Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions: Generation of a predictive model on the basis of quantitative spatial frequency measurements - A work in progress

Siva P. Raman, James L. Schroeder, Peng Huang, Yifei Chen, Stephanie Coquia, Satomi Kawamoto, Elliot K Fishman

Research output: Contribution to journalArticle

Abstract

Objective Computed tomography texture analysis (CTTA) is a method of quantifying lesion heterogeneity based on distribution of pixel intensities within a region of interest. This study investigates the ability of CTTA to distinguish different hypervascular liver lesions and compares CTTA parameters by creating a proof-of-concept model to distinguish between different lesions. Methods Following institutional review board approval, CTTA software (TexRAD Ltd) was used to retrospectively analyze 17 cases of focal nodular hyperplasia, 19 hepatic adenomas, 25 hepatocellular carcinomas, and 19 cases of normal liver parenchyma using arterial phase scans. Two radiologists read the same image series used by the CTTA software and reported their best guess diagnosis. Computed tomography texture analysis parameters were computed from regions of interest using spatial band-pass filters to quantify heterogeneity. Random-forest method was used to construct a predictive model from these parameters, and a separate regression model was created using a subset of parameters. Results The random-forest model successfully distinguished the 3 lesion types and normal liver with predicted classification performance accuracy for 91.2% for adenoma, 94.4% for focal nodular hyperplasia, and 98.6% for hepatocellular carcinoma. This error prediction was generated using a subset of data points not used in generation of the model, but not on discrete prospective cases. In contrast, the 2 human readers using the same image series data analyzed by the CTTA software had lower accuracies, of 72.2% and 65.6%, respectively. The explicit regression model with a subset of image parameters had intermediate overall accuracy of 84.9%. Conclusions Computed tomography texture analysis may prove valuable in lesion characterization. Differentiation between common hypervascular lesion types could be aided by the judicious incorporation of texture parameters into clinical analysis.

Original languageEnglish (US)
Pages (from-to)383-395
Number of pages13
JournalJournal of Computer Assisted Tomography
Volume39
Issue number3
DOIs
StatePublished - May 29 2015

Fingerprint

Tomography
Liver
Focal Nodular Hyperplasia
Software
Adenoma
Hepatocellular Carcinoma
Research Ethics Committees

Keywords

  • computed tomography
  • hypervascular liver lesions
  • quantitative imaging
  • texture analysis

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{d926fd64e67b4d36b7b36dcd9338c8cc,
title = "Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions: Generation of a predictive model on the basis of quantitative spatial frequency measurements - A work in progress",
abstract = "Objective Computed tomography texture analysis (CTTA) is a method of quantifying lesion heterogeneity based on distribution of pixel intensities within a region of interest. This study investigates the ability of CTTA to distinguish different hypervascular liver lesions and compares CTTA parameters by creating a proof-of-concept model to distinguish between different lesions. Methods Following institutional review board approval, CTTA software (TexRAD Ltd) was used to retrospectively analyze 17 cases of focal nodular hyperplasia, 19 hepatic adenomas, 25 hepatocellular carcinomas, and 19 cases of normal liver parenchyma using arterial phase scans. Two radiologists read the same image series used by the CTTA software and reported their best guess diagnosis. Computed tomography texture analysis parameters were computed from regions of interest using spatial band-pass filters to quantify heterogeneity. Random-forest method was used to construct a predictive model from these parameters, and a separate regression model was created using a subset of parameters. Results The random-forest model successfully distinguished the 3 lesion types and normal liver with predicted classification performance accuracy for 91.2{\%} for adenoma, 94.4{\%} for focal nodular hyperplasia, and 98.6{\%} for hepatocellular carcinoma. This error prediction was generated using a subset of data points not used in generation of the model, but not on discrete prospective cases. In contrast, the 2 human readers using the same image series data analyzed by the CTTA software had lower accuracies, of 72.2{\%} and 65.6{\%}, respectively. The explicit regression model with a subset of image parameters had intermediate overall accuracy of 84.9{\%}. Conclusions Computed tomography texture analysis may prove valuable in lesion characterization. Differentiation between common hypervascular lesion types could be aided by the judicious incorporation of texture parameters into clinical analysis.",
keywords = "computed tomography, hypervascular liver lesions, quantitative imaging, texture analysis",
author = "Raman, {Siva P.} and Schroeder, {James L.} and Peng Huang and Yifei Chen and Stephanie Coquia and Satomi Kawamoto and Fishman, {Elliot K}",
year = "2015",
month = "5",
day = "29",
doi = "10.1097/RCT.0000000000000217",
language = "English (US)",
volume = "39",
pages = "383--395",
journal = "Journal of Computer Assisted Tomography",
issn = "0363-8715",
publisher = "Lippincott Williams and Wilkins",
number = "3",

}

TY - JOUR

T1 - Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions

T2 - Generation of a predictive model on the basis of quantitative spatial frequency measurements - A work in progress

AU - Raman, Siva P.

