Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI

Satish Viswanath, Daniel Palumbo, Jonathan Chappelow, Pratik Patel, B. Nicholas Bloch, Neil Rofsky, Robert Lenkinski, Elizabeth Genega, Anant Madabhushi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

In magnetic resonance imaging (MRI), intensity inhomogeneity refers to an acquisition artifact which introduces a non-linear variation in the signal intensities within the image. Intensity inhomogeneity is known to significantly affect computerized analysis of MRI data (such as automated segmentation or classification procedures), hence requiring the application of bias field correction (BFC) algorithms to account for this artifact. Quantitative evaluation of BFC schemes is typically performed using generalized intensity-based measures (percent coefficient of variation, %CV) or information-theoretic measures (entropy). While some investigators have previously empirically compared BFC schemes in the context of different domains (using changes in %CV and entropy to quantify improvements), no consensus has emerged as to the best BFC scheme for any given application. The motivation for this work is that the choice of a BFC scheme for a given application should be dictated by application-specific measures rather than ad hoc measures such as entropy and %CV. In this paper, we have attempted to address the problem of determining an optimal BFC algorithm in the context of a computer-aided diagnosis (CAD) scheme for prostate cancer (CaP) detection from T2-weighted (T2w) MRI. One goal of this work is to identify a BFC algorithm that will maximize the CaP classification accuracy (measured in terms of the area under the ROC curve or AUC). A secondary aim of our work is to determine whether measures such as %CV and entropy are correlated with a classifier-based objective measure (AUC). Determining the presence or absence of these correlations is important to understand whether domain independent BFC performance measures such as %CV, entropy should be used to identify the optimal BFC scheme for any given application. In order to answer these questions, we quantitatively compared 3 different popular BFC algorithms on a cohort of 10 clinical 3 Tesla prostate T2w MRI datasets (comprising 39 2D MRI slices): N3, PABIC, and the method of Cohen et al. Results of BFC via each of the algorithms was evaluated in terms of %CV, entropy, as well as classifier AUC for CaP detection from T2w MRI. The CaP classifier was trained and evaluated on a per-pixel basis using annotations of CaP obtained via registration of T2w MRI and ex vivo whole-mount histology sections. Our results revealed that different BFC schemes resulted in a maximization of different performance measures, that is, the BFC scheme identified by minimization of %CV and entropy was not the one that maximized AUC as well. Moreover, existing BFC evaluation measures (%CV, entropy) did not correlate with AUC (application-based evaluation), but did correlate with each other, suggesting that domain-specific performance measures should be considered in making a decision regarding choice of appropriate BFC scheme. Our results also revealed that N3 provided the best correction of bias field artifacts in prostate MRI data, when the goal was to identify prostate cancer.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume7963
DOIs
StatePublished - 2011
EventMedical Imaging 2011: Computer-Aided Diagnosis - Lake Buena Vista, FL, United States
Duration: Feb 15 2011Feb 17 2011

Other

OtherMedical Imaging 2011: Computer-Aided Diagnosis
CountryUnited States
CityLake Buena Vista, FL
Period2/15/112/17/11

Fingerprint

Entropy
Magnetic resonance
magnetic resonance
Prostatic Neoplasms
cancer
Magnetic Resonance Imaging
Area Under Curve
Imaging techniques
evaluation
Artifacts
entropy
Classifiers
Prostate
classifiers
artifacts
Computer aided diagnosis
Histology
ROC Curve
Decision Making
inhomogeneity

Keywords

  • bias field correction
  • classification
  • intensity inhomogeneity
  • prostate cancer
  • T2w MRI

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Viswanath, S., Palumbo, D., Chappelow, J., Patel, P., Bloch, B. N., Rofsky, N., ... Madabhushi, A. (2011). Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 7963). [79630V] https://doi.org/10.1117/12.878813

Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI. / Viswanath, Satish; Palumbo, Daniel; Chappelow, Jonathan; Patel, Pratik; Bloch, B. Nicholas; Rofsky, Neil; Lenkinski, Robert; Genega, Elizabeth; Madabhushi, Anant.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7963 2011. 79630V.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Viswanath, S, Palumbo, D, Chappelow, J, Patel, P, Bloch, BN, Rofsky, N, Lenkinski, R, Genega, E & Madabhushi, A 2011, Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 7963, 79630V, Medical Imaging 2011: Computer-Aided Diagnosis, Lake Buena Vista, FL, United States, 2/15/11. https://doi.org/10.1117/12.878813
Viswanath S, Palumbo D, Chappelow J, Patel P, Bloch BN, Rofsky N et al. Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7963. 2011. 79630V https://doi.org/10.1117/12.878813
Viswanath, Satish ; Palumbo, Daniel ; Chappelow, Jonathan ; Patel, Pratik ; Bloch, B. Nicholas ; Rofsky, Neil ; Lenkinski, Robert ; Genega, Elizabeth ; Madabhushi, Anant. / Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7963 2011.
@inproceedings{8bd820cfee934b6abf2723e12673b140,
title = "Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI",
abstract = "In magnetic resonance imaging (MRI), intensity inhomogeneity refers to an acquisition artifact which introduces a non-linear variation in the signal intensities within the image. Intensity inhomogeneity is known to significantly affect computerized analysis of MRI data (such as automated segmentation or classification procedures), hence requiring the application of bias field correction (BFC) algorithms to account for this artifact. Quantitative evaluation of BFC schemes is typically performed using generalized intensity-based measures (percent coefficient of variation, {\%}CV) or information-theoretic measures (entropy). While some investigators have previously empirically compared BFC schemes in the context of different domains (using changes in {\%}CV and entropy to quantify improvements), no consensus has emerged as to the best BFC scheme for any given application. The motivation for this work is that the choice of a BFC scheme for a given application should be dictated by application-specific measures rather than ad hoc measures such as entropy and {\%}CV. In this paper, we have attempted to address the problem of determining an optimal BFC algorithm in the context of a computer-aided diagnosis (CAD) scheme for prostate cancer (CaP) detection from T2-weighted (T2w) MRI. One goal of this work is to identify a BFC algorithm that will maximize the CaP classification accuracy (measured in terms of the area under the ROC curve or AUC). A secondary aim of our work is to determine whether measures such as {\%}CV and entropy are correlated with a classifier-based objective measure (AUC). Determining the presence or absence of these correlations is important to understand whether domain independent BFC performance measures such as {\%}CV, entropy should be used to identify the optimal BFC scheme for any given application. In order to answer these questions, we quantitatively compared 3 different popular BFC algorithms on a cohort of 10 clinical 3 Tesla prostate T2w MRI datasets (comprising 39 2D MRI slices): N3, PABIC, and the method of Cohen et al. Results of BFC via each of the algorithms was evaluated in terms of {\%}CV, entropy, as well as classifier AUC for CaP detection from T2w MRI. The CaP classifier was trained and evaluated on a per-pixel basis using annotations of CaP obtained via registration of T2w MRI and ex vivo whole-mount histology sections. Our results revealed that different BFC schemes resulted in a maximization of different performance measures, that is, the BFC scheme identified by minimization of {\%}CV and entropy was not the one that maximized AUC as well. Moreover, existing BFC evaluation measures ({\%}CV, entropy) did not correlate with AUC (application-based evaluation), but did correlate with each other, suggesting that domain-specific performance measures should be considered in making a decision regarding choice of appropriate BFC scheme. Our results also revealed that N3 provided the best correction of bias field artifacts in prostate MRI data, when the goal was to identify prostate cancer.",
keywords = "bias field correction, classification, intensity inhomogeneity, prostate cancer, T2w MRI",
author = "Satish Viswanath and Daniel Palumbo and Jonathan Chappelow and Pratik Patel and Bloch, {B. Nicholas} and Neil Rofsky and Robert Lenkinski and Elizabeth Genega and Anant Madabhushi",
year = "2011",
doi = "10.1117/12.878813",
language = "English (US)",
isbn = "9780819485052",
volume = "7963",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI

