Validation of Automatic Target Volume Definition as Demonstrated for 11C-Choline PET/CT of Human Prostate Cancer Using Multi-modality Fusion Techniques

Park, Hyunjin; Meyer, Charles R.; Wood, David; Khan, Asra; Shah, Rajal; Hussain, Hero; Siddiqui, Javed; Seo, Jongbum; Chenevert, Thomas; Piert, Morand

doi:10.1016/j.acra.2010.01.003

« Previous Next »Academic Radiology
Volume 17, Issue 5 , Pages 614-623, May 2010

Validation of Automatic Target Volume Definition as Demonstrated for ¹¹C-Choline PET/CT of Human Prostate Cancer Using Multi-modality Fusion Techniques

Hyunjin Park, PhD
- Department of Biomedical Engineering, Gachon University of Medicine and Science, Incheon 406-799, South Korea
- Department of Radiology, University of Michigan, Ann Arbor, MI
- Address correspondence to: H.P.
Charles R. Meyer, PhD
- Department of Radiology, University of Michigan, Ann Arbor, MI
David Wood, MD
- Department of Urology, University of Michigan, Ann Arbor, MI
Asra Khan, MD
- Department of Radiology, University of Michigan, Ann Arbor, MI
Rajal Shah, MD
- Department of Pathology, University of Michigan, Ann Arbor, MI
Hero Hussain, MD
- Department of Radiology, University of Michigan, Ann Arbor, MI
Javed Siddiqui, MS
- Department of Pathology, University of Michigan, Ann Arbor, MI
Jongbum Seo, PhD
- Department of Biomedical Engineering, Yonsei University, Wonju, South Korea
Thomas Chenevert, PhD
- Department of Radiology, University of Michigan, Ann Arbor, MI
Morand Piert, MD
- Department of Radiology, University of Michigan, Ann Arbor, MI

Received 30 November 2009; accepted 6 January 2010. published online 02 March 2010.

Rationale and Objectives

Positron emission tomography (PET) is actively investigated to aid in target volume definition for radiation therapy. The objectives of this study were to apply an automatic computer algorithm to compute target volumes and to validate the algorithm using histologic data from real human prostate cancer.

Materials and Methods

Various modalities for prostate imaging were performed. In vivo imaging included T2 3-T magnetic resonance imaging and ¹¹C-choline PET. Ex vivo imaging included 3-T magnetic resonance imaging, histology, and block face photos of the prostate specimen. A novel registration method based on mutual information and thin-plate splines was applied to all modalities. Once PET is registered with histology, a voxel-by-voxel comparison between PET and histology is possible. A thresholding technique based on various fractions of the maximum standardized uptake value in the tumor was applied, and the respective computed threshold volume on PET was compared with histologic truth.

Results

Sixteen patients whose primary tumor volumes ranged from 1.2 to 12.6 cm³ were tested. PET has low spatial resolution, so only tumors > 4 cm³ were considered. Four cases met this criterion. A threshold value of 60% of the ¹¹C-choline maximum standardized uptake value resulted in the highest volume overlap between threshold volume on PET and histology. Medial axis distances between threshold volume on PET and histology showed a mean error of 7.7 ± 5.2 mm.

Conclusions

This is a proof-of-concept study demonstrating for the first time that histology-guided thresholding on PET can delineate tumor volumes in real human prostate cancer.

Key Words: Automatic target volume definition, image registration, prostate cancer, histology, thresholding, PET, MRI, multi-modality image fusion