Dataset Information

Evaluation of Deep Learning to Augment Image-Guided Radiotherapy for Head and Neck and Prostate Cancers.

ABSTRACT:

Importance

Personalized radiotherapy planning depends on high-quality delineation of target tumors and surrounding organs at risk (OARs). This process puts additional time burdens on oncologists and introduces variability among both experts and institutions.

Objective

To explore clinically acceptable autocontouring solutions that can be integrated into existing workflows and used in different domains of radiotherapy.

Design, setting, and participants

This quality improvement study used a multicenter imaging data set comprising 519 pelvic and 242 head and neck computed tomography (CT) scans from 8 distinct clinical sites and patients diagnosed either with prostate or head and neck cancer. The scans were acquired as part of treatment dose planning from patients who received intensity-modulated radiation therapy between October 2013 and February 2020. Fifteen different OARs were manually annotated by expert readers and radiation oncologists. The models were trained on a subset of the data set to automatically delineate OARs and evaluated on both internal and external data sets. Data analysis was conducted October 2019 to September 2020.

Main outcomes and measures

The autocontouring solution was evaluated on external data sets, and its accuracy was quantified with volumetric agreement and surface distance measures. Models were benchmarked against expert annotations in an interobserver variability (IOV) study. Clinical utility was evaluated by measuring time spent on manual corrections and annotations from scratch.

Results

A total of 519 participants' (519 [100%] men; 390 [75%] aged 62-75 years) pelvic CT images and 242 participants' (184 [76%] men; 194 [80%] aged 50-73 years) head and neck CT images were included. The models achieved levels of clinical accuracy within the bounds of expert IOV for 13 of 15 structures (eg, left femur, κ = 0.982; brainstem, κ = 0.806) and performed consistently well across both external and internal data sets (eg, mean [SD] Dice score for left femur, internal vs external data sets: 98.52% [0.50] vs 98.04% [1.02]; P = .04). The correction time of autogenerated contours on 10 head and neck and 10 prostate scans was measured as a mean of 4.98 (95% CI, 4.44-5.52) min/scan and 3.40 (95% CI, 1.60-5.20) min/scan, respectively, to ensure clinically accepted accuracy. Manual segmentation of the head and neck took a mean 86.75 (95% CI, 75.21-92.29) min/scan for an expert reader and 73.25 (95% CI, 68.68-77.82) min/scan for a radiation oncologist. The autogenerated contours represented a 93% reduction in time.

Conclusions and relevance

In this study, the models achieved levels of clinical accuracy within expert IOV while reducing manual contouring time and performing consistently well across previously unseen heterogeneous data sets. With the availability of open-source libraries and reliable performance, this creates significant opportunities for the transformation of radiation treatment planning.

SUBMITTER: Oktay O

PROVIDER: S-EPMC7705593 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:In adaptive radiation therapy of prostate cancer, fast and accurate registration between the planning image and treatment images of the patient is of essential importance. With the authors' recently developed deformable surface model, prostate boundaries in each treatment image can be rapidly segmented and their correspondences (or relative deformations) to the prostate boundaries in the planning image are also established automatically. However, the dense correspondences on the nonboundary regions, which are important especially for transforming the treatment plan designed in the planning image space to each treatment image space, are remained unresolved. This paper presents a novel approach to learn the statistical correlation between deformations of prostate boundary and nonboundary regions, for rapidly estimating deformations of the nonboundary regions when given the deformations of the prostate boundary at a new treatment image.The main contributions of the proposed method lie in the following aspects. First, the statistical deformation correlation will be learned from both current patient and other training patients, and further updated adaptively during the radiotherapy. Specifically, in the initial treatment stage when the number of treatment images collected from the current patient is small, the statistical deformation correlation is mainly learned from other training patients. As more treatment images are collected from the current patient, the patient-specific information will play a more important role in learning patient-specific statistical deformation correlation to effectively reflect prostate deformation of the current patient during the treatment. Eventually, only the patient-specific statistical deformation correlation is used to estimate dense correspondences when a sufficient number of treatment images have been acquired from the current patient. Second, the statistical deformation correlation will be learned by using a multiple linear regression (MLR) model, i.e., ridge regression (RR) model, which has the best prediction accuracy than other MLR models such as canonical correlation analysis (CCA) and principal component regression (PCR).To demonstrate the performance of the proposed method, we first evaluate its registration accuracy by comparing the deformation field predicted by our method with the deformation field estimated by the thin plate spline (TPS) based correspondence interpolation method on 306 serial prostate CT images of 24 patients. The average predictive error on the voxels around 5 mm of prostate boundary is 0.38 mm for our method of RR-based correlation model. Also, the corresponding maximum error is 2.89 mm. We then compare the speed for deformation interpolation by different methods. When considering the larger region of interest (ROI) with the size of 512 × 512 × 61, our method takes 24.41 seconds to interpolate the dense deformation field while TPS method needs 6.7 minutes; when considering a small ROI (surrounding prostate) with size of 112 × 110 × 93, our method takes 1.80 seconds, while TPS method needs 25 seconds.Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy, compared to the TPS-based correspondence (or deformation) interpolation approach.