AU - Schroeder, James L.

AU - Huang, Peng

AU - Chen, Yifei

AU - Coquia, Stephanie

AU - Kawamoto, Satomi

AU - Fishman, Elliot K

PY - 2015/5/29

Y1 - 2015/5/29

N2 - Objective Computed tomography texture analysis (CTTA) is a method of quantifying lesion heterogeneity based on distribution of pixel intensities within a region of interest. This study investigates the ability of CTTA to distinguish different hypervascular liver lesions and compares CTTA parameters by creating a proof-of-concept model to distinguish between different lesions. Methods Following institutional review board approval, CTTA software (TexRAD Ltd) was used to retrospectively analyze 17 cases of focal nodular hyperplasia, 19 hepatic adenomas, 25 hepatocellular carcinomas, and 19 cases of normal liver parenchyma using arterial phase scans. Two radiologists read the same image series used by the CTTA software and reported their best guess diagnosis. Computed tomography texture analysis parameters were computed from regions of interest using spatial band-pass filters to quantify heterogeneity. Random-forest method was used to construct a predictive model from these parameters, and a separate regression model was created using a subset of parameters. Results The random-forest model successfully distinguished the 3 lesion types and normal liver with predicted classification performance accuracy for 91.2% for adenoma, 94.4% for focal nodular hyperplasia, and 98.6% for hepatocellular carcinoma. This error prediction was generated using a subset of data points not used in generation of the model, but not on discrete prospective cases. In contrast, the 2 human readers using the same image series data analyzed by the CTTA software had lower accuracies, of 72.2% and 65.6%, respectively. The explicit regression model with a subset of image parameters had intermediate overall accuracy of 84.9%. Conclusions Computed tomography texture analysis may prove valuable in lesion characterization. Differentiation between common hypervascular lesion types could be aided by the judicious incorporation of texture parameters into clinical analysis.

AB - Objective Computed tomography texture analysis (CTTA) is a method of quantifying lesion heterogeneity based on distribution of pixel intensities within a region of interest. This study investigates the ability of CTTA to distinguish different hypervascular liver lesions and compares CTTA parameters by creating a proof-of-concept model to distinguish between different lesions. Methods Following institutional review board approval, CTTA software (TexRAD Ltd) was used to retrospectively analyze 17 cases of focal nodular hyperplasia, 19 hepatic adenomas, 25 hepatocellular carcinomas, and 19 cases of normal liver parenchyma using arterial phase scans. Two radiologists read the same image series used by the CTTA software and reported their best guess diagnosis. Computed tomography texture analysis parameters were computed from regions of interest using spatial band-pass filters to quantify heterogeneity. Random-forest method was used to construct a predictive model from these parameters, and a separate regression model was created using a subset of parameters. Results The random-forest model successfully distinguished the 3 lesion types and normal liver with predicted classification performance accuracy for 91.2% for adenoma, 94.4% for focal nodular hyperplasia, and 98.6% for hepatocellular carcinoma. This error prediction was generated using a subset of data points not used in generation of the model, but not on discrete prospective cases. In contrast, the 2 human readers using the same image series data analyzed by the CTTA software had lower accuracies, of 72.2% and 65.6%, respectively. The explicit regression model with a subset of image parameters had intermediate overall accuracy of 84.9%. Conclusions Computed tomography texture analysis may prove valuable in lesion characterization. Differentiation between common hypervascular lesion types could be aided by the judicious incorporation of texture parameters into clinical analysis.

KW - computed tomography

KW - hypervascular liver lesions

KW - quantitative imaging

KW - texture analysis

UR - http://www.scopus.com/inward/record.url?scp=84930069465&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930069465&partnerID=8YFLogxK

U2 - 10.1097/RCT.0000000000000217

DO - 10.1097/RCT.0000000000000217

M3 - Article

C2 - 25700222

AN - SCOPUS:84930069465

VL - 39

SP - 383

EP - 395

JO - Journal of Computer Assisted Tomography

JF - Journal of Computer Assisted Tomography

SN - 0363-8715

IS - 3

ER -