Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region

Kieselmann, Jennifer P.; Kamerling, C P; Burgos, Ninon; Menten, Martin J.; Fuller, Clifton D.; Nill, Simeon; Cardoso, M. Jorge; Oelfke, Uwe

doi:10.1088/1361-6560/aacb65

Cited by 36 publications

(63 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The study was limited to three soft tissue organs. Kieselmann et al 26 took an atlas approach to segment the parotids, spinal cord, and mandible on T1-weighted MR. Other than the limited organs in these studies, they were based on high-field diagnostic multiparametric MR images while MR-guided radiation therapy more often relies on lower magnetic fields and is more limited in variety of sequences. To our best knowledge, there has not been an automated segmentation study on the low-field H&N MR images from MR-guided radiotherapy systems.…”

Section: Introductionmentioning

confidence: 99%

Shape constrained fully convolutional DenseNet with adversarial training for multiorgan segmentation on head and neck CT and low‐field MR images

et al. 2019

View full text Add to dashboard Cite

Purpose Image‐guided radiotherapy provides images not only for patient positioning but also for online adaptive radiotherapy. Accurate delineation of organs‐at‐risk (OARs) on Head and Neck (H&N) CT and MR images is valuable to both initial treatment planning and adaptive planning, but manual contouring is laborious and inconsistent. A novel method based on the generative adversarial network (GAN) with shape constraint (SC‐GAN) is developed for fully automated H&N OARs segmentation on CT and low‐field MRI. Methods and material A deep supervised fully convolutional DenseNet is employed as the segmentation network for voxel‐wise prediction. A convolutional neural network (CNN)‐based discriminator network is then utilized to correct predicted errors and image‐level inconsistency between the prediction and ground truth. An additional shape representation loss between the prediction and ground truth in the latent shape space is integrated into the segmentation and adversarial loss functions to reduce false positivity and constrain the predicted shapes. The proposed segmentation method was first benchmarked on a public H&N CT database including 32 patients, and then on 25 0.35T MR images obtained from an MR‐guided radiotherapy system. The OARs include brainstem, optical chiasm, larynx (MR only), mandible, pharynx (MR only), parotid glands (both left and right), optical nerves (both left and right), and submandibular glands (both left and right, CT only). The performance of the proposed SC‐GAN was compared with GAN alone and GAN with the shape constraint (SC) but without the DenseNet (SC‐GAN‐ResNet) to quantify the contributions of shape constraint and DenseNet in the deep neural network segmentation. Results The proposed SC‐GAN slightly but consistently improve the segmentation accuracy on the benchmark H&N CT images compared with our previous deep segmentation network, which outperformed other published methods on the same or similar CT H&N dataset. On the low‐field MR dataset, the following average Dice's indices were obtained using improved SC‐GAN: 0.916 (brainstem), 0.589 (optical chiasm), 0.816 (mandible), 0.703 (optical nerves), 0.799 (larynx), 0.706 (pharynx), and 0.845 (parotid glands). The average surface distances ranged from 0.68 mm (brainstem) to 1.70 mm (larynx). The 95% surface distance ranged from 1.48 mm (left optical nerve) to 3.92 mm (larynx). Compared with CT, using 95% surface distance evaluation, the automated segmentation accuracy is higher on MR for the brainstem, optical chiasm, optical nerves and parotids, and lower for the mandible. The SC‐GAN performance is superior to SC‐GAN‐ResNet, which is more accurate than GAN alone on both the CT and MR datasets. The segmentation time for one patient is 14 seconds using a single GPU. Conclusion The performance of our previous shape constrained fully CNNs for H&N segmentation is further improved by incorporating GAN and DenseNet. With the novel segmentation method, we showed that the low‐field MR images acquired on a MR‐guided radiation radiotherapy syste...

show abstract

Section: Introductionmentioning

confidence: 99%

Shape constrained fully convolutional DenseNet with adversarial training for multiorgan segmentation on head and neck CT and low‐field MR images

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Many studies have shown that although it is important to quantify the degree of variation or uncertainty of the contouring, it is more important to determine the dose difference and clinical impact [10,11,14,16,17,22,23]. In earlier work, van Rooij et al [17] studied the accuracy of automatic delineation of organs at risk in the head and neck region based on deep learning techniques while using geometric indices and dosimetric indices, and they analyzed the correlation between the geometric index SDC (mean value of the DSC) and dose difference.…”

