RT Journal Article
SR Electronic
T1 Intraprostatic Tumor Segmentation on PSMA PET Images in Patients with Primary Prostate Cancer with a Convolutional Neural Network
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 823
OP 828
DO 10.2967/jnumed.120.254623
VO 62
IS 6
A1 Kostyszyn, Dejan
A1 Fechter, Tobias
A1 Bartl, Nico
A1 Grosu, Anca L.
A1 Gratzke, Christian
A1 Sigle, August
A1 Mix, Michael
A1 Ruf, Juri
A1 Fassbender, Thomas F.
A1 Kiefer, Selina
A1 Bettermann, Alisa S.
A1 Nicolay, Nils H.
A1 Spohn, Simon
A1 Kramer, Maria U.
A1 Bronsert, Peter
A1 Guo, Hongqian
A1 Qiu, Xuefeng
A1 Wang, Feng
A1 Henkenberens, Christoph
A1 Werner, Rudolf A.
A1 Baltas, Dimos
A1 Meyer, Philipp T.
A1 Derlin, Thorsten
A1 Chen, Mengxia
A1 Zamboglou, Constantinos
YR 2021
UL http://jnm.snmjournals.org/content/62/6/823.abstract
AB Accurate delineation of the intraprostatic gross tumor volume (GTV) is a prerequisite for treatment approaches in patients with primary prostate cancer (PCa). Prostate-specific membrane antigen PET (PSMA PET) may outperform MRI in GTV detection. However, visual GTV delineation underlies interobserver heterogeneity and is time consuming. The aim of this study was to develop a convolutional neural network (CNN) for automated segmentation of intraprostatic tumor (GTV-CNN) in PSMA PET. Methods: The CNN (3D U-Net) was trained on the 68Ga-PSMA PET images of 152 patients from 2 different institutions, and the training labels were generated manually using a validated technique. The CNN was tested on 2 independent internal (cohort 1: 68Ga-PSMA PET, n = 18 and cohort 2: 18F-PSMA PET, n = 19) and 1 external (cohort 3: 68Ga-PSMA PET, n = 20) test datasets. Accordance between manual contours and GTV-CNN was assessed with the Dice-Sørensen coefficient (DSC). Sensitivity and specificity were calculated for the 2 internal test datasets (cohort 1: n = 18, cohort 2: n = 11) using whole-mount histology. Results: The median DSCs for cohorts 1–3 were 0.84 (range: 0.32–0.95), 0.81 (range: 0.28–0.93), and 0.83 (range: 0.32–0.93), respectively. Sensitivities and specificities for the GTV-CNN were comparable with manual expert contours: 0.98 and 0.76 (cohort 1) and 1 and 0.57 (cohort 2), respectively. Computation time was around 6 s for a standard dataset. Conclusion: The application of a CNN for automated contouring of intraprostatic GTV in 68Ga-PSMA and 18F-PSMA PET images resulted in a high concordance with expert contours and in high sensitivities and specificities in comparison with histology as a reference. This robust, accurate and fast technique may be implemented for treatment concepts in primary prostate cancer. The trained model and the study’s source code are available in an open source repository.