Search

Accurate clinical target volume (CTV) delineation is essential to ensure proper tumor coverage in radiation therapy. This is a particularly difficult task for head-and-neck cancer patients where detailed knowledge of the pathways of microscopic tumor spread is necessary. This paper proposes a solution to auto-segment these volumes in oropharyngeal cancer patients using a two-channel 3D U-Net architecture. The first channel feeds the network with the patient’s CT image providing anatomical context, whereas the second channel provides the network with tumor location and morphological information. Radiation therapy simulation computer tomography scans and their corresponding manually delineated CTV and gross tumor volume (GTV) delineations from 285 oropharyngeal patients previously treated at MD Anderson Cancer Center were used in this study. CTV and GTV delineations underwent rigorous group peer-review prior to the start of treatment delivery. The convolutional network’s parameters were fine-tuned using a training set of 210 patients using 3-fold cross-validation. During hyper-parameter selection, we use a score based on the overlap (dice similarity coefficient (DSC)) and missed volumes (false negative dice (FND)) to minimize any possible under-treatment. Three auto-delineated models were created to estimate tight, moderate, and wide CTV margin delineations. Predictions on our test set (75 patients) resulted in auto-delineations with high overlap and close surface distance agreement (DSC  >  0.75 on 96% of cases for tight and moderate auto-delineation models and 97% of cases having mean surface distance  ⩽  5.0 mm) to the ground-truth. We found that applying a 5 mm uniform margin expansion to the auto-delineated CTVs would cover at least 90% of the physician CTV volumes for a large majority of patients; however, determination of appropriate margin expansions for auto-delineated CTVs merits further investigation. [ABSTRACT FROM AUTHOR]

We developed a novel technique to study the impact of geometric distortion of magnetic resonance imaging (MRI) on intensity-modulated radiation therapy treatment planning. The measured 3D datasets of residual geometric distortion (a 1.5 T MRI component of an MRI linear accelerator system) was fitted with a second-order polynomial model to map the spatial dependence of geometric distortions. Then the geometric distortion model was applied to computed tomography (CT) image and structure data to simulate the distortion of MRI data and structures. Fourteen CT-based treatment plans were selected from patients treated for gastrointestinal, genitourinary, thoracic, head and neck, or spinal tumors. Plans based on the distorted CT and structure data were generated (as the distorted plans). Dose deviations of the distorted plans were calculated and compared with the original plans to study the dosimetric impact of MRI distortion. The MRI geometric distortion led to notable dose deviations in five of the 14 patients, causing loss of target coverage of up to 3.68% and dose deviations to organs at risk in three patients, increasing the mean dose to the chest wall by up to 6.19 Gy in a gastrointestinal patient, and increases the maximum dose to the lung by 5.17 Gy in a thoracic patient. [ABSTRACT FROM AUTHOR]

Radiomics studies require large patient cohorts, which often include patients imaged using different imaging protocols. We aimed to determine the impact of variability in imaging protocol parameters and interscanner variability using a phantom that produced feature values similar to those of patients. Positron emission tomography (PET) scans of a Hoffman brain phantom were acquired on GE Discovery 710, Siemens mCT, and Philips Vereos scanners. A standard-protocol scan was acquired on each machine, and then each parameter that could be changed was altered individually. The phantom was contoured with 10 regions of interest (ROIs). Values for 45 features with 2 different preprocessing techniques were extracted for each image. To determine the impact of each parameter on the reliability of each radiomics feature, the intraclass correlation coefficient (ICC) was calculated with the ROIs as the subjects and the parameter values as the raters. For interscanner comparisons, we compared the standard deviation of each radiomics feature value from the standard-protocol images to the standard deviation of the same radiomics feature from PET scans of 224 patients with non-small cell lung cancer. When the pixel size was resampled prior to feature extraction, all features had good reliability (ICC > 0.75) for the field of view and matrix size. The time per bed position had excellent reliability (ICC > 0.9) on all features. When the filter cutoff was restricted to values below 6 mm, all features had good reliability. Similarly, when subsets and iterations were restricted to reasonable values used in clinics, almost all features had good reliability. The average ratio of the standard deviation of features on the phantom scans to that of the NSCLC patient scans was 0.73 using fixed-bin-width preprocessing and 0.92 using 64-level preprocessing. Most radiomics feature values had at least good reliability when imaging protocol parameters were within clinically used ranges. However, interscanner variability was about equal to interpatient variability; therefore, caution must be used when combining patients scanned on equipment from different vendors in radiomics data sets.

