Simple low-cost approaches to semantic segmentation in radiation therapy planning for prostate cancer using deep learning with non-contrast planning CT images

Abstract

Deep learning has shown great efficacy for semantic segmentation. However, there are difficulties in the collection, labeling and management of medical imaging data, because of ethical complications and the limited number of imaging studies available at a single facility. This study aimed to find a simple and low-cost method to increase the accuracy of deep learning semantic segmentation for radiation therapy of prostate cancer. Methods: In total, 556 cases with non-contrast CT images for prostate cancer radiation therapy were examined using a two-dimensional U-Net. Initially, all slices were used for the input data. Then, we removed slices of the cranial portions, which were beyond the margins of the bladder and rectum. Finally, the ground truth labels for the bladder and rectum were added as channels to the input for the prostate training dataset. Results: The highest mean dice similarity coefficients (DSCs) for each organ in the test dataset of 56 cases were 0.85 ± 0.05, 0.94 ± 0.04 and 0.85 ± 0.07 for the prostate, bladder and rectum, respectively. Removal of the cranial slices from the original images significantly increased the DSC of the rectum from 0.83 ± 0.09 to 0.85 ± 0.07 (p < 0.05). Adding bladder and rectum information to prostate training without removing the slices significantly increased the DSC of the prostate from 0.79 ± 0.05 to 0.85 ± 0.05 (p < 0.05). Conclusions: These cost-free approaches may be useful for new applications, which may include updated models and datasets. They may be applicable to other organs at risk (OARs) and clinical targets such as elective nodal irradiation.

Analytical performance of aPROMISE: automated anatomic contextualization, detection, and quantification of [18F]DCFPyL (PSMA) imaging for standardized reporting

“…However, the low soft tissue contrast and resolution in low-dose noncontrast-enhanced CT images of PET/CT provide a more difficult challenge in obtaining a clear automated volumetric segmentation of small organs. The performance of our aPROMISE algorithm in prostate segmentation in low-dose CT, without contrast, was similar to that of Nemoto T et al who also demonstrated a mean Dice score of 0.79 for prostate [22]. The Dice score of the bones and the visceral organ were observed to be 0.88 or above, indicating a much better performance of the algorithm in larger organs.…”

A Survey on Deep Learning for Precision Oncology

“…Recent studies [ 112 , 114 , 122 , 123 ] have investigated the use of deep learning for the task of predicting dose distribution of the cancer treatment in the prostate gland. Kandalan et al [ 112 ] employed a 3D U-Net for dose prediction for volumetric-modulated arc therapy (VMAT) using PTV and OARs contours in prostate cancer.…”

Deep Learning in Cancer Diagnosis and Prognosis Prediction: A Minireview on Challenges, Recent Trends, and Future Directions

“…Their model achieved a dice score of 0.88 and a precision of 76% on malignant and 75% on benign classes for the classification task using an Inception v3 architecture. The authors in [ 89 ] proposed a 2D U-Net model deploying CT images of 556 cases of prostate cancer. They achieved a dice score of 0.85, 0.94, and 0.85 for three organs, namely, prostate, bladder, and rectum, respectively.…”