Diffeomorphic Registration Research Articles

Diffeomorphic Registration Enhances Retinotopic Mapping in 3T

Diffeomorphic Registration Enhances Retinotopic Mapping in 3T

Explicit–implicit symmetric diffeomorphic deformable image registration with convolutional neural network

AbstractMedical image registration is essential and a key step in many advanced medical image tasks. In recent years, medical image registration has been applied to many clinical diagnoses, but large deformation registration is still a challenge. Deep learning‐based methods typically have higher accuracy but do not involve spatial transformation, which ignores some desirable properties, including topology preservation and the invertibility of transformation, for medical imaging studies. On the other hand, diffeomorphic registration methods achieve a differentiable spatial transformation, which guarantees topology preservation and invertibility of transformation, but registration accuracy is low. Therefore, a diffeomorphic deformation registration with CNN is proposed, based on a symmetric architecture, simultaneously estimating forward and inverse deformation fields. CNN with Efficient Channel Attention is used to better capture the spatial relationship. Deformation fields are optimized explicitly and implicitly to enhance the invertibility of transformations. An extensive experimental evaluation is performed using two 3D datasets. The proposed method is compared with different state‐of‐the‐art methods. The experimental results show excellent registration accuracy while better guaranteeing the diffeomorphic transformation.

Insights into traditional Large Deformation Diffeomorphic Metric Mapping and unsupervised deep-learning for diffeomorphic registration and their evaluation

This paper explores the connections between traditional Large Deformation Diffeomorphic Metric Mapping methods and unsupervised deep-learning approaches for non-rigid registration, particularly emphasizing diffeomorphic registration. The study provides useful insights and establishes connections between the methods, thereby facilitating a profound understanding of the methodological landscape. The methods considered in our study are extensively evaluated in T1w MRI images using traditional NIREP and Learn2Reg OASIS evaluation protocols with a focus on fairness, to establish equitable benchmarks and facilitate informed comparisons. Through a comprehensive analysis of the results, we address key questions, including the intricate relationship between accuracy and transformation quality in performance, the disentanglement of the influence of registration ingredients on performance, and the determination of benchmark methods and baselines. We offer valuable insights into the strengths and limitations of both traditional and deep-learning methods, shedding light on their comparative performance and guiding future advancements in the field.

MDB-40. DEFORMATION-HETEROGENEITY RADIOMIC MEASURES TO RISK-STRATIFY PEDIATRIC MEDULLOBLASTOMA PATIENTS: A PRELIMINARY ANALYSIS

Abstract BACKGROUND A significant challenge in medulloblastoma (MB), the most frequent malignant brain tumor in children, is to accurately identify ‘average-risk’ patients who may benefit from therapy de-escalation without affecting treatment outcomes, from ‘high-risk’ patients. Unfortunately, the current clinical risk-assessment approaches (Chang/molecular stratification) are sub-optimal in reliable risk-stratification in MB. This work evaluates the hypothesis that infiltration in aggressive tumors manifests as deformations (shifts) with larger magnitudes and positive orientations (angles) in the surrounding healthy tissue, on MRI scans, compared to average-risk tumors (smaller shift magnitudes and negative orientations). The proposed imaging marker, complementary with current approaches, may allow for more reliable risk-stratification. METHODS T1-weighted MRI scans of MB patients (2–18 years) were collected from Cincinnati Children’s Hospital Medical Center (Site1-training) (n=42) and Children’s Hospital Los Angeles (Site2-testing) (n=45). Preprocessing and annotations for the edema sub-compartment were performed. Scans were mapped to age-specific atlases using diffeomorphic registration and inverse transformations were employed to obtain the deformations. Deformation magnitudes (DefMag) were extracted within equidistant bands outside the edema by calculating the Euclidean norm of deformations, generating a magnitude statistics vector (median, skewness, kurtosis, mean, standard-deviation). Angles/Orientations (Defθ) were computed between deformation vectors and vectors connecting each voxel to the tumor’s centroid, then quantized to 6 intervals, creating an orientation vector per band. Deformation magnitudes and angles (DefMag+θ) were employed with Chang and molecular stratification within regression models for risk-stratification. RESULTS Our analysis demonstrated that deformation features, complementing molecular and Chang’s stratification, yielded best results for risk-stratification. Employing DefMag+θ+Chang+molecular on training yielded Concordance-index (CI) of 0.56, p=0.0014 (CI=0.6,p=0.03 testing). DefMag+θ+Chang’s yielded CI=0.64, p<0.0001 on training (CI=0.54,p=0.04 testing). Results were not significant when employing Chang’s or molecular classification alone. CONCLUSIONS Our results suggest that radiomic deformation features, complementing clinical and molecular approaches, may enable reliable risk-stratification in pediatric MB.

