Learning from multiple annotators for medical image segmentation.

Supervised machine learning methods have been widely developed for segmentation tasks in recent years. However, the quality of labels has high impact on the predictive performance of these algorithms. This issue is particularly acute in the medical image domain, where both the cost of annotation and the inter-observer variability are high. Different human experts contribute estimates of the "actual" segmentation labels in a typical label acquisition process, influenced by their personal biases and competency levels. The performance of automatic segmentation algorithms is limited when these noisy labels are used as the expert consensus label. In this work, we use two coupled CNNs to jointly learn, from purely noisy observations alone, the reliability of individual annotators and the expert consensus label distributions. The separation of the two is achieved by maximally describing the annotator's "unreliable behavior" (we call it "maximally unreliable") while achieving high fidelity with the noisy training data. We first create a toy segmentation dataset using MNIST and investigate the properties of the proposed algorithm. We then use three public medical imaging segmentation datasets to demonstrate our method's efficacy, including both simulated (where necessary) and real-world annotations: 1) ISBI2015 (multiple-sclerosis lesions); 2) BraTS (brain tumors); 3) LIDC-IDRI (lung abnormalities). Finally, we create a real-world multiple sclerosis lesion dataset (QSMSC at UCL: Queen Square Multiple Sclerosis Center at UCL, UK) with manual segmentations from 4 different annotators (3 radiologists with different level skills and 1 expert to generate the expert consensus label). In all datasets, our method consistently outperforms competing methods and relevant baselines, especially when the number of annotations is small and the amount of disagreement is large. The studies also reveal that the system is capable of capturing the complicated spatial characteristics of annotators' mistakes.

3D whole brain segmentation using spatially localized atlas network tiles.

Detailed whole brain segmentation is an essential quantitative technique in medical image analysis, which provides a non-invasive way of measuring brain regions from a clinical acquired structural magnetic resonance imaging (MRI). Recently, deep convolution neural network (CNN) has been applied to whole brain segmentation. However, restricted by current GPU memory, 2D based methods, downsampling based 3D CNN methods, and patch-based high-resolution 3D CNN methods have been the de facto standard solutions. 3D patch-based high resolution methods typically yield superior performance among CNN approaches on detailed whole brain segmentation (>100 labels), however, whose performance are still commonly inferior compared with state-of-the-art multi-atlas segmentation methods (MAS) due to the following challenges: (1) a single network is typically used to learn both spatial and contextual information for the patches, (2) limited manually traced whole brain volumes are available (typically less than 50) for training a network. In this work, we propose the spatially localized atlas network tiles (SLANT) method to distribute multiple independent 3D fully convolutional networks (FCN) for high-resolution whole brain segmentation. To address the first challenge, multiple spatially distributed networks were used in the SLANT method, in which each network learned contextual information for a fixed spatial location. To address the second challenge, auxiliary labels on 5111 initially unlabeled scans were created by multi-atlas segmentation for training. Since the method integrated multiple traditional medical image processing methods with deep learning, we developed a containerized pipeline to deploy the end-to-end solution. From the results, the proposed method achieved superior performance compared with multi-atlas segmentation methods, while reducing the computational time from >30 h to 15 min. The method has been made available in open source (https://github.com/MASILab/SLANTbrainSeg).

Automatic Labeling of MR Brain Images Through the Hashing Retrieval Based Atlas Forest.

The multi-atlas method is one of the efficient and common automatic labeling method, which uses the prior information provided by expert-labeled images to guide the labeling of the target. However, most multi-atlas-based methods depend on the registration that may not give the correct information during the label propagation. To address the issue, we designed a new automatic labeling method through the hashing retrieval based atlas forest. The proposed method propagates labels without registration to reduce the errors, and constructs a target-oriented learning model to integrate information among the atlases. This method innovates a coarse classification strategy to preprocess the dataset, which retains the integrity of dataset and reduces computing time. Furthermore, the method considers each voxel in the atlas as a sample and encodes these samples with hashing for the fast sample retrieval. In the stage of labeling, the method selects suitable samples through hashing learning and trains atlas forests by integrating the information from the dataset. Then, the trained model is used to predict the labels of the target. Experimental results on two datasets illustrated that the proposed method is promising in the automatic labeling of MR brain images.

An Efficient Melanoma Diagnosis Approach Using Integrated HMF Multi-Atlas Map Based Segmentation.

Melanoma is a life threading disease when it grows outside the corium layer of the skin. Mortality rates of the Melanoma cases are maximum among the skin cancer patients. The cost required for the treatment of advanced melanoma cases is very high and the survival rate is low. Numerous computerized dermoscopy systems are developed based on the combination of shape, texture and color features to facilitate early diagnosis of melanoma. The availability and cost of the dermoscopic imaging system is still an issue. To mitigate this issue, this paper presented an integrated segmentation and Third Dimensional (3D) feature extraction approach for the accurate diagnosis of melanoma. A multi-atlas method is applied for the image segmentation. The patch-based label fusion model is expressed in a Bayesian framework to improve the segmentation accuracy. A depth map is obtained from the Two-dimensional (2D) dermoscopic image for reconstructing the 3D skin lesion represented as structure tensors. The 3D shape features including the relative depth features are obtained. Streaks are the significant morphological terms of the melanoma in the radial growth phase. The proposed method yields maximum segmentation accuracy, sensibility, specificity and minimum cost function than the existing segmentation technique and classifier.

[A multi-label fusion based level set method for multiple sclerosis lesion segmentation].

A multi-label based level set model for multiple sclerosis lesion segmentation is proposed based on the shape, position and other information of lesions from magnetic resonance image. First, fuzzy c-means model is applied to extract the initial lesion region. Second, an intensity prior information term and a label fusion term are constructed using intensity information of the initial lesion region, the above two terms are integrated into a region-based level set model. The final lesion segmentation is achieved by evolving the level set contour. The experimental results show that the proposed method can accurately and robustly extract brain lesions from magnetic resonance images. The proposed method helps to reduce the work of radiologists significantly, which is useful in clinical application.

An efficient and accurate method for robust inter-dataset brain extraction and comparisons with 9 other methods.

Brain extraction is an important first step in many magnetic resonance neuroimaging studies. Due to variability in brain morphology and in the appearance of the brain due to differences in scanner acquisition parameters, the development of a generally applicable brain extraction algorithm has proven challenging. Learning-based brain extraction algorithms in particular perform well when the target and training images are sufficiently similar, but often perform worse when this condition is not met. In this study, we propose a new patch-based multi-atlas segmentation method for brain extraction which is specifically developed for accurate and robust processing across datasets. Using a diverse collection of labeled images from 5 different datasets, extensive comparisons were made with 9 other commonly used brain extraction methods, both before and after applying error correction (a machine learning method for automatically correcting segmentation errors) to each method. The proposed method performed equal to or better than the other methods in each of two segmentation scenarios: a challenging inter-dataset segmentation scenario in which no dataset-specific atlases were used (mean Dice coefficient 98.57%, volumetric correlation 0.994 across datasets following error correction), and an intra-dataset segmentation scenario in which only dataset-specific atlases were used (mean Dice coefficient 99.02%, volumetric correlation 0.998 across datasets following error correction). Furthermore, combined with error correction, the proposed method runs in less than one-tenth of the time required by the other top-performing methods in the challenging inter-dataset comparisons. Validation on an independent multi-centre dataset also confirmed the excellent performance of the proposed method.

Multi-object Model-Based Multi-atlas Segmentation Constrained Grid Cut for Automatic Segmentation of Lumbar Vertebrae from CT Images.

In this chapter, we present a multi-object model-based multi-atlas segmentation constrained grid cut method for automatic segmentation of lumbar vertebrae from a given lumbar spinal CT image. More specifically, our automatic lumbar vertebrae segmentation method consists of two steps: affine atlas-target registration-based label fusion and bone-sheetness assisted multi-label grid cut which has the inherent advantage of automatic separation of the five lumbar vertebrae from each other. We evaluate our method on 21 clinical lumbar spinal CT images with the associated manual segmentation and conduct a leave-one-out study. Our method achieved an average Dice coefficient of 93.9 ± 1.0% and an average symmetric surface distance of 0.41 ± 0.08 mm.

CERES: A new cerebellum lobule segmentation method.

The human cerebellum is involved in language, motor tasks and cognitive processes such as attention or emotional processing. Therefore, an automatic and accurate segmentation method is highly desirable to measure and understand the cerebellum role in normal and pathological brain development. In this work, we propose a patch-based multi-atlas segmentation tool called CERES (CEREbellum Segmentation) that is able to automatically parcellate the cerebellum lobules. The proposed method works with standard resolution magnetic resonance T1-weighted images and uses the Optimized PatchMatch algorithm to speed up the patch matching process. The proposed method was compared with related recent state-of-the-art methods showing competitive results in both accuracy (average DICE of 0.7729) and execution time (around 5 minutes).

Simultaneous total intracranial volume and posterior fossa volume estimation using multi-atlas label fusion.

Total intracranial volume (TICV) is an essential covariate in brain volumetric analyses. The prevalent brain imaging software packages provide automatic TICV estimates. FreeSurfer and FSL estimate TICV using a scaling factor while SPM12 accumulates probabilities of brain tissues. None of the three provide explicit skull/CSF boundary (SCB) since it is challenging to distinguish these dark structures in a T1-weighted image. However, explicit SCB not only leads to a natural way of obtaining TICV (i.e., counting voxels inside the skull) but also allows sub-definition of TICV, for example, the posterior fossa volume (PFV). In this article, they proposed to use multi-atlas label fusion to obtain TICV and PFV simultaneously. The main contributions are: (1) TICV and PFV are simultaneously obtained with explicit SCB from a single T1-weighted image. (2) TICV and PFV labels are added to the widely used BrainCOLOR atlases. (3) Detailed mathematical derivation of non-local spatial STAPLE (NLSS) label fusion is presented. As the skull is clearly distinguished in CT images, we use a semi-manual procedure to obtain atlases with TICV and PFV labels using 20 subjects who both have a MR and CT scan. The proposed method provides simultaneous TICV and PFV estimation while achieving more accurate TICV estimation compared with FreeSurfer, FSL, SPM12, and the previously proposed STAPLE based approach. The newly developed TICV and PFV labels for the OASIS BrainCOLOR atlases provide acceptable performance, which enables simultaneous TICV and PFV estimation during whole brain segmentation. The NLSS method and the new atlases have been made freely available. Hum Brain Mapp 38:599-616, 2017. © 2016 Wiley Periodicals, Inc.

Robust multi-atlas label propagation by deep sparse representation.

Recently, multi-atlas patch-based label fusion has achieved many successes in medical imaging area. The basic assumption in the current state-of-the-art approaches is that the image patch at the target image point can be represented by a patch dictionary consisting of atlas patches from registered atlas images. Therefore, the label at the target image point can be determined by fusing labels of atlas image patches with similar anatomical structures. However, such assumption on image patch representation does not always hold in label fusion since (1) the image content within the patch may be corrupted due to noise and artifact; and (2) the distribution of morphometric patterns among atlas patches might be unbalanced such that the majority patterns can dominate label fusion result over other minority patterns. The violation of the above basic assumptions could significantly undermine the label fusion accuracy. To overcome these issues, we first consider forming label-specific group for the atlas patches with the same label. Then, we alter the conventional flat and shallow dictionary to a deep multi-layer structure, where the top layer (label-specific dictionaries) consists of groups of representative atlas patches and the subsequent layers (residual dictionaries) hierarchically encode the patchwise residual information in different scales. Thus, the label fusion follows the representation consensus across representative dictionaries. However, the representation of target patch in each group is iteratively optimized by using the representative atlas patches in each label-specific dictionary exclusively to match the principal patterns and also using all residual patterns across groups collaboratively to overcome the issue that some groups might be absent of certain variation patterns presented in the target image patch. Promising segmentation results have been achieved in labeling hippocampus on ADNI dataset, as well as basal ganglia and brainstem structures, compared to other counterpart label fusion methods.

Scalable Joint Segmentation and Registration Framework for Infant Brain Images.

The first year of life is the most dynamic and perhaps the most critical phase of postnatal brain development. The ability to accurately measure structure changes is critical in early brain development study, which highly relies on the performances of image segmentation and registration techniques. However, either infant image segmentation or registration, if deployed independently, encounters much more challenges than segmentation/registration of adult brains due to dynamic appearance change with rapid brain development. In fact, image segmentation and registration of infant images can assists each other to overcome the above challenges by using the growth trajectories (i.e., temporal correspondences) learned from a large set of training subjects with complete longitudinal data. Specifically, a one-year-old image with ground-truth tissue segmentation can be first set as the reference domain. Then, to register the infant image of a new subject at earlier age, we can estimate its tissue probability maps, i.e., with sparse patch-based multi-atlas label fusion technique, where only the training images at the respective age are considered as atlases since they have similar image appearance. Next, these probability maps can be fused as a good initialization to guide the level set segmentation. Thus, image registration between the new infant image and the reference image is free of difficulty of appearance changes, by establishing correspondences upon the reasonably segmented images. Importantly, the segmentation of new infant image can be further enhanced by propagating the much more reliable label fusion heuristics at the reference domain to the corresponding location of the new infant image via the learned growth trajectories, which brings image segmentation and registration to assist each other. It is worth noting that our joint segmentation and registration framework is also flexible to handle the registration of any two infant images even with significant age gap in the first year of life, by linking their joint segmentation and registration through the reference domain. Thus, our proposed joint segmentation and registration method is scalable to various registration tasks in early brain development studies. Promising segmentation and registration results have been achieved for infant brain MR images aged from 2-week-old to 1-year-old, indicating the applicability of our method in early brain development study.

MUSE: MUlti-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters, and locally optimal atlas selection.

Atlas-based automated anatomical labeling is a fundamental tool in medical image segmentation, as it defines regions of interest for subsequent analysis of structural and functional image data. The extensive investigation of multi-atlas warping and fusion techniques over the past 5 or more years has clearly demonstrated the advantages of consensus-based segmentation. However, the common approach is to use multiple atlases with a single registration method and parameter set, which is not necessarily optimal for every individual scan, anatomical region, and problem/data-type. Different registration criteria and parameter sets yield different solutions, each providing complementary information. Herein, we present a consensus labeling framework that generates a broad ensemble of labeled atlases in target image space via the use of several warping algorithms, regularization parameters, and atlases. The label fusion integrates two complementary sources of information: a local similarity ranking to select locally optimal atlases and a boundary modulation term to refine the segmentation consistently with the target image's intensity profile. The ensemble approach consistently outperforms segmentations using individual warping methods alone, achieving high accuracy on several benchmark datasets. The MUSE methodology has been used for processing thousands of scans from various datasets, producing robust and consistent results. MUSE is publicly available both as a downloadable software package, and as an application that can be run on the CBICA Image Processing Portal (https://ipp.cbica.upenn.edu), a web based platform for remote processing of medical images.

Multi-atlas and label fusion approach for patient-specific MRI based skull estimation.

PURPOSE: MRI-based skull segmentation is a useful procedure for many imaging applications. This study describes a methodology for automatic segmentation of the complete skull from a single T1-weighted volume. METHODS: The skull is estimated using a multi-atlas segmentation approach. Using a whole head computed tomography (CT) scan database, the skull in a new MRI volume is detected by nonrigid image registration of the volume to every CT, and combination of the individual segmentations by label-fusion. We have compared Majority Voting, Simultaneous Truth and Performance Level Estimation (STAPLE), Shape Based Averaging (SBA), and the Selective and Iterative Method for Performance Level Estimation (SIMPLE) algorithms. RESULTS: The pipeline has been evaluated quantitatively using images from the Retrospective Image Registration Evaluation database (reaching an overlap of 72.46 ± 6.99%), a clinical CT-MR dataset (maximum overlap of 78.31 ± 6.97%), and a whole head CT-MRI pair (maximum overlap 78.68%). A qualitative evaluation has also been performed on MRI acquisition of volunteers. CONCLUSION: It is possible to automatically segment the complete skull from MRI data using a multi-atlas and label fusion approach. This will allow the creation of complete MRI-based tissue models that can be used in electromagnetic dosimetry applications and attenuation correction in PET/MR.

Atlas-based estimation of lung and lobar anatomy in proton MRI.

