Atlas Creation Research Articles

Hybrid unsupervised paradigm based deformable image fusion for 4D CT lung image modality

Deformable image registration plays a critical role in various clinical applications (e.g., image fusion, atlas creation, and tumors targeting). In radiation therapy, especially in the context of fast registration of computed tomography (CT) lung image modalities, it is used to determine the geometric transformation by relating the anatomic points in two images. The main challenge lies in effectively addressing the nonlinear large and small deformation between the inspiration and expiration phases. In this work, we propose an unsupervised hybrid paradigm-based registration network (HPRN) for the registration of 4D CT lung images without relying on ground truth data. The proposed HPRN exhibits effective learning of multi-scale and multi-resolution features, leading to the computation of a more accurate Deformation Vector Field (DVF). Furthermore, we incorporate the regularization, image similarity and Jacobian determinant loss functions, which results in improving capability in dealing with complex large and small deformations. We evaluate the effectiveness of the proposed model on the publicly accessible DIRLab 4DCT lung image dataset, which shows the effectiveness of the proposed framework by achieving better Target Registration Error (2.04±1.42mm) compared to other state-of-the-art unsupervised image registration algorithms.

A review of deep learning-based deformable medical image registration.

The alignment of images through deformable image registration is vital to clinical applications (e.g., atlas creation, image fusion, and tumor targeting in image-guided navigation systems) and is still a challenging problem. Recent progress in the field of deep learning has significantly advanced the performance of medical image registration. In this review, we present a comprehensive survey on deep learning-based deformable medical image registration methods. These methods are classified into five categories: Deep Iterative Methods, Supervised Methods, Unsupervised Methods, Weakly Supervised Methods, and Latest Methods. A detailed review of each category is provided with discussions about contributions, tasks, and inadequacies. We also provide statistical analysis for the selected papers from the point of view of image modality, the region of interest (ROI), evaluation metrics, and method categories. In addition, we summarize 33 publicly available datasets that are used for benchmarking the registration algorithms. Finally, the remaining challenges, future directions, and potential trends are discussed in our review.

Цифровізація розвитку міст: Urban Atlas на основі відкритих даних для міст України

Ukraine is an associate member of the European Union and in the coming years it is expected that all data and services already used by EU countries will be available to Ukraine. The lack of quality national products for assessing the development and planning of urban growth makes it impossible to assess the impact of cities on the environment and human health. The first steps to create such products for the cities of Ukraine were initiated within the European project "SMart URBan Solutions for air quality, disasters and city growth" (SMURBS), in which specialists from the Space Research Institute of NAS of Ukraine and SSA of Ukraine received the first city atlas for the Kyiv city, which was similar to the European one. However, the resulting product had significantly fewer types of land use than the European one and therefore the question of improving the developed technology arose. The main purpose of the work is to analyze the existing technology of European service Urban Atlas creation and its improvement by developing a unified algorithm for building an urban atlas using all available open geospatial and satellite data for the cities of Ukraine. The development of such technology is based on our own technology for classifying satellite time series with a spatial resolution of 10 meters to build a land cover map, as well as an algorithm for unifying open geospatial data to urban atlases Copernicus. The technology of construction of the city atlas developed in work, based on the intellectual model of classification of a land cover, can be extended to other cities of Ukraine. In the future, the creation of such a product on the basis of data for different years will allow to assess changes in land use and make a forecast for further urban expansion. The proposed information technology for constructing the city atlas will be useful for assessing the dynamics of urban growth and closely related social and economic indicators of their development. Based on it, it is also possible to assess indicators of achieving the goals of sustainable development, such as 11.3.1 "The ratio of land consumption and population growth." The study shows that the city atlas obtained for the Kyiv city has a high level of quality and has comparable land use classes with European products. It indicates that such a product can be used in government decision-making services.

Об опыте разработки «Градостроительного атласа России» и современном использовании его результатов

