Development of in-house fully residual deep convolutional neural network-based segmentation software for the male pelvic CT

Abstract

Background This study aimed to (1) develop a fully residual deep convolutional neural network (CNN)-based segmentation software for computed tomography image segmentation of the male pelvic region and (2) demonstrate its efficiency in the male pelvic region. Methods A total of 470 prostate cancer patients who had undergone intensity-modulated radiotherapy or volumetric-modulated arc therapy were enrolled. Our model was based on FusionNet, a fully residual deep CNN developed to semantically segment biological images. To develop the CNN-based segmentation software, 450 patients were randomly selected and separated into the training, validation and testing groups (270, 90, and 90 patients, respectively). In Experiment 1, to determine the optimal model, we first assessed the segmentation accuracy according to the size of the training dataset (90, 180, and 270 patients). In Experiment 2, the effect of varying the number of training labels on segmentation accuracy was evaluated. After determining the optimal model, in Experiment 3, the developed software was used on the remaining 20 datasets to assess the segmentation accuracy. The volumetric dice similarity coefficient (DSC) and the 95th-percentile Hausdorff distance (95%HD) were calculated to evaluate the segmentation accuracy for each organ in Experiment 3. Results In Experiment 1, the median DSC for the prostate were 0.61 for dataset 1 (90 patients), 0.86 for dataset 2 (180 patients), and 0.86 for dataset 3 (270 patients), respectively. The median DSCs for all the organs increased significantly when the number of training cases increased from 90 to 180 but did not improve upon further increase from 180 to 270. The number of labels applied during training had a little effect on the DSCs in Experiment 2. The optimal model was built by 270 patients and four organs. In Experiment 3, the median of the DSC and the 95%HD values were 0.82 and 3.23 mm for prostate; 0.71 and 3.82 mm for seminal vesicles; 0.89 and 2.65 mm for the rectum; 0.95 and 4.18 mm for the bladder, respectively. Conclusions We have developed a CNN-based segmentation software for the male pelvic region and demonstrated that the CNN-based segmentation software is efficient for the male pelvic region.

Evaluation of generalization ability for deep learning‐based auto‐segmentation accuracy in limited field of view CBCT of male pelvic region

“…Hirashima et al stated that segmentation using CNN is not optimal for small ROIs. 19 Nevertheless, the median DSC value in this study was comparable to those reported in previous studies. In general, intensity distributions for medical images are dependent on the characteristics of the imaging device.…”

“…0.73-0.91 for the prostate, 0.70-0.93 for the rectum, and 0.70 for the seminal vesicles using full FOV CBCT.These results were comparable to our findings. [13][14][15][16][17][19][20][21] Segmentation accuracy would be varied by intensity of image, FOV, and their combination. In this work, we discuss the effect of image intensity on the accuracy of segmentation.…”

“…[10][11][12] Researchers have conducted segmentation of soft tissue organs on CT and/or CBCT using neural encoding/decoding based on the CNN architecture, thereby exploiting supervised training. [13][14][15][16][17][18][19][20][21] The main requirement for this type of approach is anatomical correspondence between the input image (CT or CBCT), and the ground truth reference label. [13][14][15][16][17][18][19][20][21] Although CNNbased segmentation works well for full field of view (FOV) CBCT, [13][14][15][16][17]20,21 it is unknown whether CNNbased segmentation using limited FOV CBCT is possible.…”

Dose prediction for cervical cancer VMAT patients with a full-scale 3D-cGAN-based model and the comparison of different input data on the prediction results

“…This makes the use of deep learning methods feasible for predicting the dose distribution of the cervical cancer VMAT plan. On the other hand, learning the contours of the OARs from CT images is theoretically supported by related segmentation studies [41][42][43][44]. Although in our research, the 3D-cGAN model was trained and made predictions in terms of cervical cancer cases treated with VMAT, this prediction method is also applicable to other treatment sites and techniques.…”

Dose prediction for cervical cancer VMAT patients with a full-scale 3D-cGAN-based model and the comparison of different input data on the prediction results

“…This makes the use of deep learning methods feasible for predicting the dose distribution of the cervical cancer VMAT plan. On the other hand, learning the contours of the OARs from CT images is theoretically supported by related segmentation studies [ 43 – 46 ]. Although in our research, the 3D-cGAN model was trained and made predictions in terms of cervical cancer cases treated with VMAT, this prediction method is also applicable to other treatment sites and techniques.…”