Diffeomorphic transformer-based abdomen MRI-CT deformable image registration.

Stereotactic body radiotherapy (SBRT) is a well-established treatment modality for liver metastases in patients unsuitable for surgery. Both CT and MRI are useful during treatment planning for accurate target delineation and to reduce potential organs-at-risk (OAR) toxicity from radiation. MRI-CT deformable image registration (DIR) is required to propagate the contours defined on high-contrast MRI to CT images. An accurate DIR method could lead to more precisely defined treatment volumes and superior OAR sparing on the treatment plan. Therefore, it is beneficial to develop an accurate MRI-CT DIR for liver SBRT. To create a new deep learning model that can estimate the deformation vector field (DVF) for directly registering abdominal MRI-CT images. The proposed method assumed a diffeomorphic deformation. By using topology-preserved deformation features extracted from the probabilistic diffeomorphic registration model, abdominal motion can be accurately obtained and utilized for DVF estimation. The model integrated Swin transformers, which have demonstrated superior performance in motion tracking, into the convolutional neural network (CNN) for deformation feature extraction. The model was optimized using a cross-modality image similarity loss and a surface matching loss. To compute the image loss, a modality-independent neighborhood descriptor (MIND) was used between the deformed MRI and CT images. The surface matching loss was determined by measuring the distance between the warped coordinates of the surfaces of contoured structures on the MRI and CT images. To evaluate the performance of the model, a retrospective study was carried out on a group of 50 liver cases that underwent rigid registration of MRI and CT scans. The deformed MRI image was assessed against the CT image using the target registration error (TRE), Dice similarity coefficient (DSC), and mean surface distance (MSD) between the deformed contours of the MRI image and manual contours of the CT image. When compared to only rigid registration, DIR with the proposed method resulted in an increase of the mean DSC values of the liver and portal vein from 0.850±0.102 and 0.628±0.129 to 0.903±0.044 and 0.763±0.073, a decrease of the mean MSD of the liver from 7.216±4.513mm to 3.232±1.483mm, and a decrease of the TRE from 26.238±2.769mm to 8.492±1.058mm. The proposed DIR method based on a diffeomorphic transformer provides an effective and efficient way to generate an accurate DVF from an MRI-CT image pair of the abdomen. It could be utilized in the current treatment planning workflow for liver SBRT.

High-resolution hybrid micro-CT imaging pipeline for mouse brain region segmentation and volumetric morphometry.

Brain region segmentation and morphometry in humanized apolipoprotein E (APOE) mouse models with a human NOS2 background (HN) contribute to Alzheimer's disease (AD) research by demonstrating how various risk factors affect the brain. Photon-counting detector (PCD) micro-CT provides faster scan times than MRI, with superior contrast and spatial resolution to energy-integrating detector (EID) micro-CT. This paper presents a pipeline for mouse brain imaging, segmentation, and morphometry from PCD micro-CT. We used brains of 26 mice from 3 genotypes (APOE22HN, APOE33HN, APOE44HN). The pipeline included PCD and EID micro-CT scanning, hybrid (PCD and EID) iterative reconstruction, and brain region segmentation using the Small Animal Multivariate Brain Analysis (SAMBA) tool. We applied SAMBA to transfer brain region labels from our new PCD CT atlas to individual PCD brains via diffeomorphic registration. Region-based and voxel-based analyses were used for comparisons by genotype and sex. Together, PCD and EID scanning take ~5 hours to produce images with a voxel size of 22 μm, which is faster than MRI protocols for mouse brain morphometry with voxel size above 40 μm. Hybrid iterative reconstruction generates PCD images with minimal artifacts and higher spatial resolution and contrast than EID images. Our PCD atlas is qualitatively and quantitatively similar to the prior MRI atlas and successfully transfers labels to PCD brains in SAMBA. Male and female mice had significant volume differences in 26 regions, including parts of the entorhinal cortex and cingulate cortex. APOE22HN brains were larger than APOE44HN brains in clusters from the hippocampus, a region where atrophy is associated with AD. This work establishes a pipeline for mouse brain analysis using PCD CT, from staining to imaging and labeling brain images. Our results validate the effectiveness of the approach, setting a foundation for research on AD mouse models while reducing scanning durations.

