Comparative evaluation of a prototype deep learning algorithm for autosegmentation of normal tissues in head and neck radiotherapy.

PURPOSE: To introduce and validate a newly developed deep-learning (DL) auto-segmentation algorithm for head and neck (HN) organs at risk (OARs) and to compare its performance with a published commercial algorithm. METHODS: A total of 864 HN cancer cases were available to train and evaluate a prototype algorithm. The algorithm is based on a fully convolutional network with combined U-Net and V-net. A Dice loss plus Cross-Entropy Loss function with Adam optimizer was used in training. For 75 validation cases, OAR sets were generated with three DL-based models (A: the prototype model trained with gold data, B: a commercial software trained with the same data, and C: the same software trained with data from another institution). The auto-segmented structures were evaluated with Dice similarity coefficient (DSC), Hausdorff distance (HD), voxel-penalty metric (VPM) and DSC of area under dose-volume histograms. A subjective qualitative evaluation was performed on 20 random cases. RESULTS: Overall trend was for the prototype algorithm to be the closest to the gold data by all five metrics. The average DSC/VPM/HD for algorithms A, B, and C were 0.81/84.1/1.6 mm, 0.74/62.8/3.2 mm, and 0.66/46.8/3.3 mm, respectively. 93% of model A structures were evaluated to be clinically useful. CONCLUSION: The superior performance of the prototype was validated, even when trained with the same data. In addition to the challenges of perfecting the algorithms, the auto-segmentation results can differ when the same algorithm is trained at different institutions.

Technical note: Evaluation of a V-Net autosegmentation algorithm for pediatric CT scans: Performance, generalizability, and application to patient-specific CT dosimetry.

PURPOSE: This study developed and evaluated a fully convolutional network (FCN) for pediatric CT organ segmentation and investigated the generalizability of the FCN across image heterogeneities such as CT scanner model protocols and patient age. We also evaluated the autosegmentation models as part of a software tool for patient-specific CT dose estimation. METHODS: A collection of 359 pediatric CT datasets with expert organ contours were used for model development and evaluation. Autosegmentation models were trained for each organ using a modified FCN 3D V-Net. An independent test set of 60 patients was withheld for testing. To evaluate the impact of CT scanner model protocol and patient age heterogeneities, separate models were trained using a subset of scanner model protocols and pediatric age groups. Train and test sets were split to answer questions about the generalizability of pediatric FCN autosegmentation models to unseen age groups and scanner model protocols, as well as the merit of scanner model protocol or age-group-specific models. Finally, the organ contours resulting from the autosegmentation models were applied to patient-specific dose maps to evaluate the impact of segmentation errors on organ dose estimation. RESULTS: Results demonstrate that the autosegmentation models generalize to CT scanner acquisition and reconstruction methods which were not present in the training dataset. While models are not equally generalizable across age groups, age-group-specific models do not hold any advantage over combining heterogeneous age groups into a single training set. Dice similarity coefficient (DSC) and mean surface distance results are presented for 19 organ structures, for example, median DSC of 0.52 (duodenum), 0.74 (pancreas), 0.92 (stomach), and 0.96 (heart). The FCN models achieve a mean dose error within 5% of expert segmentations for all 19 organs except for the spinal canal, where the mean error was 6.31%. CONCLUSIONS: Overall, these results are promising for the adoption of FCN autosegmentation models for pediatric CT, including applications for patient-specific CT dose estimation.

Pediatric chest-abdomen-pelvis and abdomen-pelvis CT images with expert organ contours.

PURPOSE: Organ autosegmentation efforts to date have largely been focused on adult populations, due to limited availability of pediatric training data. Pediatric patients may present additional challenges for organ segmentation. This paper describes a dataset of 359 pediatric chest-abdomen-pelvis and abdomen-pelvis Computed Tomography (CT) images with expert contours of up to 29 anatomical organ structures to aid in the evaluation and development of autosegmentation algorithms for pediatric CT imaging. ACQUISITION AND VALIDATION METHODS: The dataset collection consists of axial CT images in Digital Imaging and Communications in Medicine (DICOM) format of 180 male and 179 female pediatric chest-abdomen-pelvis or abdomen-pelvis exams acquired from one of three CT scanners at Children's Wisconsin. The datasets represent random pediatric cases based upon routine clinical indications. Subjects ranged in age from 5 days to 16 years, with a mean age of 7 years. The CT acquisition, contrast, and reconstruction protocols varied across the scanner models and patients, with specifications available in the DICOM headers. Expert contours were manually labeled for up to 29 organ structures per subject. Not all contours are available for all subjects, due to limited field of view or unreliable contouring due to high noise. DATA FORMAT AND USAGE NOTES: The data are available on The Cancer Imaging Archive (TCIA_ (https://www.cancerimagingarchive.net/) under the collection Pediatric-CT-SEG. The axial CT image slices for each subject are available in DICOM format. The expert contours are stored in a single DICOM RTSTRUCT file for each subject. The contour names are listed in Table 2. POTENTIAL APPLICATIONS: This dataset will enable the evaluation and development of organ autosegmentation algorithms for pediatric populations, which exhibit variations in organ shape and size across age. Automated organ segmentation from CT images has numerous applications including radiation therapy, diagnostic tasks, surgical planning, and patient-specific organ dose estimation.

