Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging

Alberto Santamaria-Pang, Sandeep Dutta, Sokratis Makrogiannis, Amy Hara, William Pavlicek, Alvin C Silva, Brian Thomsen, Scott Robertson, Darin Okerlund, David A. Langan, Rahul Bhotika

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

9 Citations (Scopus)

Abstract

Hypodense metastases are not always completely distinguishable from benign cysts in the liver using conventional Computed Tomography (CT) imaging, since the two lesion types present with overlapping intensity distributions due to similar composition as well as other factors including beam hardening and patient motion. This problem is extremely challenging for small lesions with diameter less than 1 cm. To accurately characterize such lesions, multiple follow-up CT scans or additional Positron Emission Tomography or Magnetic Resonance Imaging exam are often conducted, and in some cases a biopsy may be required after the initial CT finding. Gemstone Spectral Imaging (GSI) with fast kVp switching enables projection-based material decomposition, offering the opportunity to discriminate tissue types based on their energy-sensitive material attenuation and density. GSI can be used to obtain monochromatic images where beam hardening is reduced or eliminated and the images come inherently pre-registered due to the fast kVp switching acquisition. We present a supervised learning method for discriminating between cysts and hypodense liver metastases using these monochromatic images. Intensity-based statistical features extracted from voxels inside the lesion are used to train optimal linear and nonlinear classifiers. Our algorithm only requires a region of interest within the lesion in order to compute relevant features and perform classification, thus eliminating the need for an accurate segmentation of the lesion. We report classifier performance using M-fold cross-validation on a large lesion database with radiologist-provided lesion location and labels as the reference standard. Our results demonstrate that (a) classification using a single projection-based spectral CT image, i.e., a monochromatic image at a specified keV, outperforms classification using an image-based dual energy CT pair, i.e., low and high kVp images derived from the same fast kVp acquisition and (b) classification using monochromatic images can achieve very high accuracy in separating benign liver cysts and metastases, especially for small lesions.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
PublisherSPIE
Volume7624
ISBN (Print)9780819480255
DOIs
StatePublished - 2010
EventMedical Imaging 2010: Computer-Aided Diagnosis - San Diego, United States
Duration: Feb 16 2010Feb 18 2010

Other

OtherMedical Imaging 2010: Computer-Aided Diagnosis
CountryUnited States
CitySan Diego
Period2/16/102/18/10

Fingerprint

liver
Liver
lesions
Tomography
tomography
Imaging techniques
Cysts
Neoplasm Metastasis
cysts
metastasis
Hardening
Classifiers
energy
Positron emission tomography
Biopsy
classifiers
Supervised learning
hardening
Positron-Emission Tomography
acquisition

Keywords

  • Gemstone Spectral Imaging CT
  • Liver Lesion Classification
  • Projection-Based Material Decomposition

ASJC Scopus subject areas

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

Cite this

Santamaria-Pang, A., Dutta, S., Makrogiannis, S., Hara, A., Pavlicek, W., Silva, A. C., ... Bhotika, R. (2010). Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 7624). [76240V] SPIE. https://doi.org/10.1117/12.844059

Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging. / Santamaria-Pang, Alberto; Dutta, Sandeep; Makrogiannis, Sokratis; Hara, Amy; Pavlicek, William; Silva, Alvin C; Thomsen, Brian; Robertson, Scott; Okerlund, Darin; Langan, David A.; Bhotika, Rahul.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7624 SPIE, 2010. 76240V.

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

Santamaria-Pang, A, Dutta, S, Makrogiannis, S, Hara, A, Pavlicek, W, Silva, AC, Thomsen, B, Robertson, S, Okerlund, D, Langan, DA & Bhotika, R 2010, Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 7624, 76240V, SPIE, Medical Imaging 2010: Computer-Aided Diagnosis, San Diego, United States, 2/16/10. https://doi.org/10.1117/12.844059
Santamaria-Pang A, Dutta S, Makrogiannis S, Hara A, Pavlicek W, Silva AC et al. Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7624. SPIE. 2010. 76240V https://doi.org/10.1117/12.844059
Santamaria-Pang, Alberto ; Dutta, Sandeep ; Makrogiannis, Sokratis ; Hara, Amy ; Pavlicek, William ; Silva, Alvin C ; Thomsen, Brian ; Robertson, Scott ; Okerlund, Darin ; Langan, David A. ; Bhotika, Rahul. / Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7624 SPIE, 2010.
@inproceedings{f6145168913249348e5861dcfe4b75a2,
title = "Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging",
abstract = "Hypodense metastases are not always completely distinguishable from benign cysts in the liver using conventional Computed Tomography (CT) imaging, since the two lesion types present with overlapping intensity distributions due to similar composition as well as other factors including beam hardening and patient motion. This problem is extremely challenging for small lesions with diameter less than 1 cm. To accurately characterize such lesions, multiple follow-up CT scans or additional Positron Emission Tomography or Magnetic Resonance Imaging exam are often conducted, and in some cases a biopsy may be required after the initial CT finding. Gemstone Spectral Imaging (GSI) with fast kVp switching enables projection-based material decomposition, offering the opportunity to discriminate tissue types based on their energy-sensitive material attenuation and density. GSI can be used to obtain monochromatic images where beam hardening is reduced or eliminated and the images come inherently pre-registered due to the fast kVp switching acquisition. We present a supervised learning method for discriminating between cysts and hypodense liver metastases using these monochromatic images. Intensity-based statistical features extracted from voxels inside the lesion are used to train optimal linear and nonlinear classifiers. Our algorithm only requires a region of interest within the lesion in order to compute relevant features and perform classification, thus eliminating the need for an accurate segmentation of the lesion. We report classifier performance using M-fold cross-validation on a large lesion database with radiologist-provided lesion location and labels as the reference standard. Our results demonstrate that (a) classification using a single projection-based spectral CT image, i.e., a monochromatic image at a specified keV, outperforms classification using an image-based dual energy CT pair, i.e., low and high kVp images derived from the same fast kVp acquisition and (b) classification using monochromatic images can achieve very high accuracy in separating benign liver cysts and metastases, especially for small lesions.",
keywords = "Gemstone Spectral Imaging CT, Liver Lesion Classification, Projection-Based Material Decomposition",
author = "Alberto Santamaria-Pang and Sandeep Dutta and Sokratis Makrogiannis and Amy Hara and William Pavlicek and Silva, {Alvin C} and Brian Thomsen and Scott Robertson and Darin Okerlund and Langan, {David A.} and Rahul Bhotika",
year = "2010",
doi = "10.1117/12.844059",
language = "English (US)",
isbn = "9780819480255",
volume = "7624",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",

