Marko Rak

Closed-Form Hierarchical Finite Element Models for Part-Based Object Detection

In this work we address part-based object detection under variability of part shapes and spatial relations. Our approach bases on the hierarchical finite element modeling concept of Engel and Tönnies [ET09a, ET09b]. They model object parts by elastic materials, which adapt to image structures via image-derived forces. Spatial part relations are realized through additional layers of elastic material forming an elastic hierarchy. We present a closed-form solution to this concept, reformulating the hierarchical optimization problem into the optimization of a non-hierarchical finite element model. This allows us to apply standard finite element techniques to hierarchical problems and to provide an efficient framework for part-based object detection. We demonstrate our approach at the example of lumbar column detection in magnetic resonance imaging on a data set of 49 subjects. Given a rough model initialization, our approach solved the detection problem reliably in 45 out of 49 cases, showing computation times of only a few seconds per subject.

Visualization and Analysis of Lumbar Spine Canal Variability in Cohort Study Data

AbstractLarge-scale longitudinal epidemiological studies, such as the Study of Health in Pomerania (SHIP), investigatethousands of individuals with common characteristics or experiences (a cohort) including a multitude of socio-demographic and biological factors. Unique for SHIP is the inclusion of medical image data acquired via anextensive whole-body MRI protocol. Based on this data, we study the variability of the lumbar spine and itsrelation to a subset of socio-demographic and biological factors. We focus on the shape of the lumbar spinal canalwhich plays a crucial role in understanding the causes of lower back pain.We propose an approach for the reproducible analysis of lumbar spine canal variability in a cohort. It is basedon the centerline of each individual canal, which is derived from a semi-automatic, model-based detection of thelumbar spine. The centerlines are clustered by means of Agglomerative Hierarchical Clustering to form groupswith low intra-group and high inter-group shape variability. The number of clusters is computed automatically.The clusters are visualized by means of representatives to reduce visual clutter and simplify a comparison betweensubgroups of the cohort. Special care is taken to convey the shape of the spinal canal also orthogonal to the viewplane. We demonstrate our approach for 490 individuals drawn from the SHIP data. We present preliminary resultsof investigating the clusters with respect to their associated socio-demographic and biological factors.Categories and Subject Descriptors

Deformable part models for object detection in medical images

BackgroundObject detection in 3-D medical images is often necessary for constraining a segmentation or registration task. It may be a task in its own right as well, when instances of a structure, e.g. the lymph nodes, are searched. Problems from occlusion, illumination and projection do not arise, making the problem simpler than object detection in photographies. However, objects of interest are often not well contrasted against the background. Influence from noise and other artifacts is much stronger and shape and appearance may vary substantially within a class.MethodsDeformable models capture the characteristic shape of an anatomic object and use constrained deformation for hypothesing object boundaries in image regions of low or non-existing contrast. Learning these constraints requires a large sample data base. We show that training may be replaced by readily available user knowledge defining a prototypical deformable part model. If structures have a strong part-relationship, or if they may be found based on spatially related guiding structures, or if the deformation is rather restricted, the supporting data information suffices for solving the detection task. We use a finite element model to represent anatomic variation by elastic deformation. Complex shape variation may be represented by a hierarchical model with simpler part variation. The hierarchy may be represented explicitly as a hierarchy of sub-shapes, or implicitly by a single integrated model. Data support and model deformation of the complete model can be represented by an energy term, serving as quality-of-fit function for object detection.ResultsThe model was applied to detection and segmentation tasks in various medical applications in 2- and 3-D scenes. It has been shown that model fitting and object detection can be carried out efficiently by a combination of a local and global search strategy using models that are parameterized for the different tasks.ConclusionsA part-based elastic model represents complex within-class object variation without training. The hierarchy of parts may specify relationship to neighboring anatomical objects in object detection or a part-decomposition of a complex anatomic structure. The intuitive way to incorporate domain knowledge has a high potential to serve as easily adaptable method to a wide range of different detection tasks in medical image analysis.

Image analysis in epidemiological applications

Abstract A rising number of epidemiological studies apply imaging technologies. Images are not features themselves but provide raw data from which features are extracted. Different to other applications of analysis of medical images the data is analyzed statistically across the cohort. It results in unique requirements regarding the development of methods to efficiently integrate varying domain knowledge into the process without compromising comparability of results across subjects or time. Examples from two different strategies are presented and discussed.

Interactive Visual Analysis of Lumbar Back Pain - What the Lumbar Spine Tells About Your Life.