LDDMM-Face: Large deformation diffeomorphic metric learning for cross-annotation face alignment

We propose an innovative, flexible, and consistent cross-annotation face alignment framework, LDDMM-Face, the key contribution of which is a deformation layer that naturally embeds facial geometry in a diffeomorphic way. This enables and solves cross-annotation face alignment tasks that were impossible in the existing works. Instead of predicting facial landmarks via a heatmap or coordinate regression, we formulate the face alignment task in a diffeomorphic registration manner and predict momenta that uniquely parameterize the deformation between the initial boundary and true boundary. We then perform large deformation diffeomorphic metric mapping (LDDMM) simultaneously for curve and landmark to localize the facial landmarks. The novel embedding of LDDMM into a deep network allows LDDMM-Face to consistently annotate facial landmarks without ambiguity and flexibly handle various annotation schemes, and can even predict dense annotations from sparse ones. To the best of our knowledge, this is the first study to leverage learning-based diffeomorphic mapping for face alignment. Our method can be easily integrated into various face alignment networks. We extensively evaluate LDDMM-Face on five benchmark datasets: 300W, WFLW, HELEN, COFW-68, and AFLW. LDDMM-Face distinguishes itself with outstanding performance when dealing with within-dataset cross-annotation learning (sparse-to-dense) and cross-dataset learning (different training and testing datasets). In addition, LDDMM-Face shows promising results on the most challenging task of cross-dataset cross-annotation learning (different training and testing datasets with different annotations). Our codes are available at https://github.com/CRazorback/LDDMM-Face.

ENHANCING 3T RETINOTOPIC MAPS USING DIFFEOMORPHIC REGISTRATION.

Retinotopic mapping aims to uncover the relationship between visual stimuli on the retina and neural responses on the visual cortical surface. This study advances retinotopic mapping by applying diffeomorphic registration to the 3T NYU retinotopy dataset, encompassing analyze-PRF and mrVista data. Diffeomorphic Registration for Retinotopic Maps (DRRM) quantifies the diffeomorphic condition, ensuring accurate alignment of retinotopic maps without topological violations. Leveraging the Beltrami coefficient and topological condition, DRRM significantly enhances retinotopic map accuracy. Evaluation against existing methods demonstrates DRRM's superiority on various datasets, including 3T and 7T retinotopy data. The application of diffeomorphic registration improves the interpretability of low-quality retinotopic maps, holding promise for clinical applications.

Normal and equivolumetric coordinate systems for cortical areas

We describe coordinate systems adapted for the space between two surfaces, such as those delineating the highly folded cortex in mammalian brains. These systems are estimated in order to satisfy geometric priors, including streamline normality or equivolumetric conditions on layers. We give a precise mathematical formulation of these problems, and present numerical simulations based on diffeomorphic registration methods, comparing them with recent approaches.Our method involves•Diffeomorphic registration of inner and outer folded surfaces.•Followed by equivolumetric reparametrization of layers to yield coordinate system.

A bi-variant variational model for diffeomorphic image registration with relaxed Jacobian determinant constraints