В 1991-1994 годах по заданию Минстроя России «Ин­ститут геобиосферных исследований РАЕН» разрабатывал «Градостроительный атлас России» (М 1:8 000 000). В статье с точки зрения научной значимости и возможности современ­ного использования дана оценка выполненных работ. Целью разработки атласа являлось обобщение самой современной на тот период информации о территории России, её оценки с позиций расселения и градостроительства, а задачами - систематизация, анализ и прогноз. Атлас был призван стать символом научного престижа отрасли, предназначался для широкого применения в проектировании, управлении, при обучении специалистов градостроительных специальностей.
 Работа выполнялась в начале «перестройки», в пери­од максимальной открытости информации, интенсивного межотраслевого сотрудничества министерств, ведомств, специалистов различных областей науки, заинтересованных в комплексной оценке территорий страны, её регионов и городов. Это было предпосылкой новизны и уникальности полученных результатов.
 В работе над «Градостроительным атласом России» приняли участие ведущие специалисты в области градостроительства, географии, экологии, геологии, экономики и других научных дисциплин. На начальном этапе были разработаны методология, методика, структура, содержание атласа и ряда тематических карт, их предполагалось около 50. К концу 1993 года было раз­работано 15 карт, пробным тиражом изданы две - посвящённые наиболее актуальным проблемам того времени («Радиационная обстановка России» и «Транспорт России. Основные комму­никации»). Несмотря на прекрасные отзывы специалистов и организаций в период кризиса в 1994 году работы были при­остановлены и впоследствии не возобновлялись.
 Сегодня, когда в соответствии с «майскими» указами Президента России В.В.Путина наступило время активных действий в сфере обоснований стратегического планирования пространственного развития и практического решения акту­альных задач расселения и градостроительства, целесообразно использовать наработки «Градостроительного атласа России», в том числе в области методологии. Методология создания атласа в определённой степени является прообразом формирования будущих отраслевых информационных систем, использующих методы электронного картографирования.
 Значительная часть разработанных карт не потеряла своей актуальности или требует небольшой актуализации, после чего они смогут стать составной частью аналитической базы пространственного развития.

Optimization and validation of automated hippocampal subfield segmentation across the lifespan.

Automated segmentation of hippocampal (HC) subfields from magnetic resonance imaging (MRI) is gaining popularity, but automated procedures that afford high speed and reproducibility have yet to be extensively validated against the standard, manual morphometry. We evaluated the concurrent validity of an automated method for hippocampal subfields segmentation (automated segmentation of hippocampal subfields, ASHS; Yushkevich et al., ) using a customized atlas of the HC body, with manual morphometry as a standard. We built a series of customized atlases comprising the entorhinal cortex (ERC) and subfields of the HC body from manually segmented images, and evaluated the correspondence of automated segmentations with manual morphometry. In samples with age ranges of 6-24 and 62-79 years, 20 participants each, we obtained validity coefficients (intraclass correlations, ICC) and spatial overlap measures (dice similarity coefficient) that varied substantially across subfields. Anterior and posterior HC body evidenced the greatest discrepancies between automated and manual segmentations. Adding anterior and posterior slices for atlas creation and truncating automated output to the ranges manually defined by multiple neuroanatomical landmarks substantially improved the validity of automated segmentation, yielding ICC above 0.90 for all subfields and alleviating systematic bias. We cross-validated the developed atlas on an independent sample of 30 healthy adults (age 31-84) and obtained good to excellent agreement: ICC (2) = 0.70-0.92. Thus, with described customization steps implemented by experts trained in MRI neuroanatomy, ASHS shows excellent concurrent validity, and can become a promising method for studying age-related changes in HC subfield volumes.

Human brain atlasing: past, present and future.

We have recently witnessed an explosion of large-scale initiatives and projects addressing mapping, modeling, simulation and atlasing of the human brain, including the BRAIN Initiative, the Human Brain Project, the Human Connectome Project (HCP), the Big Brain, the Blue Brain Project, the Allen Brain Atlas, the Brainnetome, among others. Besides these large and international initiatives, there are numerous mid-size and small brain atlas-related projects. My contribution to these global efforts has been to create adult human brain atlases in health and disease, and to develop atlas-based applications. For over two decades with my R&D lab I developed 35 brain atlases, licensed to 67 companies and made available in about 100 countries. This paper has two objectives. First, it provides an overview of the state of the art in brain atlasing. Second, as it is already 20 years from the release of our first brain atlas, I summarise my past and present efforts, share my experience in atlas creation, validation and commercialisation, compare with the state of the art, and propose future directions.

A concept for holistic whole body MRI data analysis, Imiomics.

PurposeTo present and evaluate a whole-body image analysis concept, Imiomics (imaging–omics) and an image registration method that enables Imiomics analyses by deforming all image data to a common coordinate system, so that the information in each voxel can be compared between persons or within a person over time and integrated with non-imaging data.MethodsThe presented image registration method utilizes relative elasticity constraints of different tissue obtained from whole-body water-fat MRI.The registration method is evaluated by inverse consistency and Dice coefficients and the Imiomics concept is evaluated by example analyses of importance for metabolic research using non-imaging parameters where we know what to expect.The example analyses include whole body imaging atlas creation, anomaly detection, and cross-sectional and longitudinal analysis.ResultsThe image registration method evaluation on 128 subjects shows low inverse consistency errors and high Dice coefficients. Also, the statistical atlas with fat content intensity values shows low standard deviation values, indicating successful deformations to the common coordinate system.The example analyses show expected associations and correlations which agree with explicit measurements, and thereby illustrate the usefulness of the proposed Imiomics concept.ConclusionsThe registration method is well-suited for Imiomics analyses, which enable analyses of relationships to non-imaging data, e.g. clinical data, in new types of holistic targeted and untargeted big-data analysis.