AU - Viswanath, Satish

AU - Palumbo, Daniel

AU - Chappelow, Jonathan

AU - Patel, Pratik

AU - Bloch, B. Nicholas

AU - Rofsky, Neil

AU - Lenkinski, Robert

AU - Genega, Elizabeth

AU - Madabhushi, Anant

PY - 2011

Y1 - 2011

N2 - In magnetic resonance imaging (MRI), intensity inhomogeneity refers to an acquisition artifact which introduces a non-linear variation in the signal intensities within the image. Intensity inhomogeneity is known to significantly affect computerized analysis of MRI data (such as automated segmentation or classification procedures), hence requiring the application of bias field correction (BFC) algorithms to account for this artifact. Quantitative evaluation of BFC schemes is typically performed using generalized intensity-based measures (percent coefficient of variation, %CV) or information-theoretic measures (entropy). While some investigators have previously empirically compared BFC schemes in the context of different domains (using changes in %CV and entropy to quantify improvements), no consensus has emerged as to the best BFC scheme for any given application. The motivation for this work is that the choice of a BFC scheme for a given application should be dictated by application-specific measures rather than ad hoc measures such as entropy and %CV. In this paper, we have attempted to address the problem of determining an optimal BFC algorithm in the context of a computer-aided diagnosis (CAD) scheme for prostate cancer (CaP) detection from T2-weighted (T2w) MRI. One goal of this work is to identify a BFC algorithm that will maximize the CaP classification accuracy (measured in terms of the area under the ROC curve or AUC). A secondary aim of our work is to determine whether measures such as %CV and entropy are correlated with a classifier-based objective measure (AUC). Determining the presence or absence of these correlations is important to understand whether domain independent BFC performance measures such as %CV, entropy should be used to identify the optimal BFC scheme for any given application. In order to answer these questions, we quantitatively compared 3 different popular BFC algorithms on a cohort of 10 clinical 3 Tesla prostate T2w MRI datasets (comprising 39 2D MRI slices): N3, PABIC, and the method of Cohen et al. Results of BFC via each of the algorithms was evaluated in terms of %CV, entropy, as well as classifier AUC for CaP detection from T2w MRI. The CaP classifier was trained and evaluated on a per-pixel basis using annotations of CaP obtained via registration of T2w MRI and ex vivo whole-mount histology sections. Our results revealed that different BFC schemes resulted in a maximization of different performance measures, that is, the BFC scheme identified by minimization of %CV and entropy was not the one that maximized AUC as well. Moreover, existing BFC evaluation measures (%CV, entropy) did not correlate with AUC (application-based evaluation), but did correlate with each other, suggesting that domain-specific performance measures should be considered in making a decision regarding choice of appropriate BFC scheme. Our results also revealed that N3 provided the best correction of bias field artifacts in prostate MRI data, when the goal was to identify prostate cancer.

AB - In magnetic resonance imaging (MRI), intensity inhomogeneity refers to an acquisition artifact which introduces a non-linear variation in the signal intensities within the image. Intensity inhomogeneity is known to significantly affect computerized analysis of MRI data (such as automated segmentation or classification procedures), hence requiring the application of bias field correction (BFC) algorithms to account for this artifact. Quantitative evaluation of BFC schemes is typically performed using generalized intensity-based measures (percent coefficient of variation, %CV) or information-theoretic measures (entropy). While some investigators have previously empirically compared BFC schemes in the context of different domains (using changes in %CV and entropy to quantify improvements), no consensus has emerged as to the best BFC scheme for any given application. The motivation for this work is that the choice of a BFC scheme for a given application should be dictated by application-specific measures rather than ad hoc measures such as entropy and %CV. In this paper, we have attempted to address the problem of determining an optimal BFC algorithm in the context of a computer-aided diagnosis (CAD) scheme for prostate cancer (CaP) detection from T2-weighted (T2w) MRI. One goal of this work is to identify a BFC algorithm that will maximize the CaP classification accuracy (measured in terms of the area under the ROC curve or AUC). A secondary aim of our work is to determine whether measures such as %CV and entropy are correlated with a classifier-based objective measure (AUC). Determining the presence or absence of these correlations is important to understand whether domain independent BFC performance measures such as %CV, entropy should be used to identify the optimal BFC scheme for any given application. In order to answer these questions, we quantitatively compared 3 different popular BFC algorithms on a cohort of 10 clinical 3 Tesla prostate T2w MRI datasets (comprising 39 2D MRI slices): N3, PABIC, and the method of Cohen et al. Results of BFC via each of the algorithms was evaluated in terms of %CV, entropy, as well as classifier AUC for CaP detection from T2w MRI. The CaP classifier was trained and evaluated on a per-pixel basis using annotations of CaP obtained via registration of T2w MRI and ex vivo whole-mount histology sections. Our results revealed that different BFC schemes resulted in a maximization of different performance measures, that is, the BFC scheme identified by minimization of %CV and entropy was not the one that maximized AUC as well. Moreover, existing BFC evaluation measures (%CV, entropy) did not correlate with AUC (application-based evaluation), but did correlate with each other, suggesting that domain-specific performance measures should be considered in making a decision regarding choice of appropriate BFC scheme. Our results also revealed that N3 provided the best correction of bias field artifacts in prostate MRI data, when the goal was to identify prostate cancer.

KW - bias field correction

KW - classification

KW - intensity inhomogeneity

KW - prostate cancer

KW - T2w MRI

UR - http://www.scopus.com/inward/record.url?scp=79955758439&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955758439&partnerID=8YFLogxK

U2 - 10.1117/12.878813

DO - 10.1117/12.878813

M3 - Conference contribution

AN - SCOPUS:79955758439

SN - 9780819485052

VL - 7963

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -