Hin zu einem universellen MRT-Atlas von Prostata und Prostatazonen: Vergleich von verschiedenen MRT-Herstellern und Bildaufnahmeparametern

Translated title of the contribution: Towards a universal MRI atlas of the prostate and prostate zones: Comparison of MRI vendor and image acquisition parameters

Kyle Padgett, Amy Swallen, Sara Pirozzi, Jon Piper, Felix M. Chinea, Matthew C Abramowitz, Aaron Nelson, Alan Pollack, Radka Stoyanova

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background and purpose: The aim of this study was to evaluate an automatic multi-atlas-based segmentation method for generating prostate, peripheral (PZ), and transition zone (TZ) contours on MRIs with and without fat saturation (±FS), and compare MRIs from different vendor MRI systems. Methods: T2-weighted (T2) and fat-saturated (T2FS) MRIs were acquired on 3T GE (GE, Waukesha, WI, USA) and Siemens (Erlangen, Germany) systems. Manual prostate and PZ contours were used to create atlas libraries. As a test MRI is entered, the procedure for atlas segmentation automatically identifies the atlas subjects that best match the test subject, followed by a normalized intensity-based free-form deformable registration. The contours are transformed to the test subject, and Dice similarity coefficients (DSC) and Hausdorff distances between atlas-generated and manual contours were used to assess performance. Results: Three atlases were generated based on GE_T2 (n = 30), GE_T2FS (n = 30), and Siem_T2FS (n = 31). When test images matched the contrast and vendor of the atlas, DSCs of 0.81 and 0.83 for T2 ± FS were obtained (baseline performance). Atlases performed with higher accuracy when segmenting (i) T2FS vs. T2 images, likely due to a superior contrast between prostate vs. surrounding tissue; (ii) prostate vs. zonal anatomy; (iii) in the mid-gland vs. base and apex. Atlases performance declined when tested with images with differing contrast and MRI vendor. Conversely, combined atlases showed similar performance to baseline. Conclusion: The MRI atlas-based segmentation method achieved good results for prostate, PZ, and TZ compared to expert contoured volumes. Combined atlases performed similarly to matching atlas and scan type. The technique is fast, fully automatic, and implemented on commercially available clinical platform.

Original languageGerman
JournalStrahlentherapie und Onkologie
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Atlases
Prostate
Fats
Libraries
Germany
Anatomy

Keywords

  • Magnetic resonance imaging
  • Prostate neoplasms
  • Radiation oncologists
  • Radiotherapy
  • Segmentation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Oncology

Cite this

@article{1d57ecd7ac2b42eb887f1a1d9a61a586,
title = "Hin zu einem universellen MRT-Atlas von Prostata und Prostatazonen: Vergleich von verschiedenen MRT-Herstellern und Bildaufnahmeparametern",
abstract = "Background and purpose: The aim of this study was to evaluate an automatic multi-atlas-based segmentation method for generating prostate, peripheral (PZ), and transition zone (TZ) contours on MRIs with and without fat saturation (±FS), and compare MRIs from different vendor MRI systems. Methods: T2-weighted (T2) and fat-saturated (T2FS) MRIs were acquired on 3T GE (GE, Waukesha, WI, USA) and Siemens (Erlangen, Germany) systems. Manual prostate and PZ contours were used to create atlas libraries. As a test MRI is entered, the procedure for atlas segmentation automatically identifies the atlas subjects that best match the test subject, followed by a normalized intensity-based free-form deformable registration. The contours are transformed to the test subject, and Dice similarity coefficients (DSC) and Hausdorff distances between atlas-generated and manual contours were used to assess performance. Results: Three atlases were generated based on GE_T2 (n = 30), GE_T2FS (n = 30), and Siem_T2FS (n = 31). When test images matched the contrast and vendor of the atlas, DSCs of 0.81 and 0.83 for T2 ± FS were obtained (baseline performance). Atlases performed with higher accuracy when segmenting (i) T2FS vs. T2 images, likely due to a superior contrast between prostate vs. surrounding tissue; (ii) prostate vs. zonal anatomy; (iii) in the mid-gland vs. base and apex. Atlases performance declined when tested with images with differing contrast and MRI vendor. Conversely, combined atlases showed similar performance to baseline. Conclusion: The MRI atlas-based segmentation method achieved good results for prostate, PZ, and TZ compared to expert contoured volumes. Combined atlases performed similarly to matching atlas and scan type. The technique is fast, fully automatic, and implemented on commercially available clinical platform.",
keywords = "Magnetic resonance imaging, Prostate neoplasms, Radiation oncologists, Radiotherapy, Segmentation",
author = "Kyle Padgett and Amy Swallen and Sara Pirozzi and Jon Piper and Chinea, {Felix M.} and Abramowitz, {Matthew C} and Aaron Nelson and Alan Pollack and Radka Stoyanova",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/s00066-018-1348-5",
language = "German",
journal = "Strahlentherapie und Onkologie",
issn = "0179-7158",
publisher = "Urban und Vogel",

}

TY - JOUR

T1 - Hin zu einem universellen MRT-Atlas von Prostata und Prostatazonen

T2 - Vergleich von verschiedenen MRT-Herstellern und Bildaufnahmeparametern

AU - Padgett, Kyle

AU - Swallen, Amy

AU - Pirozzi, Sara

AU - Piper, Jon

AU - Chinea, Felix M.

AU - Abramowitz, Matthew C

AU - Nelson, Aaron

AU - Pollack, Alan

AU - Stoyanova, Radka

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Background and purpose: The aim of this study was to evaluate an automatic multi-atlas-based segmentation method for generating prostate, peripheral (PZ), and transition zone (TZ) contours on MRIs with and without fat saturation (±FS), and compare MRIs from different vendor MRI systems. Methods: T2-weighted (T2) and fat-saturated (T2FS) MRIs were acquired on 3T GE (GE, Waukesha, WI, USA) and Siemens (Erlangen, Germany) systems. Manual prostate and PZ contours were used to create atlas libraries. As a test MRI is entered, the procedure for atlas segmentation automatically identifies the atlas subjects that best match the test subject, followed by a normalized intensity-based free-form deformable registration. The contours are transformed to the test subject, and Dice similarity coefficients (DSC) and Hausdorff distances between atlas-generated and manual contours were used to assess performance. Results: Three atlases were generated based on GE_T2 (n = 30), GE_T2FS (n = 30), and Siem_T2FS (n = 31). When test images matched the contrast and vendor of the atlas, DSCs of 0.81 and 0.83 for T2 ± FS were obtained (baseline performance). Atlases performed with higher accuracy when segmenting (i) T2FS vs. T2 images, likely due to a superior contrast between prostate vs. surrounding tissue; (ii) prostate vs. zonal anatomy; (iii) in the mid-gland vs. base and apex. Atlases performance declined when tested with images with differing contrast and MRI vendor. Conversely, combined atlases showed similar performance to baseline. Conclusion: The MRI atlas-based segmentation method achieved good results for prostate, PZ, and TZ compared to expert contoured volumes. Combined atlases performed similarly to matching atlas and scan type. The technique is fast, fully automatic, and implemented on commercially available clinical platform.

AB - Background and purpose: The aim of this study was to evaluate an automatic multi-atlas-based segmentation method for generating prostate, peripheral (PZ), and transition zone (TZ) contours on MRIs with and without fat saturation (±FS), and compare MRIs from different vendor MRI systems. Methods: T2-weighted (T2) and fat-saturated (T2FS) MRIs were acquired on 3T GE (GE, Waukesha, WI, USA) and Siemens (Erlangen, Germany) systems. Manual prostate and PZ contours were used to create atlas libraries. As a test MRI is entered, the procedure for atlas segmentation automatically identifies the atlas subjects that best match the test subject, followed by a normalized intensity-based free-form deformable registration. The contours are transformed to the test subject, and Dice similarity coefficients (DSC) and Hausdorff distances between atlas-generated and manual contours were used to assess performance. Results: Three atlases were generated based on GE_T2 (n = 30), GE_T2FS (n = 30), and Siem_T2FS (n = 31). When test images matched the contrast and vendor of the atlas, DSCs of 0.81 and 0.83 for T2 ± FS were obtained (baseline performance). Atlases performed with higher accuracy when segmenting (i) T2FS vs. T2 images, likely due to a superior contrast between prostate vs. surrounding tissue; (ii) prostate vs. zonal anatomy; (iii) in the mid-gland vs. base and apex. Atlases performance declined when tested with images with differing contrast and MRI vendor. Conversely, combined atlases showed similar performance to baseline. Conclusion: The MRI atlas-based segmentation method achieved good results for prostate, PZ, and TZ compared to expert contoured volumes. Combined atlases performed similarly to matching atlas and scan type. The technique is fast, fully automatic, and implemented on commercially available clinical platform.

KW - Magnetic resonance imaging

KW - Prostate neoplasms

KW - Radiation oncologists

KW - Radiotherapy

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85052336345&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052336345&partnerID=8YFLogxK

U2 - 10.1007/s00066-018-1348-5

DO - 10.1007/s00066-018-1348-5

M3 - Article

C2 - 30140944

AN - SCOPUS:85052336345

JO - Strahlentherapie und Onkologie

JF - Strahlentherapie und Onkologie

SN - 0179-7158

ER -