Epidemiology aims to provide insight into disease causations. Hence, subject groups (cohorts) are analyzed to correlate the subjects’ varying lifestyles, their medical properties and diseases. Recently, these cohort studies comprise medical image data. We assess potential relations between image-derived variables of the lumbar spine with lower back pain in a cross-sectional study. Therefore, an Interactive Visual Analysis (IVA) framework was created and tested with 2,540 segmented lumbar spine data sets. The segmentation results are evaluated and quantified by employing shape-describing variables, such as spine canal curvature and torsion. We analyze mutual dependencies among shape-describing variables and non-image variables, e.g., pain indicators. Therefore, we automatically train a decision tree classifier for each non-image variable. We provide an IVA technique to compare classifiers with a decision tree quality plot. As a first result, we conclude that image-based variables are only sufficient to describe lifestyle factors within the data. A correlation between lumbar spine shape and lower back pain could not be found with the automatically trained classifiers. However, the presented approach is a valuable extension for the IVA of epidemiological data. Hence, relations between non-image variables were successfully detected and described.

Measurement of the Aortic Diameter in Plain Axial Cardiac Cine MRI

We address the task of aortic diameter measurement in (noncontrast- enhanced) plain axial cardiac cine MRI. To this end, we set up a likelihood maximization problem which allows us to recover globally optimal aorta locations and diameters of the cine sequence efficiently. Our approach provides intuitive means of manual post-correction and requires little user interaction, making large-scale image analysis feasible. Experiments on a data set of 20 cine sequences with 30 time frames showed (at least) pixel-accurate diameter measurements which are also highly stable against re-parameterization.

Texture-Based Detection of Myositis in Ultrasonographies

Muscle ultrasonography is a convenient technique to visualize healthy and pathological muscle tissue as it is non-invasive and image acquisition can be done in real-time. In this paper, a texture-based approach is presented to detect myositis in ultrasound images automatically. We compute different texture features like wavelet transform features and first-order grey-level intensity statistics of a relevant central image patch carrying structure and intensity information of muscle tissue. Using a combination of these information we reached an accuracy of classification of 92.20 % with our approach on a training data set of 63 clinically pre-classified data sets.

Star Convex Cuts with Encoding Swaps for Fast Whole-Spine Vertebrae Segmentation in MRI

We propose an automatic approach for fast vertebral body segmentation in three-dimensional magnetic resonance images of the whole spine. Previous works are limited to the lower thoracolumbar section and often take minutes to compute, which can be problematic in clinical routine or for data sets with numerous subjects. We address these limitations by a graph cut formulation. Our formulation involves appearance and shape information as well as star-convexity constraints to ensure a topologically correct segmentation for each vertebra. For close targets such as adjacent vertebrae, implementing star-convexity without fusing targets (naive binary formulations) or increasing run time/loosing optimality guarantees (multi-label formulations) is challenging. We provide a solution based on encoding swaps, which preserve optimality and ensure topological correctness between vertebrae. We validated our approach on two data sets. The first contains T1and T2-weighted whole-spine images of 64 subjects. The second comprises 23 T2-weighted thoracolumbar images and is publicly available. Our results are competitive to previous works (or better) at a fraction of the run time. We yielded Dice coefficients of 85.1± 4.4 % and 89.7± 2.3 % with run times of 1.65±0.28 s and 2.73±0.36 s per vertebra on consumer hardware.

Extracting the Aorta Centerline in Contrast-Enhanced MRI

We propose a semi-automatic approach for aorta centerline extraction in contrast-enhanced MRI, making aorta length analysis feasible on large scale. Starting from user-specified start and end regions, we extract the aorta path in between the regions automatically. The extraction is formulated as an optimization problem, seeking for the path that most likely runs central to the aorta. To this end, we exploit that the aorta distinguishes from the surrounding by strong image gradients that point inwards to the aorta’s center due to contrast-enhanced imaging. We also include additional means of manual guidance to resolveerroneous cases. Experiments on data of 19 subjects yielded results that are close to the inter-reader variability. The average distance to the ground truth was 1.89 ± 1.54 mm, while aorta lengths deviated by only 0.66 ± 0.49 %.

A Learning-Free Approach to Whole Spine Vertebra Localization in MRI

In recent years, analysis of magnetic resonance images of the spine gained considerable interest with vertebra localization being a key step for higher level analysis. Approaches based on trained appearance - which are de facto standard - may be inappropriate for certain tasks, because processing usually takes several minutes or training data is unavailable. Learning-free approaches have yet to show there competitiveness for whole-spine localization. Our work fills this gap. We combine a fast engineered detector with a novel vertebrae appearance similarity concept. The latter can compete with trained appearance, which we show on a data set of 64 \(T_1\)- and 64 \(T_2\)-weighted images. Our detection took \(27.7 \pm 3.78\) s with a detection rate of 96.0 % and a distance to ground truth of \(3.45 \pm 2.2\) mm, which is well below the slice thickness.