Section: Discussionmentioning

confidence: 99%

“…In addition, there is considerable variability between them, and different auto-segmentation studies use different geometric indices to evaluate the contouring results; further, different indices have different properties [12][13][14]. Under the assumption of a reference contour, the method for clinically assessing the accuracy of radiotherapy (RT) contours is to determine and predict the deviation of its dosimetric indices based on the dose distribution of the radiation treatment plan [10,[15][16][17]. The relationships between geometric indices and dosimetric indices are yet to be further studied.…”

Section: Introductionmentioning

confidence: 99%

“…The relationships between geometric indices and dosimetric indices are yet to be further studied. At present, the research on the evaluation of geometric indices and dose parameters of auto-segmentation results is carried out under the condition of unknown contouring error [10,11,16]. The purpose of our study is to explore the evaluation accuracy of the geometric indices for evaluating the contours under the known errors.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Clinically oriented contour evaluation using geometric and dosimetric indices based on simple geometric transformations

Xian

Xiao

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract Background: In radiotherapy, geometric indices are often used to evaluate the accuracy of contouring; however, the ability of geometric indices to identify the error of contouring results is limited, and they do not consider any clinical background. Based on the reference contouring, we systematically introduced the known geometric errors to study the relationship between geometric indices and dosimetric indices and evaluated the clinical feasibility of assessing the accuracy of contouring based on geometric indices alone.Materials and Methods: A C-shaped target, organ at risk (Core), and intensity-modulated radiotherapy (IMRT) plan outlined in the American Association of Physicists in Medicine (AAPM) TG-119 report (The report of Task Group 119 of the AAPM) were used as references. Translation, scaling, rotation (except for the Core), and sine function transformation were performed to simulate the test contours. The corresponding dosimetric indices were obtained from the original dose distribution of the radiotherapy plan, and correlations (R²) between geometric and dosimetric indices were quantified through linear regression. The Wilcoxon signed rank test was used to compare the differences between the geometric indices of three different directions of translation transformations.Results: The correlations between the geometric and dosimetric indices were inconsistent for the contouring of the target and Core after the geometric transformation. Except for the sine function transformation (R²: 0.04–0.023, P > 0.05), the other geometric transformations of the planning target volume (PTV) had correlations with the dosimetric indices D98% and Dmean (R²: 0.689–0.988), 80% of which were strongly correlated. The correlation results for the other geometric transformations in the Core were similar to those in the PTV except for the posterior direction transformation. The results of Wilcoxon signed rank test showed that only the P-values of volumetric geometric indices of PTV were less than 0.05.Conclusions: The dosimetric indices are heavily influenced by the contour differences, thus highlighting their importance in the evaluation process. Clinically, an assessment of the contour accuracy of the region of interest is not feasible based on geometric indices alone, and should be combined with dosimetric indices.

show abstract

“…Obrazowanie MRI, przed zastosowaniem hybryd MR-RT, było wykorzystywane głównie jako pomocnicze do wyznaczenia objętości napromienianej na skanach CT po wcześniejszej fuzji obrazów. Obecnie systemy pracujące w oparciu o MR-only RT umożliwiają skanowanie pacjenta bezpośrednio przed lub w trakcie leczenia i dostosowanie planu zgodnie ze zaktualizowanymi informacjami na temat anatomii pacjenta dla tej samej frakcji leczenia [6]. Do wyznaczenia nowych konturów targetu i narządów krytycznych na skanach MR możemy wybrać poniższe możliwości:…”

Section: Wyznaczenie Konturówunclassified

Najnowsze osiągnięcia w radioterapii adaptacyjnej MRgRT w świetle doniesień zaprezentowanych podczas konferencji ESTRO 37

Sobocka-Kurdyk

2019

LOS

View full text Add to dashboard Cite

show abstract

Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region

Cited by 36 publications

References 50 publications

Shape constrained fully convolutional DenseNet with adversarial training for multiorgan segmentation on head and neck CT and low‐field MR images

Shape constrained fully convolutional DenseNet with adversarial training for multiorgan segmentation on head and neck CT and low‐field MR images

Clinically oriented contour evaluation using geometric and dosimetric indices based on simple geometric transformations

Najnowsze osiągnięcia w radioterapii adaptacyjnej MRgRT w świetle doniesień zaprezentowanych podczas konferencji ESTRO 37

Contact Info

Product

Resources

About