Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI

Satish Viswanath, B. Nicolas Bloch, Mark Rosen, Jonathan Chappelow, Robert Toth, Neil Rofsky, Robert Lenkinski, Elisabeth Genega, Arjun Kalyanpur, Anant Madabhushi

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

29 Citations (Scopus)

Abstract

Screening and detection of prostate cancer (CaP) currently lacks an image-based protocol which is reflected in the high false negative rates currently associated with blinded sextant biopsies. Multi-protocol magnetic resonance imaging (MRI) offers high resolution functional and structural data about internal body structures (such as the prostate). In this paper we present a novel comprehensive computer-aided scheme for CaP detection from high resolution in vivo multi-protocol MRI by integrating functional and structural information obtained via dynamic-contrast enhanced (DCE) and T2-weighted (T2-w) MRI, respectively. Our scheme is fully-automated and comprises (a) prostate segmentation, (b) multimodal image registration, and (c) data representation and multi-classifier modules for information fusion. Following prostate boundary segmentation via an improved active shape model, the DCE/T2-w protocols and the T2-w/ex vivo histological prostatectomy specimens are brought into alignment via a deformable, multi-attribute registration scheme. T2-w/histology alignment allows for the mapping of true CaP extent onto the in vivo MRI, which is used for training and evaluation of a multi-protocol MRI CaP classifier. The meta-classifier used is a random forest constructed by bagging multiple decision tree classifiers, each trained individually on T2-w structural, textural and DCE functional attributes. 3-fold classifier cross validation was performed using a set of 18 images derived from 6 patient datasets on a per-pixel basis. Our results show that the results of CaP detection obtained from integration of T2-w structural textural data and DCE functional data (area under the ROC curve of 0.815) significantly outperforms detection based on either of the individual modalities (0.704 (T2-w) and 0.682 (DCE)). It was also found that a meta-classifier trained directly on integrated T2-w and DCE data (data-level integration) significantly outperformed a decision-level meta-classifier, constructed by combining the classifier outputs from the individual T2-w and DCE channels.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume7260
DOIs
StatePublished - 2009
EventMedical Imaging 2009: Computer-Aided Diagnosis - Lake Buena Vista, FL, United States
Duration: Feb 10 2009Feb 12 2009

Other

OtherMedical Imaging 2009: Computer-Aided Diagnosis
CountryUnited States
CityLake Buena Vista, FL
Period2/10/092/12/09

Fingerprint

Magnetic resonance
classifiers
magnetic resonance
Prostatic Neoplasms
Classifiers
cancer
Magnetic Resonance Imaging
Imaging techniques
Prostate
Decision Trees
Sextants
Prostatectomy
sextants
ROC Curve
alignment
Area Under Curve
Histology
histology
Information fusion
Biopsy

Keywords

  • 3 Tesla
  • Bagging
  • CAD
  • Data fusion
  • DCE-MRI
  • Decision fusion
  • Decision trees
  • Multimodal integration
  • Non-rigid registration
  • Prostate cancer
  • Random forests
  • Segmentation
  • Supervised learning
  • T2-w 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., Bloch, B. N., Rosen, M., Chappelow, J., Toth, R., Rofsky, N., ... Madabhushi, A. (2009). Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 7260). [72603I] https://doi.org/10.1117/12.811899

Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI. / Viswanath, Satish; Bloch, B. Nicolas; Rosen, Mark; Chappelow, Jonathan; Toth, Robert; Rofsky, Neil; Lenkinski, Robert; Genega, Elisabeth; Kalyanpur, Arjun; Madabhushi, Anant.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7260 2009. 72603I.