Radiomics studies require many patients in order to power them, thus patients are often combined from different institutions and using different imaging protocols. Various studies have shown that imaging protocols affect radiomics feature values. We examined whether using data from cohorts with controlled imaging protocols improved patient outcome models. We retrospectively reviewed 726 CT and 686 PET images from head and neck cancer patients, who were divided into training or independent testing cohorts. For each patient, radiomics features with different preprocessing were calculated and two clinical variables-HPV status and tumor volume-were also included. A Cox proportional hazards model was built on the training data by using bootstrapped Lasso regression to predict overall survival. The effect of controlled imaging protocols on model performance was evaluated by subsetting the original training and independent testing cohorts to include only patients whose images were obtained using the same imaging protocol and vendor. Tumor volume, HPV status, and two radiomics covariates were selected for the CT model, resulting in an AUC of 0.72. However, volume alone produced a higher AUC, whereas adding radiomics features reduced the AUC. HPV status and one radiomics feature were selected as covariates for the PET model, resulting in an AUC of 0.59, but neither covariate was significantly associated with survival. Limiting the training and independent testing to patients with the same imaging protocol reduced the AUC for CT patients to 0.55, and no covariates were selected for PET patients. Radiomics features were not consistently associated with survival in CT or PET images of head and neck patients, even within patients with the same imaging protocol.

Strong magnetic fields affect radiation dose deposition in MRI-guided radiation therapy systems, particularly at interfaces between tissues of differing densities such as those in the thorax. In this study, we evaluated the impact of a 1.5 T magnetic field on radiation-induced lung damage in C57L/J mice. We irradiated 140 mice to the whole thorax with parallel-opposed Co-60 beams to doses of 0, 9.0, 10.0, 10.5, 11.0, 12.0, or 13.0 Gy (20 mice per dose group). Ten mice per dose group were irradiated while a 1.5 T magnetic field was applied transverse to the radiation beam and ten mice were irradiated with the magnetic field set to 0 T. We compared survival and noninvasive assays of radiation-induced lung damage, namely respiratory rate and metrics derived from thoracic cone-beam CTs, between the two sets of mice. We report two main results. First, the presence of a transverse 1.5 T field during irradiation had no impact on survival of C57L/J mice. Second, there was a small but statistically significant effect on noninvasive assays of radiation-induced lung damage. These results provide critical safety data for the clinical introduction of MRI-guided radiation therapy systems.

PURPOSE:To evaluate the uncertainty of radiomics features from contrast-enhanced breath-hold helical CT scans of non-small cell lung cancer for both manual and semi-automatic segmentation due to intra-observer, inter-observer, and inter-software reliability. METHODS:Three radiation oncologists manually delineated lung tumors twice from 10 CT scans using two software tools (3D-Slicer and MIM Maestro). Additionally, three observers without formal clinical training were instructed to use two semi-automatic segmentation tools, Lesion Sizing Toolkit (LSTK) and GrowCut, to delineate the same tumor volumes. The accuracy of the semi-automatic contours was assessed by comparison with physician manual contours using Dice similarity coefficients and Hausdorff distances. Eighty-three radiomics features were calculated for each delineated tumor contour. Informative features were identified based on their dynamic range and correlation to other features. Feature reliability was then evaluated using intra-class correlation coefficients (ICC). Feature range was used to evaluate the uncertainty of the segmentation methods. RESULTS:From the initial set of 83 features, 40 radiomics features were found to be informative, and these 40 features were used in the subsequent analyses. For both intra-observer and inter-observer reliability, LSTK had higher reliability than GrowCut and the two manual segmentation tools. All observers achieved consistently high ICC values when using LSTK, but the ICC value varied greatly for each observer when using GrowCut and the manual segmentation tools. For inter-software reliability, features were not reproducible across the software tools for either manual or semi-automatic segmentation methods. Additionally, no feature category was found to be more reproducible than another feature category. Feature ranges of LSTK contours were smaller than those of manual contours for all features. CONCLUSION:Radiomics features extracted from LSTK contours were highly reliable across and among observers. With semi-automatic segmentation tools, observers without formal clinical training were comparable to physicians in evaluating tumor segmentation.

Visceral adiposity is a risk factor for many chronic diseases. Existing methods to quantify visceral adipose tissue volume using computed tomographic (CT) images often use a single slice, are manual, and are time consuming, making them impractical for large population studies. We developed and validated a method to accurately, rapidly, and robustly measure visceral adipose tissue volume using CT images.In-house software, Medical Executable for the Efficient and Robust Quantification of Adipose Tissue (MEERQAT), was developed to calculate visceral adipose tissue volume using a series of CT images within a manually identified region of interest. To distinguish visceral and subcutaneous adipose tissue, ellipses are drawn through the rectus abdominis and transverse abdominis using manual and automatic processes. Visceral and subcutaneous adipose tissue volumes are calculated by counting the numbers of voxels corresponding to adipose tissue in the region of interest. MEERQAT's ellipse interpolation method was validated by comparing visceral adipose volume from 10 patients' CT scans with corresponding results from manually delineated scans. Accuracy of visceral adipose quantification was tested using a phantom consisting of animal fat and tissues. Robustness of the method was tested by determining intra-observer and inter-observer coefficients of variation (CV).The mean difference in visceral adipose tissue volume between manual and elliptical delineation methods was -0.54 ± 4.81%. In the phantom, our measurement differed from the known adipose volume by ≤ 7.5% for all scanning parameters. Mean inter-observer CV for visceral adipose tissue volume was 0.085, and mean intra-observer CV for visceral adipose tissue volume was 0.059.We have developed and validated a robust method of accurately and quickly determining visceral adipose tissue volume in any defined region of interest using CT imaging.