Diffeomorphic registration is a widely used technique for finding a smooth and invertible transformation between two coordinate systems, which are measured using template and reference images. The point-wise volume-preserving constraint det⁡(∇φ(x))=1 is effective in some cases, but may be too restrictive in others, especially when local deformations are relatively large. This can result in poor matching when enforcing large local deformations. In this paper, we propose a new bi-variant diffeomorphic image registration model that introduces a soft constraint on the Jacobian equation det⁡(∇φ(x))=f(x)>0. This allows local deformations to shrink and grow within a flexible range 0<κm<det⁡(∇φ(x))<κM. The Jacobian determinant of transformation is explicitly controlled by optimizing the relaxation function f(x). To prevent deformation folding and improve the smoothness of the transformation, a positive constraint is imposed on the optimization of the relaxation function f(x), and a regularizer is used to ensure the smoothness of f(x). Furthermore, the positivity constraint ensures that f(x) is as close to one as possible, which helps to achieve a volume-preserving transformation on average. We also analyze the existence of the minimizer for the variational model and propose a penalty-splitting algorithm with a multilevel strategy to solve this model. Numerical experiments demonstrate the convergence of the proposed algorithm and show that the positivity constraint can effectively control the range of relative volume without compromising the accuracy of the registration. Moreover, the proposed model generates diffeomorphic maps for large local deformations and outperforms several existing registration models in terms of performance.

MEBRAINS 1.0: A new population-based macaque atlas

Abstract Due to their fundamental relevance, the number of anatomical macaque brain templates is constantly growing. Novel templates aim to alleviate limitations of previously published atlases and offer the foundation to integrate multiscale multimodal data. Typical limitations of existing templates include their reliance on one subject, their unimodality (usually only T1 or histological images), or lack of anatomical details. The MEBRAINS template overcomes these limitations by using a combination of T1 and T2 images, from the same 10 animals (Macaca mulatta), which are averaged by the multi-brain toolbox for diffeomorphic registration and segmentation. The resulting volumetric T1 and T2 templates are supplemented with high-quality white and gray matter surfaces built with FreeSurfer. Human-curated segmentations of pial surface, the white/gray matter interface, and major subcortical nuclei were used to analyze the relative quality of the MEBRAINS template. Additionally, 9 computed tomography (CT) scans of the same monkeys were registered to the T1 modality and co-registered to the template. Through its main features (multi-subject, multimodal, volume-and-surface, traditional, and deep learning-based segmentations), MEBRAINS aims to improve integration of multimodal multi-scale macaque data and is quantitatively equal to, or better than, currently widely used macaque templates. We provide a detailed description of the algorithms/methods used to create the template aiming to furnish future researchers with a map-like perspective which should facilitate identification of an optimal pipeline for the task they have at hand. Finally, recently published 3D maps of the macaque inferior parietal lobe, (pre)motor and prefrontal cortex were warped to the MEBRAINS surface template, thus populating it with a parcellation scheme based on cyto- and receptor architectonic analyses. The template is integrated in the EBRAINS and Scalable Brain Atlas web-based infrastructures, each of which comes with its own suite of spatial registration tools.

NIMG-84. COMPARISON OF TRADITIONAL AND ADVANCED DEEP-LEARNING PATIENT SPECIFIC REGISTRATION METHODS IN PRESERVING TUMOR HABITAT IN GLIOBLASTOMA TUMORS: A PRELIMINARY STUDY

Abstract PURPOSE Registration of patient specific pre and post operative MRI scans in Glioblastoma patients is crucial to assess tumor progression, effect of radiation, and chemotherapy and survival prediction. However, in brain tumors longitudinal registration is challenging as tumor resection and mass effects causes tissue deformations and missing intensity correspondence across longitudinal scans. In this study, we compared effectiveness of traditional registration algorithms with state-of-the-art computationally extensive deep learning (DL) models in preserving intensity of tumor regions. METHODS Brain MRI scans (T1w,T1c,T2w,FLAIR) of 14 glioblastoma patients with first (tp1), intermediate (tp2) and follow up post operative scans (tp3) were obtained from two publicly available dataset (LUMIERE, XCURE). Preprocessing involved co-registration of all modalities within a timepoint, skull stripping, and uniformity in image dimensions, orientation, and origin across all timepoints. Tumor scans at tp2 and tp3 were registered to tp1 for all modalities independently, using different registration algorithms such as the traditional affine and symmetric diffeomorphic (SyN) registration with and without tumor masks using the ANTs software, along with the MICCAI-registration challenge winner DL algorithm (DIRAC). Intensity distribution of registered and unregistered scans for tp2 and tp3 were compared using the earth mover’s distance (EMD). RESULTS Traditional affine (tp2-0.074±0.06, tp3:0.058±0.051) and SyN (tp2-0.048±0.031 ,tp3:0.075±0.074) registration algorithm using T2w modality provided lower average EMD than DL method (tp2-0.624±0.69,tp3-0.595±0.67). Intensity distribution was better preserved for scans from intermediate timepoints compared to the post operative scans. CONCLUSION Our preliminary assessment suggests that traditional registration methods when implemented on pre-aligned MRI scans may better preserve image intensities in brain tumors scans compared to advanced and computationally expensive DL methods.