Stereotaxic Magnetic Resonance Imaging Brain Atlases for Infants from 3 to 12 Months

Background: Accurate labeling of brain structures within an individual or group is a key issue in neuroimaging. Methods for labeling infant brains have depended on the labels done on adult brains or average magnetic resonance imaging (MRI) templates based on adult brains. However, the features of adult brains differ in several ways from infant brains, so the creation of a labeled stereotaxic atlas based on infants would be helpful. The current work builds on the recent creation of age-appropriate average MRI templates during the first year (3, 4.5, 6, 7.5, 9, and 12 months) by creating anatomical label sets for each template. Methods: We created stereotaxic atlases for the age-specific average MRI templates. Manual delineation of cortical and subcortical areas was done on the average templates based on infants during the first year. We also applied a procedure for automatic computation of macroanatomical atlases for individual infant participants using two manually segmented adult atlases (Hammers, LONI Probabilistic Brain Atlas-LPBA40). To evaluate our methods, we did manual delineation of several cortical areas on selected individuals from each age. Linear and nonlinear registration of the individual and average template was used to transform the average atlas into the individual participant's space, and the average-transformed atlas was compared to the individual manually delineated brain areas. We also applied these methods to an external data set - not used in the atlas creation - to test generalizability of the atlases. Results: Age-appropriate manual atlases were the best fit to the individual manually delineated regions, with more error seen at greater age discrepancy. There was a close fit between the manually delineated and the automatically labeled regions for individual participants and for the age-appropriate template-based atlas transformed into participant space. There was close correspondence between automatic labeling of individual brain regions and those from the age-appropriate template. These relationships held even when tested on an external set of images. Conclusion: We have created age-appropriate labeled templates for use in the study of infant development at 6 ages (3, 4.5, 6, 7.5, 9, and 12 months). Comparison with manual methods was quite good. We developed three stereotaxic atlases (one manual, two automatic) for each infant age, which should allow more fine-grained analysis of brain structure for these populations than was previously possible with existing tools. The template-based atlases constructed in the current study are available online (http://jerlab.psych.sc.edu/NeurodevelopmentalMRIDatabase).

Developmental brain ADC atlas creation from clinical images

Yangming Ou1, Nathaniel Reynolds1, Randy L. Gollub1, Rudolph Pienaar3, Yanbing Wang1, T. David Wang1, Darren Sack1, Katherine P. Andriole4, Steven Pieper5, Christopher Herrick1, Shawn N. Murphy1, P. Ellen Grant3, Lilla Zollei1 1A. A. Martinos Biomedical Imaging Center, Massachusetts General Hospital, Charlestown, MA, 2Information Systems, Partners HealthCare System, Inc, Charlestown, MA, 3Children’s Hospital Boston, Boston, MA, 4Brigham Women’s Hospital, Boston, MA, 5Isomics, Inc.. Poster #3973

Development and evaluation of an articulated registration algorithm for human skeleton registration

Accurate registration over multiple scans is necessary to assess treatment response of bone diseases (e.g. metastatic bone lesions). This study aimed to develop and evaluate an articulated registration algorithm for the whole-body skeleton registration in human patients. In articulated registration, whole-body skeletons are registered by auto-segmenting into individual bones using atlas-based segmentation, and then rigidly aligning them. Sixteen patients (weight = 80–117 kg, height = 168–191 cm) with advanced prostate cancer underwent the pre- and mid-treatment PET/CT scans over a course of cancer therapy. Skeletons were extracted from the CT images by thresholding (HU>150). Skeletons were registered using the articulated, rigid, and deformable registration algorithms to account for position and postural variability between scans. The inter-observers agreement in the atlas creation, the agreement between the manually and atlas-based segmented bones, and the registration performances of all three registration algorithms were all assessed using the Dice similarity index—DSIobserved, DSIatlas, and DSIregister. Hausdorff distance (dHausdorff) of the registered skeletons was also used for registration evaluation. Nearly negligible inter-observers variability was found in the bone atlases creation as the DSIobserver was 96 ± 2%. Atlas-based and manual segmented bones were in excellent agreement with DSIatlas of 90 ± 3%. Articulated (DSIregsiter = 75 ± 2%, dHausdorff = 0.37 ± 0.08 cm) and deformable registration algorithms (DSIregister = 77 ± 3%, dHausdorff = 0.34 ± 0.08 cm) considerably outperformed the rigid registration algorithm (DSIregsiter = 59 ± 9%, dHausdorff = 0.69 ± 0.20 cm) in the skeleton registration as the rigid registration algorithm failed to capture the skeleton flexibility in the joints. Despite superior skeleton registration performance, deformable registration algorithm failed to preserve the local rigidity of bones as over 60% of the skeletons were deformed. Articulated registration is superior to rigid and deformable registrations by capturing global flexibility while preserving local rigidity inherent in skeleton registration. Therefore, articulated registration can be employed to accurately register the whole-body human skeletons, and it enables the treatment response assessment of various bone diseases.

