between the organ procurement and transplant network and liver imaging reporting and data system classifications.

Authors:  Mustafa R Bashir; Rong Huang; Nicholas Mayes; Daniele Marin; Carl L Berg; Rendon C Nelson; Tracy A Jaffe
Journal:  J Magn Reson Imaging       Date:  2014-11-05       Impact factor: 4.813

6.  Accuracy of the diagnostic evaluation of hepatocellular carcinoma with LI-RADS.

Authors:  Weimin Liu; Jie Qin; Ruomi Guo; Sidong Xie; Hang Jiang; Xiaohong Wang; Zhuang Kang; Jin Wang; Hong Shan
Journal:  Acta Radiol       Date:  2017-06-26       Impact factor: 1.990

Review 7.  Hepatocellular carcinoma: epidemiology and molecular carcinogenesis.

Authors:  Hashem B El-Serag; K Lenhard Rudolph
Journal:  Gastroenterology       Date:  2007-06       Impact factor: 22.682

8.  Rate of observation and inter-observer agreement for LI-RADS major features at CT and MRI in 184 pathology proven hepatocellular carcinomas.

Authors:  Eric C Ehman; Spencer C Behr; Sarah E Umetsu; Nicholas Fidelman; Ben M Yeh; Linda D Ferrell; Thomas A Hope
Journal:  Abdom Radiol (NY)       Date:  2016-05

9.  Classification of focal liver lesions on ultrasound images by extracting hybrid textural features and using an artificial neural network.

Authors:  Yoo Na Hwang; Ju Hwan Lee; Ga Young Kim; Yuan Yuan Jiang; Sung Min Kim
Journal:  Biomed Mater Eng       Date:  2015       Impact factor: 1.300

10.  Deep Learning with Convolutional Neural Network for Differentiation of Liver Masses at Dynamic Contrast-enhanced CT: A Preliminary Study.

Authors:  Koichiro Yasaka; Hiroyuki Akai; Osamu Abe; Shigeru Kiryu
Journal:  Radiology       Date:  2017-10-23       Impact factor: 11.105

View more
  57 in total

1.  Radiomics analysis of contrast-enhanced CT for classification of hepatic focal lesions in colorectal cancer patients: its limitations compared to radiologists.

Authors:  Heejin Bae; Hansang Lee; Sungwon Kim; Kyunghwa Han; Hyungjin Rhee; Dong-Kyu Kim; Hyuk Kwon; Helen Hong; Joon Seok Lim
Journal:  Eur Radiol       Date:  2021-05-10       Impact factor: 5.315

Review 2.  The overview of the deep learning integrated into the medical imaging of liver: a review.

Authors:  Kailai Xiang; Baihui Jiang; Dong Shang
Journal:  Hepatol Int       Date:  2021-07-15       Impact factor: 6.047

3.  Deep learning with ultrasonography: automated classification of liver fibrosis using a deep convolutional neural network.

Authors:  Jeong Hyun Lee; Ijin Joo; Tae Wook Kang; Yong Han Paik; Dong Hyun Sinn; Sang Yun Ha; Kyunga Kim; Choonghwan Choi; Gunwoo Lee; Jonghyon Yi; Won-Chul Bang
Journal:  Eur Radiol       Date:  2019-09-02       Impact factor: 5.315

4.  Assessment of knee pain from MR imaging using a convolutional Siamese network.

Authors:  Gary H Chang; David T Felson; Shangran Qiu; Ali Guermazi; Terence D Capellini; Vijaya B Kolachalama
Journal:  Eur Radiol       Date:  2020-02-13       Impact factor: 5.315

Review 5.  Up-to-Date Role of CT/MRI LI-RADS in Hepatocellular Carcinoma.

Authors:  Guilherme Moura Cunha; Victoria Chernyak; Kathryn J Fowler; Claude B Sirlin
Journal:  J Hepatocell Carcinoma       Date:  2021-05-31

6.  Radiomics Response Signature for Identification of Metastatic Colorectal Cancer Sensitive to Therapies Targeting EGFR Pathway.

Authors:  Laurent Dercle; Lin Lu; Lawrence H Schwartz; Min Qian; Sabine Tejpar; Peter Eggleton; Binsheng Zhao; Hubert Piessevaux
Journal:  J Natl Cancer Inst       Date:  2020-09-01       Impact factor: 13.506

7.  Automatized Hepatic Tumor Volume Analysis of Neuroendocrine Liver Metastases by Gd-EOB MRI-A Deep-Learning Model to Support Multidisciplinary Cancer Conference Decision-Making.

Authors:  Uli Fehrenbach; Siyi Xin; Alexander Hartenstein; Timo Alexander Auer; Franziska Dräger; Konrad Froböse; Henning Jann; Martina Mogl; Holger Amthauer; Dominik Geisel; Timm Denecke; Bertram Wiedenmann; Tobias Penzkofer
Journal:  Cancers (Basel)       Date:  2021-05-31       Impact factor: 6.639

8.  Protein functional annotation of simultaneously improved stability, accuracy and false discovery rate achieved by a sequence-based deep learning.

Authors:  Jiajun Hong; Yongchao Luo; Yang Zhang; Junbiao Ying; Weiwei Xue; Tian Xie; Lin Tao; Feng Zhu
Journal:  Brief Bioinform       Date:  2020-07-15       Impact factor: 11.622

9.  A Deep Convolutional Neural Network for Annotation of Magnetic Resonance Imaging Sequence Type.

Authors:  Sara Ranjbar; Kyle W Singleton; Pamela R Jackson; Cassandra R Rickertsen; Scott A Whitmire; Kamala R Clark-Swanson; J Ross Mitchell; Kristin R Swanson; Leland S Hu
Journal:  J Digit Imaging       Date:  2020-04       Impact factor: 4.056

Review 10.  [Artificial intelligence and radiomics in MRI-based prostate diagnostics].

Authors:  Charlie Alexander Hamm; Nick Lasse Beetz; Lynn Jeanette Savic; Tobias Penzkofer
Journal:  Radiologe       Date:  2020-01       Impact factor: 0.635

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.