Deterministic linear Boltzmann transport equation solver for patient-specific CT dose estimation: Comparison against a Monte Carlo benchmark for realistic scanner configurations and patient models.

PURPOSE: Epidemiological evidence suggests an increased risk of cancer related to computed tomography (CT) scans, with children exposed to greater risk. The purpose of this work is to test the reliability of a linear Boltzmann transport equation (LBTE) solver for rapid and patient-specific CT dose estimation. This includes building a flexible LBTE framework for modeling modern clinical CT scanners and to validate the resulting dose maps across a range of realistic scanner configurations and patient models. METHODS: In this study, computational tools were developed for modeling CT scanners, including a bowtie filter, overrange collimation, and tube current modulation. The LBTE solver requires discretization in the spatial, angular, and spectral dimensions, which may affect the accuracy of scanner modeling. To investigate these effects, this study evaluated the LBTE dose accuracy for different discretization parameters, scanner configurations, and patient models (male, female, adults, pediatric). The method used to validate the LBTE dose maps was the Monte Carlo code Geant4, which provided ground truth dose maps. LBTE simulations were implemented on a GeForce GTX 1080 graphic unit, while Geant4 was implemented on a distributed cluster of CPUs. RESULTS: The agreement between Geant4 and the LBTE solver quantifies the accuracy of the LBTE, which was similar across the different protocols and phantoms. The results suggest that 18 views per rotation provides sufficient accuracy, as no significant improvement in the accuracy was observed by increasing the number of projection views. Considering this discretization, the LBTE solver average simulation time was approximately 30 s. However, in the LBTE solver the phantom model was implemented with a lower voxel resolution with respect to Geant4, as it is limited by the memory of the GPU. Despite this discretization, the results showed a good agreement between the LBTE and Geant4, with root mean square error of the dose in organs of approximately 3.5% for most of the studied configurations. CONCLUSIONS: The LBTE solver is proposed as an alternative to Monte Carlo for patient-specific organ dose estimation. This study demonstrated accurate organ dose estimates for the rapid LBTE solver when considering realistic aspects of CT scanners and a range of phantom models. Future plans will combine the LBTE framework with deep learning autosegmentation algorithms to provide near real-time patient-specific organ dose estimation.

Using federated data sources and Varian Learning Portal framework to train a neural network model for automatic organ segmentation.

PURPOSE: In this study we trained a deep neural network model for female pelvis organ segmentation using data from several sites without any personal data sharing. The goal was to assess its prediction power compared with the model trained in a centralized manner. METHODS: Varian Learning Portal (VLP) is a distributed machine learning (ML) infrastructure enabling privacy-preserving research across hospitals from different regions or countries, within the framework of a trusted consortium. Such a framework is relevant in the case when there is a high level of trust among the participating sites, but there are legal restrictions which do not allow the actual data sharing between them. We trained an organ segmentation model for the female pelvic region using the synchronous data distributed framework provided by the VLP. RESULTS: The prediction performance of the model trained using the federated framework offered by VLP was on the same level as the performance of the model trained in a centralized manner where all training data was pulled together in one centre. CONCLUSIONS: VLP infrastructure can be used for GPU-based training of a deep neural network for organ segmentation for the female pelvic region. This organ segmentation instance is particularly difficult due to the high variation in the organs' shape and size. Being able to train the model using data from several clinics can help, for instance, by exposing the model to a larger range of data variations. VLP framework enables such a distributed training approach without sharing protected health information.

Fast and robust adaptation of organs-at-risk delineations from planning scans to match daily anatomy in pre-treatment scans for online-adaptive radiotherapy of abdominal tumors.

PURPOSE: To validate a novel deformable image registration (DIR) method for online adaptation of planning organ-at-risk (OAR) delineations to match daily anatomy during hypo-fractionated RT of abdominal tumors. MATERIALS AND METHODS: For 20 liver cancer patients, planning OAR delineations were adapted to daily anatomy using the DIR on corresponding repeat CTs. The DIR's accuracy was evaluated for the entire cohort by comparing adapted and expert-drawn OAR delineations using geometric (Dice Similarity Coefficient (DSC), Modified Hausdorff Distance (MHD) and Mean Surface Error (MSE)) and dosimetric (Dmax and Dmean) measures. RESULTS: For all OARs, DIR achieved average DSC, MHD and MSE of 86%, 2.1 mm, and 1.7 mm, respectively, within 20 s for each repeat CT. Compared to the baseline (translations), the average improvements ranged from 2% (in heart) to 24% (in spinal cord) in DSC, and 25% (in heart) to 44% (in right kidney) in MHD and MSE. Furthermore, differences in dose statistics (Dmax, Dmean and D2%) using delineations from an expert and the proposed DIR were found to be statistically insignificant (p > 0.01). CONCLUSION: The validated DIR showed potential for online-adaptive radiotherapy of abdominal tumors as it achieved considerably high geometric and dosimetric correspondences with the expert-drawn OAR delineations, albeit in a fraction of time required by experts.

