Horst K. Hahn

IWT-interactive watershed transform: a hierarchical method for efficient interactive and automated segmentation of multidimensional gray-scale images

In this paper we present the Interactive Watershed Transform (IWT) for efficient segmentation of multidimensional grayscale images. The IWT builds upon a fast immersion-based watershed transform (WT) followed by a hierarchical organization of the resulting basins in a tree structure. Each local image minimum is represented as an atomic basin at the lowest hierarchy level. The fast WT consists of two steps. First, all image elements are sorted according to their image intensity using a Bucket Sort algorithm. Second, each element is processed exactly once with respect to its neighborhood (e. g., 4, 6, and 8 direct neighbors for 2d, 3d, and 4d transform, respectively) in the specified order. Sort-ing, processing, and tree generation are of order O(n). After computing the WT, one global parameter, the so-called preflooding height, and an arbitrary number of markers are evaluated in real-time to control tree partitioning and basin merging. The IWT has been successfully applied to a large variety of medical images, e. g., for segmentation and volu-metry of neuroanatomic structures as well as bone segmentation, without making assumptions on the objects’ shapes. The IWT combines automation and efficient interactive control in a coherent algorithm while completely avoiding oversegmentation which is the major problem of the classical WT.

Lung lobe segmentation by anatomy-guided 3D watershed transform

Since the lobes are mostly independent anatomic compartments of the lungs, they play a major role in diagnosis and therapy of lung diseases. The exact localization of the lobe-separating fissures in CT images often represents a non-trivial task even for experts. Therefore, a lung lobe segmentation method suitable to work robustly under clinical conditions must take advantage of additional anatomic information. Due to the absence of larger blood vessels in the vicinity of the fissures, a distance transform performed on a previously generated vessel mask allows a reliable estimation of the boundaries even in cases where the fissures themselves are invisible. To make use of image regions with visible fissures, we linearly combine the original data with the distance map. The segmentation itself is performed on the combined image using an interactive 3D watershed algorithm which allows an iterative refinement of the results. The proposed method was successfully applied to CT scans of 24 patients. Preliminary intra- and inter-observer studies conducted for one of the datasets showed a volumetric variability of well below 1%. The achieved structural decomposition of the lungs not only assists in subsequent image processing steps but also allows a more accurate prediction of lobe-specific functional parameters.

Automatic Quantification of DTI Parameters Along Fiber Bundles

We introduce a novel technique that allows for an automatic quantification of MR DTI parameters along arbitrarily oriented fiber bundles. Most previous methods require either a manual placement of ROIs, are limited to single fiber tracts, or are limited to bundles which are perpendicular to one of the three image planes. Thus, the quantification process is made much more time-efficient and robust by our new approach. We compare our technique with a manual quantification of an expert and show the similarity of the results. Furthermore, we demonstrate how to visualize the parameters at a certain position of the fiber bundle so that areas of interest can easily be examined.

Clinical relevance of model based computer-assisted diagnosis and therapy

The ability to acquire and store radiological images digitally has made this data available to mathematical and scientific methods. With the step from subjective interpretation to reproducible measurements and knowledge, it is also possible to develop and apply models that give additional information which is not directly visible in the data. In this context, it is important to know the characteristics and limitations of each model. Four characteristics assure the clinical relevance of models for computer-assisted diagnosis and therapy: ability of patient individual adaptation, treatment of errors and uncertainty, dynamic behavior, and in-depth evaluation. We demonstrate the development and clinical application of a model in the context of liver surgery. Here, a model for intrahepatic vascular structures is combined with individual, but in the degree of vascular details limited anatomical information from radiological images. As a result, the model allows for a dedicated risk analysis and preoperative planning of oncologic resections as well as for living donor liver transplantations. The clinical relevance of the method was approved in several evaluation studies of our medical partners and more than 2900 complex surgical cases have been analyzed since 2002.

Multispectral brain tumor segmentation based on histogram model adaptation

Brain tumor segmentation and quantification from MR images is a challenging task. The boundary of a tumor and its volume are important parameters that can have direct impact on surgical treatment, radiation therapy, or on quantitative measurements of tumor regression rates. Although a wide range of different methods has already been proposed, a commonly accepted approach is not yet established. Today, the gold standard at many institutions still consists of a manual tumor outlining, which is potentially subjective, and a time consuming and tedious process. We propose a new method that allows for fast multispectral segmentation of brain tumors. An efficient initialization of the segmentation is obtained using a novel probabilistic intensity model, followed by an iterative refinement of the initial segmentation. A progressive region growing that combines probability and distance information provides a new, flexible tumor segmentation. In order to derive a robust model for brain tumors that can be easily applied to a new dataset, we retain information not on the anatomical, but on the global cross-subject intensity variability. Therefore, a set of multispectral histograms from different patient datasets is registered onto a reference histogram using global affine and non-rigid registration methods. The probability model is then generated from manual expert segmentations that are transferred to the histogram feature domain. A forward and backward transformation of a manual segmentation between histogram and image domain allows for a statistical analysis of the accuracy and robustness of the selected features. Experiments are carried out on patient datasets with different tumor shapes, sizes, locations, and internal texture.