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

Viswanath, S, Bloch, BN, Rosen, M, Chappelow, J, Toth, R, Rofsky, N, Lenkinski, R, Genega, E, Kalyanpur, A & Madabhushi, A 2009, Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 7260, 72603I, Medical Imaging 2009: Computer-Aided Diagnosis, Lake Buena Vista, FL, United States, 2/10/09. https://doi.org/10.1117/12.811899
Viswanath S, Bloch BN, Rosen M, Chappelow J, Toth R, Rofsky N et al. Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7260. 2009. 72603I https://doi.org/10.1117/12.811899
Viswanath, Satish ; Bloch, B. Nicolas ; Rosen, Mark ; Chappelow, Jonathan ; Toth, Robert ; Rofsky, Neil ; Lenkinski, Robert ; Genega, Elisabeth ; Kalyanpur, Arjun ; Madabhushi, Anant. / Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7260 2009.
@inproceedings{e1da104cd029478abe18ec5bc52b9253,
title = "Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI",
abstract = "Screening and detection of prostate cancer (CaP) currently lacks an image-based protocol which is reflected in the high false negative rates currently associated with blinded sextant biopsies. Multi-protocol magnetic resonance imaging (MRI) offers high resolution functional and structural data about internal body structures (such as the prostate). In this paper we present a novel comprehensive computer-aided scheme for CaP detection from high resolution in vivo multi-protocol MRI by integrating functional and structural information obtained via dynamic-contrast enhanced (DCE) and T2-weighted (T2-w) MRI, respectively. Our scheme is fully-automated and comprises (a) prostate segmentation, (b) multimodal image registration, and (c) data representation and multi-classifier modules for information fusion. Following prostate boundary segmentation via an improved active shape model, the DCE/T2-w protocols and the T2-w/ex vivo histological prostatectomy specimens are brought into alignment via a deformable, multi-attribute registration scheme. T2-w/histology alignment allows for the mapping of true CaP extent onto the in vivo MRI, which is used for training and evaluation of a multi-protocol MRI CaP classifier. The meta-classifier used is a random forest constructed by bagging multiple decision tree classifiers, each trained individually on T2-w structural, textural and DCE functional attributes. 3-fold classifier cross validation was performed using a set of 18 images derived from 6 patient datasets on a per-pixel basis. Our results show that the results of CaP detection obtained from integration of T2-w structural textural data and DCE functional data (area under the ROC curve of 0.815) significantly outperforms detection based on either of the individual modalities (0.704 (T2-w) and 0.682 (DCE)). It was also found that a meta-classifier trained directly on integrated T2-w and DCE data (data-level integration) significantly outperformed a decision-level meta-classifier, constructed by combining the classifier outputs from the individual T2-w and DCE channels.",
keywords = "3 Tesla, Bagging, CAD, Data fusion, DCE-MRI, Decision fusion, Decision trees, Multimodal integration, Non-rigid registration, Prostate cancer, Random forests, Segmentation, Supervised learning, T2-w MRI",
author = "Satish Viswanath and Bloch, {B. Nicolas} and Mark Rosen and Jonathan Chappelow and Robert Toth and Neil Rofsky and Robert Lenkinski and Elisabeth Genega and Arjun Kalyanpur and Anant Madabhushi",
year = "2009",
doi = "10.1117/12.811899",
language = "English (US)",
isbn = "9780819475114",
volume = "7260",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Integrating structural and functional imaging for computer assisted detection of prostate cancer on multi-protocol in vivo 3 tesla MRI

