Development and validation of a deep learning algorithm for auto-delineation of clinical target volume and organs at risk in cervical cancer radiotherapy

Abstract

The delineation of the clinical target volume (CTV) is a crucial, laborious and subjective step in cervical cancer radiotherapy. The aim of this study was to propose and evaluate a novel end-to-end convolutional neural network (CNN) for fully automatic and accurate CTV in cervical cancer. Methods: A total of 237 computed tomography (CT) scans of patients with locally advanced cervical cancer were collected and evaluated. A novel 2.5D CNN network, called DpnUNet, was developed for CTV delineation and further applied for CTV and organ-at-risk (OAR) delineation simultaneously. Comprehensive comparisons and experiments were performed. The mean Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (95HD) and subjective evaluation were used to assess the performance of this method. Results: The mean DSC and 95HD values were 0.86 and 5.34 mm for the delineated CTVs. The clinical experts' subjective assessments showed that 90% of the predicted contours were acceptable for clinical usage. The mean DSC and 95HD values were 0.91 and 4.05 mm for bladder, 0.85 and 2.16 mm for bone marrow, 0.90 and 1.27 mm for left femoral head, 0.90 and 1.51 mm for right femoral head, 0.82 and 4.29 mm for rectum, 0.85 and 4.35 mm for bowel bag, 0.82 and 4.96 mm for spinal cord respectively. The average delineation time for one patient's CT images was within 15 seconds. Conclusion:The experimental results demonstrate that the CTV and OARs delineated for cervical cancer by DpnUNet was in close agreement with the ground truth. DpnUNet could significantly reduce the radiation oncologists' contouring time.

Evaluation of deep learning-based autosegmentation in breast cancer radiotherapy

“…Our group also previously demonstrated the potential of deep learning-based autosegmentation of target volumes and OARs in breast cancer radiotherapy [ 7 , 8 ]. A training set for a proposed deep learning-based autocontouring system (ACS) is generally generated by a single expert or a small group of experts [ 9 ]. Therefore, generalization is often discussed as an issue of external validity.…”

An Adversarial Deep-Learning-Based Model for Cervical Cancer CTV Segmentation With Multicenter Blinded Randomized Controlled Validation

“…The proposed model is based on DpnUNet ( 13 ), which originated from the architecture of U-Net ( 19 ), but replaces all the encoder and decoder components with DPN components. Considering that the original DpnUNet is still underperforming compared with manual delineation in clinical practice, an extra convolutional layer is added at the end of DpnUNet, in which the output channels are one and the kernel size is 1 × 1.…”

“…A recent study has first applied a deep-learning-based method called DpnUNet to CTV segmentation in cervical cancer. The authors’ previous experimental results demonstrated that 88.65% of the contours generated by DpnUNet were acceptable for clinical usage ( 13 ). The mean dice similarity coefficient (DSC) and the 95 th Hausdorff distance (95HD) were 0.86 and 5.34 for the delineated CTVs.…”

First experience of autonomous, un-supervised treatment planning integrated in adaptive MR-guided radiotherapy and delivered to a patient with prostate cancer

“…Recently, the potentials of artificial intelligence (AI) in healthcare to automatize, standardize and speed-up processes has been discussed extensively [2][3][4][5][6]. In RT, the automation of single workflow steps has recently been investigated by several groups, especially with respect to automatic contouring and plan optimization [7][8][9][10][11][12][13][14][15][16]. For automatic segmentation of organs at risk (OAR), deep learning (DL)-based solutions showed promising results, with first reports on introduction into clinical routine and commercialization [7][8][9][10][11][12].…”

“…In RT, the automation of single workflow steps has recently been investigated by several groups, especially with respect to automatic contouring and plan optimization [7][8][9][10][11][12][13][14][15][16]. For automatic segmentation of organs at risk (OAR), deep learning (DL)-based solutions showed promising results, with first reports on introduction into clinical routine and commercialization [7][8][9][10][11][12]. Solutions for automatic treatment plan optimization were developed and are today partly available for routine planning in commercial treatment planning systems (TPS) [13][14][15][16].…”