Hierarchical Density-Based Clustering of White Matter Tracts in the Human Brain

Diffusion tensor magnetic resonance imaging (DTI) provides a promising way of estimating the neural fiber pathways in the human brain non-invasively via white matter tractography. However, it is difficult to analyze the vast number of resulting tracts quantitatively. Automatic tract clustering would be useful for the neuroscience community, as it can contribute to accurate neurosurgical planning, tract-based analysis, or white matter atlas creation. In this paper, the authors propose a new framework for automatic white matter tract clustering using a hierarchical density-based approach. A novel fiber similarity measure based on dynamic time warping allows for an effective and efficient evaluation of fiber similarity. A lower bounding technique is used to further speed up the computation. Then the algorithm OPTICS is applied, to sort the data into a reachability plot, visualizing the clustering structure of the data. Interactive and automatic clustering algorithms are finally introduced to obtain the clusters. Extensive experiments on synthetic data and real data demonstrate the effectiveness and efficiency of our fiber similarity measure and show that the hierarchical density-based clustering method can group these tracts into meaningful bundles on multiple scales as well as eliminating noisy fibers.

Atlas-based Segmentation in Prostate IMRT: Timesavings in the Clinical Workflow

Methods A commercially available package was used for atlas creation and automatic contour generation (mimvista Corp, Cleveland, Oh). An atlas was created from 98 previously contoured imRT patients treated at our institution. Bladder and urethral contrast was used in 38 patients. Ten of the most recent patients were randomly selected for analysis, five with contrast and five without. Prior to contour generation, the patient to be contoured was removed from the atlas. The atlas was stratified by bladder size ( 400mL), and only the subjects matching the patient’s estimated bladder size and contrast usage were considered. The most similar atlas subject was automatically selected and contours were automatically deformed onto the patient’s CT scan. The time to manually contour the patient by a resident (GK) and subsequently edit by an attending physician (AL) was recorded per contour for the prostate (CTV), bladder, rectum, and both femurs. (Figure 1 and 2) Contouring time was similarly recorded with the atlas contours used as a starting point. Contouring times were then compared and analyzed between the atlas and non-atlas contours.

Automatic Tractography Segmentation Using a High-Dimensional White Matter Atlas

We propose a new white matter atlas creation method that learns a model of the common white matter structures present in a group of subjects. We demonstrate that our atlas creation method, which is based on group spectral clustering of tractography, discovers structures corresponding to expected white matter anatomy such as the corpus callosum, uncinate fasciculus, cingulum bundles, arcuate fasciculus, and corona radiata. The white matter clusters are augmented with expert anatomical labels and stored in a new type of atlas that we call a high-dimensional white matter atlas. We then show how to perform automatic segmentation of tractography from novel subjects by extending the spectral clustering solution, stored in the atlas, using the Nystrom method. We present results regarding the stability of our method and parameter choices. Finally we give results from an atlas creation and automatic segmentation experiment. We demonstrate that our automatic tractography segmentation identifies corresponding white matter regions across hemispheres and across subjects, enabling group comparison of white matter anatomy.

Magnetic resonance angiography: From anatomical knowledge modeling to vessel segmentation

Magnetic resonance angiography (MRA) has become a common way to study cerebral vascular structures. Indeed, it enables to obtain information on flowing blood in a totally non-invasive and non-irradiant fashion. MRA exams are generally performed for three main applications: detection of vascular pathologies, neurosurgery planning, and vascular landmark detection for brain functional analysis. This large field of applications justifies the necessity to provide efficient vessel segmentation tools. Several methods have been proposed during the last fifteen years. However, the obtained results are still not fully satisfying. A solution to improve brain vessel segmentation from MRA data could consist in integrating high-level a priori knowledge in the segmentation process. A preliminary attempt to integrate such knowledge is proposed here. It is composed of two methods devoted to phase contrast MRA (PC MRA) data. The first method is a cerebral vascular atlas creation process, composed of three steps: knowledge extraction, registration, and data fusion. Knowledge extraction is performed using a vessel size determination algorithm based on skeletonization, while a topology preserving non-rigid registration method is used to fuse the information into the atlas. The second method is a segmentation process involving adaptive sets of gray-level hit-or-miss operators. It uses anatomical knowledge modeled by the cerebral vascular atlas to adapt the parameters of these operators (number, size, and orientation) to the searched vascular structures. These two methods have been tested by creating an atlas from a 18 MRA database, and by using it to segment 30 MRA images, comparing the results to those obtained from a region-growing segmentation method.