PURPOSE: To propose an accurate methodological framework for automatically segmenting pulmonary proton MRI based on an optimal consensus of a spatially normalized library of annotated lung atlases. METHODS: A library of 62 manually annotated lung atlases comprising 48 mixed healthy, chronic obstructive pulmonary disease, and asthmatic subjects of a large age range with multiple ventilation levels is used to produce an optimal segmentation in proton MRI, based on a consensus of the spatially normalized library. An extension of this methodology is used to provide best-guess estimates of lobar subdivisions in proton MRI from annotated computed tomography data. RESULTS: A leave-one-out evaluation strategy was used for evaluation. Jaccard overlap measures for the left and right lungs were used for performance comparisons relative to the current state-of-the-art (0.966 ± 0.018 and 0.970 ± 0.016, respectively). Best-guess estimates for the lobes exhibited comparable performance levels (left upper: 0.882 ± 0.059, left lower: 0.868 ± 0.06, right upper: 0.852 ± 0.067, right middle: 0.657 ± 0.130, right lower: 0.873 ± 0.063). CONCLUSION: An annotated atlas library approach can be used to provide good lung and lobe estimation in proton MRI. The proposed framework is useful for subsequent anatomically based analysis of structural and/or functional pulmonary image data. Magn Reson Med 76:315-320, 2016. © 2015 Wiley Periodicals, Inc.

An algorithm for optimal fusion of atlases with different labeling protocols.

In this paper we present a novel label fusion algorithm suited for scenarios in which different manual delineation protocols with potentially disparate structures have been used to annotate the training scans (hereafter referred to as "atlases"). Such scenarios arise when atlases have missing structures, when they have been labeled with different levels of detail, or when they have been taken from different heterogeneous databases. The proposed algorithm can be used to automatically label a novel scan with any of the protocols from the training data. Further, it enables us to generate new labels that are not present in any delineation protocol by defining intersections on the underling labels. We first use probabilistic models of label fusion to generalize three popular label fusion techniques to the multi-protocol setting: majority voting, semi-locally weighted voting and STAPLE. Then, we identify some shortcomings of the generalized methods, namely the inability to produce meaningful posterior probabilities for the different labels (majority voting, semi-locally weighted voting) and to exploit the similarities between the atlases (all three methods). Finally, we propose a novel generative label fusion model that can overcome these drawbacks. We use the proposed method to combine four brain MRI datasets labeled with different protocols (with a total of 102 unique labeled structures) to produce segmentations of 148 brain regions. Using cross-validation, we show that the proposed algorithm outperforms the generalizations of majority voting, semi-locally weighted voting and STAPLE (mean Dice score 83%, vs. 77%, 80% and 79%, respectively). We also evaluated the proposed algorithm in an aging study, successfully reproducing some well-known results in cortical and subcortical structures.

Automatic clustering of white matter fibers in brain diffusion MRI with an application to genetics.

To understand factors that affect brain connectivity and integrity, it is beneficial to automatically cluster white matter (WM) fibers into anatomically recognizable tracts. Whole brain tractography, based on diffusion-weighted MRI, generates vast sets of fibers throughout the brain; clustering them into consistent and recognizable bundles can be difficult as there are wide individual variations in the trajectory and shape of WM pathways. Here we introduce a novel automated tract clustering algorithm based on label fusion--a concept from traditional intensity-based segmentation. Streamline tractography generates many incorrect fibers, so our top-down approach extracts tracts consistent with known anatomy, by mapping multiple hand-labeled atlases into a new dataset. We fuse clustering results from different atlases, using a mean distance fusion scheme. We reliably extracted the major tracts from 105-gradient high angular resolution diffusion images (HARDI) of 198 young normal twins. To compute population statistics, we use a pointwise correspondence method to match, compare, and average WM tracts across subjects. We illustrate our method in a genetic study of white matter tract heritability in twins.

Brain extraction based on locally linear representation-based classification.

Brain extraction is an important procedure in brain image analysis. Although numerous brain extraction methods have been presented, enhancing brain extraction methods remains challenging because brain MRI images exhibit complex characteristics, such as anatomical variability and intensity differences across different sequences and scanners. To address this problem, we present a Locally Linear Representation-based Classification (LLRC) method for brain extraction. A novel classification framework is derived by introducing the locally linear representation to the classical classification model. Under this classification framework, a common label fusion approach can be considered as a special case and thoroughly interpreted. Locality is important to calculate fusion weights for LLRC; this factor is also considered to determine that Local Anchor Embedding is more applicable in solving locally linear coefficients compared with other linear representation approaches. Moreover, LLRC supplies a way to learn the optimal classification scores of the training samples in the dictionary to obtain accurate classification. The International Consortium for Brain Mapping and the Alzheimer's Disease Neuroimaging Initiative databases were used to build a training dataset containing 70 scans. To evaluate the proposed method, we used four publicly available datasets (IBSR1, IBSR2, LPBA40, and ADNI3T, with a total of 241 scans). Experimental results demonstrate that the proposed method outperforms the four common brain extraction methods (BET, BSE, GCUT, and ROBEX), and is comparable to the performance of BEaST, while being more accurate on some datasets compared with BEaST.