AU - Viswanath, Satish

AU - Bloch, B. Nicolas

AU - Rosen, Mark

AU - Chappelow, Jonathan

AU - Toth, Robert

AU - Rofsky, Neil

AU - Lenkinski, Robert

AU - Genega, Elisabeth

AU - Kalyanpur, Arjun

AU - Madabhushi, Anant

PY - 2009

Y1 - 2009

N2 - Screening and detection of prostate cancer (CaP) currently lacks an image-based protocol which is reflected in the high false negative rates currently associated with blinded sextant biopsies. Multi-protocol magnetic resonance imaging (MRI) offers high resolution functional and structural data about internal body structures (such as the prostate). In this paper we present a novel comprehensive computer-aided scheme for CaP detection from high resolution in vivo multi-protocol MRI by integrating functional and structural information obtained via dynamic-contrast enhanced (DCE) and T2-weighted (T2-w) MRI, respectively. Our scheme is fully-automated and comprises (a) prostate segmentation, (b) multimodal image registration, and (c) data representation and multi-classifier modules for information fusion. Following prostate boundary segmentation via an improved active shape model, the DCE/T2-w protocols and the T2-w/ex vivo histological prostatectomy specimens are brought into alignment via a deformable, multi-attribute registration scheme. T2-w/histology alignment allows for the mapping of true CaP extent onto the in vivo MRI, which is used for training and evaluation of a multi-protocol MRI CaP classifier. The meta-classifier used is a random forest constructed by bagging multiple decision tree classifiers, each trained individually on T2-w structural, textural and DCE functional attributes. 3-fold classifier cross validation was performed using a set of 18 images derived from 6 patient datasets on a per-pixel basis. Our results show that the results of CaP detection obtained from integration of T2-w structural textural data and DCE functional data (area under the ROC curve of 0.815) significantly outperforms detection based on either of the individual modalities (0.704 (T2-w) and 0.682 (DCE)). It was also found that a meta-classifier trained directly on integrated T2-w and DCE data (data-level integration) significantly outperformed a decision-level meta-classifier, constructed by combining the classifier outputs from the individual T2-w and DCE channels.

AB - Screening and detection of prostate cancer (CaP) currently lacks an image-based protocol which is reflected in the high false negative rates currently associated with blinded sextant biopsies. Multi-protocol magnetic resonance imaging (MRI) offers high resolution functional and structural data about internal body structures (such as the prostate). In this paper we present a novel comprehensive computer-aided scheme for CaP detection from high resolution in vivo multi-protocol MRI by integrating functional and structural information obtained via dynamic-contrast enhanced (DCE) and T2-weighted (T2-w) MRI, respectively. Our scheme is fully-automated and comprises (a) prostate segmentation, (b) multimodal image registration, and (c) data representation and multi-classifier modules for information fusion. Following prostate boundary segmentation via an improved active shape model, the DCE/T2-w protocols and the T2-w/ex vivo histological prostatectomy specimens are brought into alignment via a deformable, multi-attribute registration scheme. T2-w/histology alignment allows for the mapping of true CaP extent onto the in vivo MRI, which is used for training and evaluation of a multi-protocol MRI CaP classifier. The meta-classifier used is a random forest constructed by bagging multiple decision tree classifiers, each trained individually on T2-w structural, textural and DCE functional attributes. 3-fold classifier cross validation was performed using a set of 18 images derived from 6 patient datasets on a per-pixel basis. Our results show that the results of CaP detection obtained from integration of T2-w structural textural data and DCE functional data (area under the ROC curve of 0.815) significantly outperforms detection based on either of the individual modalities (0.704 (T2-w) and 0.682 (DCE)). It was also found that a meta-classifier trained directly on integrated T2-w and DCE data (data-level integration) significantly outperformed a decision-level meta-classifier, constructed by combining the classifier outputs from the individual T2-w and DCE channels.

KW - 3 Tesla

KW - Bagging

KW - CAD

KW - Data fusion

KW - DCE-MRI

KW - Decision fusion

KW - Decision trees

KW - Multimodal integration

KW - Non-rigid registration

KW - Prostate cancer

KW - Random forests

KW - Segmentation

KW - Supervised learning

KW - T2-w MRI

UR - http://www.scopus.com/inward/record.url?scp=66749101612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=66749101612&partnerID=8YFLogxK

U2 - 10.1117/12.811899

DO - 10.1117/12.811899

M3 - Conference contribution

C2 - 25301989

AN - SCOPUS:66749101612

SN - 9780819475114

VL - 7260

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

ER -