Feasibility of real-time motion management with helical tomotherapy.

PURPOSE: This study investigates the potential application of image-based motion tracking and real-time motion correction to a helical tomotherapy system. METHODS: A kV x-ray imaging system was added to a helical tomotherapy system, mounted 90 degrees offset from the MV treatment beam, and an optical camera system was mounted above the foot of the couch. This experimental system tracks target motion by acquiring an x-ray image every few seconds during gantry rotation. For respiratory (periodic) motion, software correlates internal target positions visible in the x-ray images with marker positions detected continuously by the camera, and generates an internal-external correlation model to continuously determine the target position in three-dimensions (3D). Motion correction is performed by continuously updating jaw positions and MLC leaf patterns to reshape (effectively re-pointing) the treatment beam to follow the 3D target motion. For motion due to processes other than respiration (e.g., digestion), no correlation model is used - instead, target tracking is achieved with the periodically acquired x-ray images, without correlating with a continuous camera signal. RESULTS: The system's ability to correct for respiratory motion was demonstrated using a helical treatment plan delivered to a small (10 mm diameter) target. The phantom was moved following a breathing trace with an amplitude of 15 mm. Film measurements of delivered dose without motion, with motion, and with motion correction were acquired. Without motion correction, dose differences within the target of up to 30% were observed. With motion correction enabled, dose differences in the moving target were less than 2%. Nonrespiratory system performance was demonstrated using a helical treatment plan for a 55 mm diameter target following a prostate motion trace with up to 14 mm of motion. Without motion correction, dose differences up to 16% and shifts of greater than 5 mm were observed. Motion correction reduced these to less than a 6% dose difference and shifts of less than 2 mm. CONCLUSIONS: Real-time motion tracking and correction is technically feasible on a helical tomotherapy system. In one experiment, dose differences due to respiratory motion were greatly reduced. Dose differences due to nonrespiratory motion were also reduced, although not as much as in the respiratory case due to less frequent tracking updates. In both cases, beam-on time was not increased by motion correction, since the system tracks and corrects for motion simultaneously with treatment delivery.

Identifying a minimal rheological configuration: a tool for effective and efficient constitutive modeling of soft tissues.

We describe a modeling methodology intended as a preliminary step in the identification of appropriate constitutive frameworks for the time-dependent response of biological tissues. The modeling approach comprises a customizable rheological network of viscous and elastic elements governed by user-defined 1D constitutive relationships. The model parameters are identified by iterative nonlinear optimization, minimizing the error between experimental and model-predicted structural (load-displacement) tissue response under a specific mode of deformation. We demonstrate the use of this methodology by determining the minimal rheological arrangement, constitutive relationships, and model parameters for the structural response of various soft tissues, including ex vivo perfused porcine liver in indentation, ex vivo porcine brain cortical tissue in indentation, and ex vivo human cervical tissue in unconfined compression. Our results indicate that the identified rheological configurations provide good agreement with experimental data, including multiple constant strain rate load/unload tests and stress relaxation tests. Our experience suggests that the described modeling framework is an efficient tool for exploring a wide array of constitutive relationships and rheological arrangements, which can subsequently serve as a basis for 3D constitutive model development and finite-element implementations. The proposed approach can also be employed as a self-contained tool to obtain simplified 1D phenomenological models of the structural response of biological tissue to single-axis manipulations for applications in haptic technologies.

A nonrigid image registration framework for identification of tissue mechanical parameters.

We present a modular framework for mechanically regularized nonrigid image registration of 3D ultrasound and for identification of tissue mechanical parameters. Mechanically regularized deformation fields are computed from sparsely estimated local displacements. We enforce image-based local motion estimates by applying concentrated forces at mesh nodes of a mechanical finite-element model. The concentrated forces are generated by the elongation of regularization springs connected to the mesh nodes as their free ends are displaced according to local motion estimates. The regularization energy corresponding to the potential energy stored in the springs is minimized when the mechanical response of the model matches the observed response of the organ. We demonstrate that this technique is suitable for identification of material parameters of a nonlinear viscoelastic liver model and demonstrate its benefits over traditional indentation methods in terms of improved volumetric agreement between the model response and the experiment.

Stereo display of 3D ultrasound images for surgical robot guidance.

The recent advent of real-time 3-D ultrasound (3DUS) imaging enables a variety of surgical procedures to be performed within the beating heart. Implementation of these procedures is hampered by the difficulty of manipulating tissue guided by the distorted, low resolution 3DUS images and the dexterity constraints imposed by the confined intracardiac space. This paper investigates the use of surgical robotics in conjunction with 3DUS to overcome these limitations. In addition, it describes the development of a graphics processor based volume Tenderer for real-time stereo visualization of the ultrasound data. Stereo displayed 3DUS was compared to ID-displayed 3DUS and endoscopic guidance with a user study. Five subjects performed in vitro surgical tasks using a surgical robot. Results indicate that subjects were able to complete surgical tasks 35 % faster with stereo-displayed 3DUS images compared to conventional two dimensional display of 3DUS.