Project description:BackgroundWe reported the concept of patient-specific deep learning (DL) for real-time markerless tumor segmentation in image-guided radiotherapy (IGRT). The method was aimed to control the attention of convolutional neural networks (CNNs) by artificial differences in co-occurrence probability (CoOCP) in training datasets, that is, focusing CNN attention on soft tissues while ignoring bones. However, the effectiveness of this attention-based data augmentation has not been confirmed by explainable techniques. Furthermore, compared to reasonable ground truths, the feasibility of tumor segmentation in clinical kilovolt (kV) X-ray fluoroscopic (XF) images has not been confirmed.PurposeThe first aim of this paper was to present evidence that the proposed method provides an explanation and control of DL behavior. The second purpose was to validate the real-time lung tumor segmentation in clinical kV XF images for IGRT.MethodsThis retrospective study included 10 patients with lung cancer. Patient-specific and XF angle-specific image pairs comprising digitally reconstructed radiographs (DRRs) and projected-clinical-target-volume (pCTV) images were calculated from four-dimensional computer tomographic data and treatment planning information. The training datasets were primarily augmented by random overlay (RO) and noise injection (NI): RO aims to differentiate positional CoOCP in soft tissues and bones, and NI aims to make a difference in the frequency of occurrence of local and global image features. The CNNs for each patient-and-angle were automatically optimized in the DL training stage to transform the training DRRs into pCTV images. In the inference stage, the trained CNNs transformed the test XF images into pCTV images, thus identifying target positions and shapes.ResultsThe visual analysis of DL attention heatmaps for a test image demonstrated that our method focused CNN attention on soft tissue and global image features rather than bones and local features. The processing time for each patient-and-angle-specific dataset in the training stage was ∼30 min, whereas that in the inference stage was 8 ms/frame. The estimated three-dimensional 95 percentile tracking error, Jaccard index, and Hausdorff distance for 10 patients were 1.3-3.9 mm, 0.85-0.94, and 0.6-4.9 mm, respectively.ConclusionsThe proposed attention-based data augmentation with both RO and NI made the CNN behavior more explainable and more controllable. The results obtained demonstrated the feasibility of real-time markerless lung tumor segmentation in kV XF images for IGRT.

Project description:BackgroundThe imaging dose of cone-beam computed tomography (CBCT) in image-guided radiotherapy (IGRT) poses adverse effects on patient health. To improve the quality of sparse-view low-dose CBCT images, a projection synthesis convolutional neural network (SynCNN) model is proposed.MethodsIncluded in this retrospective, single-center study were 223 patients diagnosed with brain tumours from Beijing Cancer Hospital. The proposed SynCNN model estimated two pairs of orthogonally direction-separable spatial kernels to synthesize the missing projection in between the input neighboring sparse-view projections via local convolution operations. The SynCNN model was trained on 150 real patients to learn patterns for inter-view projection synthesis. CBCT data from 30 real patients were used to validate the SynCNN, while data from a phantom and 43 real patients were used to test the SynCNN externally. Sparse-view projection datasets with 1/2, 1/4, and 1/8 of the original sampling rate were simulated, and the corresponding full-view projection datasets were restored using the SynCNN model. The tomographic images were then reconstructed with the Feldkamp-Davis-Kress algorithm. The root-mean-square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) metrics were measured in both the projection and image domains. Five experts were invited to grade the image quality blindly for 40 randomly selected evaluation groups with a four-level rubric, where a score greater than or equal to 2 was considered acceptable image quality. The running time of the SynCNN model was recorded. The SynCNN model was directly compared with the three other methods on 1/4 sparse-view reconstructions.ResultsThe phantom and patient studies showed that the missing projections were accurately synthesized. In the image domain, for the phantom study, compared with images reconstructed from sparse-view projections, images with SynCNN synthesis exhibited significantly improved qualities with decreased values in RMSE and increased values in PSNR and SSIM. For the patient study, between the results with and without the SynCNN synthesis, the averaged RMSE decreased by 3.4×10-4, 10.3×10-4, and 21.7×10-4, the averaged PSNR increased by 3.4, 6.6, and 9.4 dB, and the averaged SSIM increased by 5.2×10-2, 18.9×10-2 and 33.9×10-2, for the 1/2, 1/4, and 1/8 sparse-view reconstructions, respectively. In expert subjective evaluation, both the median scores and acceptance rates of the images with SynCNN synthesis were higher than those reconstructed from sparse-view projections. It took the model less than 0.01 s to synthesize an inter-view projection. Compared with the three other methods, the SynCNN model obtained the best scores in terms of the three metrics in both domains.ConclusionsThe proposed SynCNN model effectively improves the quality of sparse-view CBCT images at a low time cost. With the SynCNN model, the CBCT imaging dose in IGRT could be reduced potentially.