Fractal Aspects of Three-Dimensional Vascular Constructive Optimization

We study various properties of constructive optimization in 3D vascular systems. After some remarks on existing approaches to vascular modeling and on the theory of vascular optimality, we briefly describe an algorithm called Global Constructive Optimization (GCO). Twenty-one vascular systems are modeled in three different groups: planar, spherical, and liver shaped. Based on the Strahler ordering scheme, these models are characterized and compared to data from liver corrosion casts. A good correspondence could be observed between modeled and real portal venous systems. The branching characteristics of the hepatic vein still pose open questions. Finally, a concept for the modeling of vascular interdigitation based on optimality principles is suggested.

White matter lesion phantom for diffusion tensor data and its application to the assessment of fiber tracking

For risk analysis prior to interventional treatment of brain tumors it is important to identify the functional brain areas affected by the tumor and to estimate their connectivity. Fiber Tracking (FT) on Diffusion Tensor (DT) data has the potential to facilitate this task. Our work is organized in two parts. First, we derive a relationship between diffusion anisotropy and orientation uncertainty of the DT by considering image noise. In order to assess a given FT algorithm with respect to the reconstruction of locally disturbed fiber bundles, this relationship is used for the simulation of white mat-ter lesions in DT data. Then, a deflection based FT algorithm is assessed with our software phantom. The FT algorithm is modified and its parameters are adjusted in order to obtain a fiber bundle reconstruction, which is robust to local fiber disturbance. Thus, it is demonstrated how to evaluate and improve FT algorithms with respect to the reconstruction of locally disturbed fiber bundles on the basis of phantom data with known ground truth. This is expected to improve functional and structural risk analysis for the interventional treatment of brain tumors.

Fast and Robust Quantification of Parahippocampal Atrophy via Temporal Horn Index

We propose a fast and robust method to obtain the temporal horn index (THI) as an indirect but sensitive regional measure for hippocampal and parahippocampal atrophy, based on MRI. The THI is defined as the temporal horn volume to lateral ventricular volume ratio. The proposed method relies on efficient 3D interactive segmentation and a fully automated histogram analysis. It provides consistent THI measurements within a few minutes even for extremely small temporal horns of less than 0.1 ml. The THI obtained by volumetric MRI analysis is sensitive to hippocampal and parahippocampal atrophy and is expected to provide an early marker for pathologic changes associated with Alzheimer’s and Parkinson’s disease.

Efficient fiber clustering using parameterized polynomials

In the past few years, fiber clustering algorithms have shown to be a very powerful tool for grouping white matter connections tracked in DTI images into anatomically meaningful bundles. They improve the visualization and perception, and could enable robust quantification and comparison between individuals. However, most existing techniques perform a coarse approximation of the fibers due to the high complexity of the underlying clustering problem or do not allow for an efficient clustering in real time. In this paper, we introduce new algorithms and data structures which overcome both problems. The fibers are represented very precisely and efficiently by parameterized polynomials defining the x-, y-, and z-component individually. A two-step clustering method determines possible clusters having a Gaussian distributed structure within one component and, afterwards, verifies their existences by principal component analysis (PCA) with respect to the other two components. As the PCA has to be performed only n times for a constant number of points, the clustering can be done in linear time O(n), where n denotes the number of fibers. This drastically improves on existing techniques, which have a high, quadratic running time, and it allows for an efficient whole brain fiber clustering. Furthermore, our new algorithms can easily be used for detecting corresponding clusters in different brains without time-consuming registration methods. We show a high reliability, robustness and efficiency of our new algorithms based on several artificial and real fiber sets that include different elements of fiber architecture such as fiber kissing, crossing and nested fiber bundles.

HWT - hybrid watershed transform: optimal combination of hierarchical interactive and automated image segmentation

In quantitative medical imaging and therapy planning, the optimal combination of automated and interactively defined information is crucial for image segmentation methods to be both efficient and effective. We propose to combine an efficient hierarchical region merging scheme that collects per-region statistics across hierarchy levels with a trainable classification engine that facilitates automated region labeling based on an arbitrary number of reference segmentations. When applying the classification engine, we propose to use a corridor of non-classified regions resulting in a sparse labeling with extremely low false-classification rate, and to attribute labels to the remaining basins through successive merging with ready-labeled basins. The proposed hierarchical region merging scheme also permits to efficiently include interactively defined labels. We denominate this general approach as Hybrid Hierarchical Interactive Image Segmentation Scheme (HIS2). More specifically, we present an extension of the Interactive Watershed Transform, which we combine with a trainable two-class Support Vector Machine based on Gaussian radial basis functions. Finally, we present a novel asymmetric marker scheme, which provides a powerful means of regionally correcting remaining inaccuracies while preserving full detail of the automatic labeling procedure. We denominate the complete algorithm as Hybrid Watershed Transform (HWT), which we apply to one challenging segmentation problem in clinical imaging, namely efficient bone removal in large computed tomography angiographic data sets. Efficiency and accuracy of the proposed methodology is evaluated on multi-slice images from nine different sites. As a result, its ability to rapidly and automatically generate robust and precise segmentation results in combination with a versatile manual correction mechanism could be proven without requiring specific anatomical or geometrical models.