Endoscopic examinations are frequently-used procedures for patients with head and neck cancer undergoing radiotherapy, but radiation treatment plans are created on computed tomography (CT) scans. Image registration between endoscopic video and CT could be used to improve treatment planning and analysis of radiation-related normal tissue toxicity. The purpose of this study was to explore the feasibility of endoscopy-CT image registration without prospective physical tracking of the endoscope during the examination.A novel registration technique called Location Search was developed. This technique uses physical constraints on the endoscope's view direction to search for the virtual endoscope coordinates that maximize the similarity between the endoscopic video frame and the virtual endoscopic image. Its performance was tested on phantom and patient images and compared to an established registration technique, Frame-To-Frame Tracking.In phantoms, Location Search had average registration errors of 0.55 ± 0.60 cm for point measurements and 0.29 ± 0.15 cm for object surface measurements. Frame-To-Frame Tracking achieved similar results on some frames, but it failed on others due to the virtual endoscope becoming lost. This weakness was more pronounced in patients, where Frame-To-Frame tracking could not make it through the nasal cavity. On successful patient video frames, Location Search was able to find endoscope positions with an average distance of 0.98 ± 0.53 cm away from the ground truth positions. However, it failed on many frames due to false similarity matches caused by anatomical structural differences between the endoscopic video and the virtual endoscopic images.Endoscopy-CT image registration without prospective physical tracking of the endoscope is possible, but more development is required to achieve an accuracy suitable for clinical translation.

When imaging studies (e.g. CT) are used to quantify morphological changes in an anatomical structure, it is necessary to understand the extent and source of motion which can give imaging artifacts (e.g. blurring or local distortion). The objective of this study was to assess the magnitude of esophageal motion due to cardiac motion. We used retrospective electrocardiogram-gated contrast-enhanced computed tomography angiography images for this study. The anatomic region from the carina to the bottom of the heart was taken at deep-inspiration breath hold with the patients' arms raised above their shoulders, in a position similar to that used for radiation therapy. The esophagus was delineated on the diastolic phase of cardiac motion, and deformable registration was used to sequentially deform the images in nearest-neighbor phases among the 10 cardiac phases, starting from the diastolic phase. Using the 10 deformation fields generated from the deformable registration, the magnitude of the extreme displacements was then calculated for each voxel, and the mean and maximum displacement was calculated for each computed tomography slice for each patient. The average maximum esophageal displacement due to cardiac motion for all patients was 5.8 mm (standard deviation: 1.6 mm, maximum: 10.0 mm) in the transverse direction. For 21 of 26 patients, the largest esophageal motion was found in the inferior region of the heart; for the other patients, esophageal motion was approximately independent of superior-inferior position. The esophagus motion was larger at cardiac phases where the electrocardiogram R-wave occurs. In conclusion, the magnitude of esophageal motion near the heart due to cardiac motion is similar to that due to other sources of motion, including respiratory motion and intra-fraction motion. A larger cardiac motion will result into larger esophagus motion in a cardiac cycle.

Purpose We assessed automated contouring of normal structures for patients with head-and-neck cancer (HNC) using a multiatlas deformable-image-registration algorithm to better provide a fully automated radiation treatment planning solution for low- and middle-income countries, provide quantitative analysis, and determine acceptability worldwide. Methods Autocontours of eight normal structures (brain, brainstem, cochleae, eyes, lungs, mandible, parotid glands, and spinal cord) from 128 patients with HNC were retrospectively scored by a dedicated HNC radiation oncologist. Contours from a 10-patient subset were evaluated by five additional radiation oncologists from international partner institutions, and interphysician variability was assessed. Quantitative agreement of autocontours with independently physician-drawn structures was assessed using the Dice similarity coefficient and mean surface and Hausdorff distances. Automated contouring was then implemented clinically and has been used for 166 patients, and contours were quantitatively compared with the physician-edited autocontours using the same metrics. Results Retrospectively, 87% of normal structure contours were rated as acceptable for use in dose-volume-histogram-based planning without edit. Upon clinical implementation, 50% of contours were not edited for use in treatment planning. The mean (± standard deviation) Dice similarity coefficient of autocontours compared with physician-edited autocontours for parotid glands (0.92 ± 0.10), brainstem (0.95 ± 0.09), and spinal cord (0.92 ± 0.12) indicate that only minor edits were performed. The average mean surface and Hausdorff distances for all structures were less than 0.15 mm and 1.8 mm, respectively. Conclusion Automated contouring of normal structures generates reliable contours that require only minimal editing, as judged by retrospective ratings from multiple international centers and clinical integration. Autocontours are acceptable for treatment planning with no or, at most, minor edits, suggesting that automated contouring is feasible for clinical use and in the ongoing development of automated radiation treatment planning algorithms.