Nonlocal regularization for active appearance model: Application to medial temporal lobe segmentation.

The human medial temporal lobe (MTL) is an important part of the limbic system, and its substructures play key roles in learning, memory, and neurodegeneration. The MTL includes the hippocampus (HC), amygdala (AG), parahippocampal cortex (PHC), entorhinal cortex, and perirhinal cortex--structures that are complex in shape and have low between-structure intensity contrast, making them difficult to segment manually in magnetic resonance images. This article presents a new segmentation method that combines active appearance modeling and patch-based local refinement to automatically segment specific substructures of the MTL including HC, AG, PHC, and entorhinal/perirhinal cortex from MRI data. Appearance modeling, relying on eigen-decomposition to analyze statistical variations in image intensity and shape information in study population, is used to capture global shape characteristics of each structure of interest with a generative model. Patch-based local refinement, using nonlocal means to compare the image local intensity properties, is applied to locally refine the segmentation results along the structure borders to improve structure delimitation. In this manner, nonlocal regularization and global shape constraints could allow more accurate segmentations of structures. Validation experiments against manually defined labels demonstrate that this new segmentation method is computationally efficient, robust, and accurate. In a leave-one-out validation on 54 normal young adults, the method yielded a mean Dice κ of 0.87 for the HC, 0.81 for the AG, 0.73 for the anterior parts of the parahippocampal gyrus (entorhinal and perirhinal cortex), and 0.73 for the posterior parahippocampal gyrus.

Investigation of morphometric variability of subthalamic nucleus, red nucleus, and substantia nigra in advanced Parkinson's disease patients using automatic segmentation and PCA-based analysis.

Subthalamic nucleus (STN) deep brain stimulation (DBS) is an effective surgical therapy to treat Parkinson's disease (PD). Conventional methods employ standard atlas coordinates to target the STN, which, along with the adjacent red nucleus (RN) and substantia nigra (SN), are not well visualized on conventional T1w MRIs. However, the positions and sizes of the nuclei may be more variable than the standard atlas, thus making the pre-surgical plans inaccurate. We investigated the morphometric variability of the STN, RN and SN by using label-fusion segmentation results from 3T high resolution T2w MRIs of 33 advanced PD patients. In addition to comparing the size and position measurements of the cohort to the Talairach atlas, principal component analysis (PCA) was performed to acquire more intuitive and detailed perspectives of the measured variability. Lastly, the potential correlation between the variability shown by PCA results and the clinical scores was explored.

Three-dimensional hip cartilage quality assessment of morphology and dGEMRIC by planar maps and automated segmentation.

OBJECTIVE: The quantitative interpretation of hip cartilage magnetic resonance imaging (MRI) has been limited by the difficulty of identifying and delineating the cartilage in a three-dimensional (3D) dataset, thereby reducing its routine usage. In this paper a solution is suggested by unfolding the cartilage to planar two-dimensional (2D) maps on which both morphology and biochemical degeneration patterns can be investigated across the entire hip joint. DESIGN: Morphological TrueFISP and biochemical delayed gadolinium enhanced MRI of cartilage (dGEMRIC) hip images were acquired isotropically for 15 symptomatic subjects with mild or no radiographic osteoarthritis (OA). A multi-template based label fusion technique was used to automatically segment the cartilage tissue, followed by a geometric projection algorithm to generate the planar maps. The segmentation performance was investigated through a leave-one-out study, for two different fusion methods and as a function of the number of utilized templates. RESULTS: For each of the generated planar maps, various patterns could be seen, indicating areas of healthy and degenerated cartilage. Dice coefficients for cartilage segmentation varied from 0.76 with four templates to 0.82 with 14 templates. Regional analysis suggests even higher segmentation performance in the superior half of the cartilage. CONCLUSIONS: The proposed technique is the first of its kind to provide planar maps that enable straightforward quantitative assessment of hip cartilage morphology and dGEMRIC values. This technique may have important clinical applications for patient selection for hip preservation surgery, as well as for epidemiological studies of cartilage degeneration patterns. It is also shown that 10-15 templates are sufficient for accurate segmentation in this application.