PC-Reg: A pyramidal prediction–correction approach for large deformation image registration

Deformable image registration plays an important role in medical image analysis. Deep neural networks such as VoxelMorph and TransMorph are fast, but limited to small deformations and face challenges in the presence of large deformations. To tackle large deformations in medical image registration, we propose PC-Reg, a pyramidal Prediction and Correction method for deformable registration, which treats multi-scale registration akin to solving an ordinary differential equation (ODE) across scales. Starting with a zero-initialized deformation at the coarse level, PC-Reg follows the predictor–corrector regime and progressively predicts a residual flow and a correction flow to update the deformation vector field through different scales. The prediction in each scale can be regarded as a single step of ODE integration. PC-Reg can be easily extended to diffeomorphic registration and is able to alleviate the multiscale accumulated upsampling and diffeomorphic integration error. Further, to transfer details from full resolution to low scale, we introduce a distillation loss, where the output is used as the target label for intermediate outputs. Experiments on inter-patient deformable registration show that the proposed method significantly improves registration not only for large but also for small deformations.

SequenceMorph: A Unified Unsupervised Learning Framework for Motion Tracking on Cardiac Image Sequences.

Modern medical imaging techniques, such as ultrasound (US) and cardiac magnetic resonance (MR) imaging, have enabled the evaluation of myocardial deformation directly from an image sequence. While many traditional cardiac motion tracking methods have been developed for the automated estimation of the myocardial wall deformation, they are not widely used in clinical diagnosis, due to their lack of accuracy and efficiency. In this paper, we propose a novel deep learning-based fully unsupervised method, SequenceMorph, for in vivo motion tracking in cardiac image sequences. In our method, we introduce the concept of motion decomposition and recomposition. We first estimate the inter-frame (INF) motion field between any two consecutive frames, by a bi-directional generative diffeomorphic registration neural network. Using this result, we then estimate the Lagrangian motion field between the reference frame and any other frame, through a differentiable composition layer. Our framework can be extended to incorporate another registration network, to further reduce the accumulated errors introduced in the INF motion tracking step, and to refine the Lagrangian motion estimation. By utilizing temporal information to perform reasonable estimations of spatio-temporal motion fields, this novel method provides a useful solution for image sequence motion tracking. Our method has been applied to US (echocardiographic) and cardiac MR (untagged and tagged cine) image sequences; the results show that SequenceMorph is significantly superior to conventional motion tracking methods, in terms of the cardiac motion tracking accuracy and inference efficiency.

Craniofacial Soft-Tissue Anthropomorphic Database with Magnetic Resonance Imaging and Unbiased Diffeomorphic Registration.

Objective assessment of craniofacial surgery outcomes in a pediatric population is challenging because of the complexity of patient presentations, diversity of procedures performed, and rapid craniofacial growth. There is a paucity of robust methods to quantify anatomical measurements by age and objectively compare craniofacial dysmorphology and postoperative outcomes. Here, the authors present data in developing a racially and ethnically sensitive anthropomorphic database, providing plastic and craniofacial surgeons with "normal" three-dimensional anatomical parameters with which to appraise and optimize aesthetic and reconstructive outcomes. Patients with normal craniofacial anatomy undergoing head magnetic resonance imaging (MRI) scans from 2008 to 2021 were included in this retrospective study. Images were used to construct composite (template) images with diffeomorphic image registration method using the Advanced Normalization Tools package. Composites were thresholded to generate binary three-dimensional segmentations used for anatomical measurements in Materalise Mimics. High-resolution MRI scans from 130 patients generated 12 composites from an average of 10 MRI sequences each: four 3-year-olds, four 4-year-olds, and four 5-year-olds (two male, two female, two Black, and two White). The average head circumference of 3-, 4-, and 5-year-old composites was 50.3, 51.5, and 51.7 cm, respectively, comparable to normative data published by the World Health Organization. Application of diffeomorphic registration-based image template algorithm to MRI is effective in creating composite templates to represent "normal" three-dimensional craniofacial and soft-tissue anatomy. Future research will focus on development of automated computational tools to characterize anatomical normality, generation of indices to grade preoperative severity, and quantification of postoperative results to reduce subjectivity bias.

