Deep Learning Enables Prostate MRI Segmentation: A Large Cohort Evaluation With Inter-Rater Variability Analysis

Abstract

Whole-prostate gland (WPG) segmentation plays a significant role in prostate volume measurement, treatment, and biopsy planning. This study evaluated a previously developed automatic WPG segmentation, deep attentive neural network (DANN), on a large, continuous patient cohort to test its feasibility in a clinical setting. With IRB approval and HIPAA compliance, the study cohort included 3,698 3T MRI scans acquired between 2016 and 2020. In total, 335 MRI scans were used to train the model, and 3,210 and 100 were used to conduct the qualitative and quantitative evaluation of the model. In addition, the DANN-enabled prostate volume estimation was evaluated by using 50 MRI scans in comparison with manual prostate volume estimation. For qualitative evaluation, visual grading was used to evaluate the performance of WPG segmentation by two abdominal radiologists, and DANN demonstrated either acceptable or excellent performance in over 96% of the testing cohort on the WPG or each prostate sub-portion (apex, midgland, or base). Two radiologists reached a substantial agreement on WPG and midgland segmentation (κ = 0.75 and 0.63) and moderate agreement on apex and base segmentation (κ = 0.56 and 0.60). For quantitative evaluation, DANN demonstrated a dice similarity coefficient of 0.93 ± 0.02, significantly higher than other baseline methods, such as DeepLab v3+ and UNet (both p values < 0.05). For the volume measurement, 96% of the evaluation cohort achieved differences between the DANN-enabled and manual volume measurement within 95% limits of agreement. In conclusion, the study showed that the DANN achieved sufficient and consistent WPG segmentation on a large, continuous study cohort, demonstrating its great potential to serve as a tool to measure prostate volume.

Segmentation of the prostate, its zones, anterior fibromuscular stroma, and urethra on the MRIs and multimodality image fusion using U-Net model

“…Herein, for future works, we propose to use synthetic image generation methods such as a Cycle Generative Adversarial Network (CycleGAN) to produce MRI from the provided CT scans in prostate cancer patients. We should also point out the lack of a multi-reader, which would have been interesting as it was discussed by Becker et al (27) to evaluate inter-rater reliability (39,(65)(66)(67)(68) for four-zone prostate segmentation and assess the radiologist's level of expertise. Also, we propose expanding the scope of the study even further by: (I) not providing bounding boxes.…”

A Brief Review of Artificial Intelligence in Genitourinary Oncological Imaging

“…38,39 As in the case of lesion identification, segmentation algorithms also benefit from larger databases. In a recent study, 40 two readers visually assessed segmentation accuracy of a neural network that was based on 3210 MRI scans and rated more than 96% of the segmentations as either acceptable (failure of segmentation is less than 30% of the gland) or excellent (failure of segmentation is less than 10% of the gland) with only 4% of being unacceptable. As more readers are incorporated into studies, the actual utility of the algorithm is more accurately assessed.…”

Pretreatment information–aided automatic segmentation for online magnetic resonance imaging‐guided prostate radiotherapy

“…For prostate cancer, many studies investigated the segmentation of prostate and/or clinical target volumes (CTV) with DL and yielded promising results. [16][17][18][19][20][21][22][23][24][25] But the anatomical morphology varies greatly in the pelvic region from patient to patient, possibly due to variations in bladder/rectum filling and other unstable factors. Moreover, significant intensity inhomogeneities among MR images across patients can be observed even with the same scanner.…”

“…In recent years, DL became widely used for medical image segmentation due to its excellent performance. For prostate cancer, many studies investigated the segmentation of prostate and/or clinical target volumes (CTV) with DL and yielded promising results 16–25 . But the anatomical morphology varies greatly in the pelvic region from patient to patient, possibly due to variations in bladder/rectum filling and other unstable factors.…”