}

TY - GEN

T1 - Automated liver lesion characterization using fast kVp switching dual energy computed tomography imaging

AU - Santamaria-Pang, Alberto

AU - Dutta, Sandeep

AU - Makrogiannis, Sokratis

AU - Hara, Amy

AU - Pavlicek, William

AU - Silva, Alvin C

AU - Thomsen, Brian

AU - Robertson, Scott

AU - Okerlund, Darin

AU - Langan, David A.

AU - Bhotika, Rahul

PY - 2010

Y1 - 2010

N2 - Hypodense metastases are not always completely distinguishable from benign cysts in the liver using conventional Computed Tomography (CT) imaging, since the two lesion types present with overlapping intensity distributions due to similar composition as well as other factors including beam hardening and patient motion. This problem is extremely challenging for small lesions with diameter less than 1 cm. To accurately characterize such lesions, multiple follow-up CT scans or additional Positron Emission Tomography or Magnetic Resonance Imaging exam are often conducted, and in some cases a biopsy may be required after the initial CT finding. Gemstone Spectral Imaging (GSI) with fast kVp switching enables projection-based material decomposition, offering the opportunity to discriminate tissue types based on their energy-sensitive material attenuation and density. GSI can be used to obtain monochromatic images where beam hardening is reduced or eliminated and the images come inherently pre-registered due to the fast kVp switching acquisition. We present a supervised learning method for discriminating between cysts and hypodense liver metastases using these monochromatic images. Intensity-based statistical features extracted from voxels inside the lesion are used to train optimal linear and nonlinear classifiers. Our algorithm only requires a region of interest within the lesion in order to compute relevant features and perform classification, thus eliminating the need for an accurate segmentation of the lesion. We report classifier performance using M-fold cross-validation on a large lesion database with radiologist-provided lesion location and labels as the reference standard. Our results demonstrate that (a) classification using a single projection-based spectral CT image, i.e., a monochromatic image at a specified keV, outperforms classification using an image-based dual energy CT pair, i.e., low and high kVp images derived from the same fast kVp acquisition and (b) classification using monochromatic images can achieve very high accuracy in separating benign liver cysts and metastases, especially for small lesions.

AB - Hypodense metastases are not always completely distinguishable from benign cysts in the liver using conventional Computed Tomography (CT) imaging, since the two lesion types present with overlapping intensity distributions due to similar composition as well as other factors including beam hardening and patient motion. This problem is extremely challenging for small lesions with diameter less than 1 cm. To accurately characterize such lesions, multiple follow-up CT scans or additional Positron Emission Tomography or Magnetic Resonance Imaging exam are often conducted, and in some cases a biopsy may be required after the initial CT finding. Gemstone Spectral Imaging (GSI) with fast kVp switching enables projection-based material decomposition, offering the opportunity to discriminate tissue types based on their energy-sensitive material attenuation and density. GSI can be used to obtain monochromatic images where beam hardening is reduced or eliminated and the images come inherently pre-registered due to the fast kVp switching acquisition. We present a supervised learning method for discriminating between cysts and hypodense liver metastases using these monochromatic images. Intensity-based statistical features extracted from voxels inside the lesion are used to train optimal linear and nonlinear classifiers. Our algorithm only requires a region of interest within the lesion in order to compute relevant features and perform classification, thus eliminating the need for an accurate segmentation of the lesion. We report classifier performance using M-fold cross-validation on a large lesion database with radiologist-provided lesion location and labels as the reference standard. Our results demonstrate that (a) classification using a single projection-based spectral CT image, i.e., a monochromatic image at a specified keV, outperforms classification using an image-based dual energy CT pair, i.e., low and high kVp images derived from the same fast kVp acquisition and (b) classification using monochromatic images can achieve very high accuracy in separating benign liver cysts and metastases, especially for small lesions.

KW - Gemstone Spectral Imaging CT

KW - Liver Lesion Classification

KW - Projection-Based Material Decomposition

UR - http://www.scopus.com/inward/record.url?scp=79953653269&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953653269&partnerID=8YFLogxK

U2 - 10.1117/12.844059

DO - 10.1117/12.844059

M3 - Conference contribution

AN - SCOPUS:79953653269

SN - 9780819480255

VL - 7624

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

PB - SPIE

ER -