Symmetric pyramid network for medical image inverse consistent diffeomorphic registration.

Over the past few years, deep learning-based image registration methods have achieved remarkable performance in medical image analysis. However, many existing methods struggle to ensure accurate registration while preserving the desired diffeomorphic properties and inverse consistency of the final deformation field. To address the problem, this paper presents a novel symmetric pyramid network for medical image inverse consistent diffeomorphic registration. Specifically, we first encode the multi-scale images to the feature pyramids via a shared-weights encoder network and then progressively conduct the feature-level diffeomorphic registration. The feature-level registration is implemented symmetrically to ensure inverse consistency. We independently carry out the forward and backward feature-level registration and average the estimated bidirectional velocity fields for more robust estimation. Finally, we employ symmetric multi-scale similarity loss to train the network. Experimental results on three public datasets, including Mindboggle101, CANDI, and OAI, show that our method significantly outperforms others, demonstrating that the proposed network can achieve accurate alignment and generate the deformation fields with expected properties. Our code will be available at https://github.com/zhangliutong/SPnet.

Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration.

Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with ifdefined tmiformat R2.5a state of the art method methods in the literature in a cohort of 72 fetal datasets in the range of 20-36 weeks gestational age. Results suggest improved image resolution Quantitative as well as radiological assessment suggest improved image quality and more accurate prediction of gestational age at scan is obtained when comparing to a state of the art reconstruction method methods. In addition, gestational age prediction results from our volumetric reconstructions compare favourably are competitive with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of 0.618 weeks ( R2=0.958 ) is achieved when combining fetal brain and trunk information.

Diffeomorphic Registration of 3D Surfaces with Point and Curve Landmarks

Diffeomorphic Registration of 3D Surfaces with Point and Curve Landmarks

Diffeomorphic Registration Using Sinkhorn Divergences

The diffeomorphic registration framework enables one to define an optimal matching function between two probability measures with respect to a data-fidelity loss function. The nonconvexity of the optimization problem renders the choice of this loss function crucial to avoid poor local minima. Recent work showed experimentally the efficiency of entropy-regularized optimal transportation costs, as they are computationally fast and differentiable while having few minima. Following this approach, we provide in this paper a new framework based on Sinkhorn divergences, unbiased entropic optimal transportation costs, and prove the statistical consistency with rate of the empirical optimal deformations.

A Novel 3D-to-3D Diffeomorphic Registration Algorithm With Applications to Left Ventricle Segmentation in MR and Ultrasound Sequences

Left ventricular segmentation is a difficult and time-consuming task performed by clinicians, requiring the use of manual contours. We propose a novel three-dimensional diffeomorphic registration algorithm for endocardial segmentation of the left ventricle in magnetic resonance images and ultrasound temporal sequences. The proposed diffeomorphic registration method computes a voxel-to-voxel correspondence in three-dimensional space and is parameterized by one radial and three curl components to emulate the cardiac deformations. In addition, the proposed method allows for enforcing constraints to control the amount of deformation. The method was evaluated on 521 temporal frames from 20 patients from the Automated Cardiac Diagnosis Challenge magnetic resonance imaging dataset and 213 frames from 10 patients undergoing ultrasound scans from the Mazankowski Alberta Heart Institute. The algorithm was compared against six other registration methods, the Symmetric Normalization algorithm from the Dipy package, two variants of the Demons algorithm from the Insight Toolkit software package, two versions of RealTiTracker, and Elastix. The proposed method yielded overall Dice scores of 98.10 (0.90)% for the MRI dataset and 92.90 (2.42)% for the ultrasound dataset. The robustness of the algorithm is demonstrated by the high performance on multiple imaging modalities and various patient abnormalities.