Shiri Gordon

Unsupervised image-set clustering using an information theoretic framework

In this paper, we combine discrete and continuous image models with information-theoretic-based criteria for unsupervised hierarchical image-set clustering. The continuous image modeling is based on mixture of Gaussian densities. The unsupervised image-set clustering is based on a generalized version of a recently introduced information-theoretic principle, the information bottleneck principle. Images are clustered such that the mutual information between the clusters and the image content is maximally preserved. Experimental results demonstrate the performance of the proposed framework for image clustering on a large image set. Information theoretic tools are used to evaluate cluster quality. Particular emphasis is placed on the application of the clustering for efficient image search and retrieval.

Unsupervised Image Clustering Using the Information Bottleneck Method

A new method for unsupervised image category clustering is presented, based on a continuous version of a recently introduced information theoretic principle, the information bottleneck (IB). The clustering method is based on hierarchical grouping: Utilizing a Gaussian mixture model, each image in a given archive is first represented as a set of coherent regions in a selected feature space. Images are next grouped such that the mutual information between the clusters and the image content is maximally preserved. The appropriate number of clusters can be determined directly from the IB principle. Experimental results demonstrate the performance of the proposed clustering method on a real image database.

Image segmentation of uterine cervix images for indexing in PACS

The National Cancer Institute has collected a large database of digitized 35 mm slides of the uterine cervix, the idea being to build a system enabling to study the evolution of lesions related to cervical cancer. In taking the first few steps towards this goal, the objective of this work is to develop and evaluate methodologies required for visual-based (i.e. content-based) indexing and retrieval that substantially improve information management of such a database. In this paper we model the properties of three tissue types using color and texture features, and use these models for image segmentation. Statistical modeling and segmentation tools are used for the task.

Automatic Detection of Anatomical Landmarks in Uterine Cervix Images

The work focuses on a unique medical repository of digital cervicographic images (ldquoCervigramsrdquo) collected by the National Cancer Institute (NCI) in longitudinal multiyear studies. NCI, together with the National Library of Medicine (NLM), is developing a unique Web-accessible database of the digitized cervix images to study the evolution of lesions related to cervical cancer. Tools are needed for automated analysis of the cervigram content to support cancer research. We present a multistage scheme for segmenting and labeling regions of anatomical interest within the cervigrams. In particular, we focus on the extraction of the cervix region and fine detection of the cervix boundary; specular reflection is eliminated as an important preprocessing step; in addition, the entrance to the endocervical canal (the ldquoosrdquo), is detected. Segmentation results are evaluated on three image sets of cervigrams that were manually labeled by NCI experts.

Evaluation of uterine cervix segmentations using ground truth from multiple experts

This work is focused on the generation and utilization of a reliable ground truth (GT) segmentation for a large medical repository of digital cervicographic images (cervigrams) collected by the National Cancer Institute (NCI). NCI invited twenty experts to manually segment a set of 939 cervigrams into regions of medical and anatomical interest. Based on this unique data, the objectives of the current work are to: (1) Automatically generate a multi-expert GT segmentation map; (2) Use the GT map to automatically assess the complexity of a given segmentation task; (3) Use the GT map to evaluate the performance of an automated segmentation algorithm. The multi-expert GT map is generated via the STAPLE (Simultaneous Truth and Performance Level Estimation) algorithm, which is a well-known method to generate a GT segmentation from multiple observations. A new measure of segmentation complexity, which relies on the inter-observer variability within the GT map, is defined. This measure is used to identify images that were found difficult to segment by the experts and to compare the complexity of different segmentation tasks. An accuracy measure, which evaluates the performance of automated segmentation algorithms is presented. Two algorithms for cervix boundary detection are compared using the proposed accuracy measure. The measure is shown to reflect the actual segmentation quality achieved by the algorithms. The methods and conclusions presented in this work are general and can be applied to different images and segmentation tasks. Here they are applied to the cervigram database including a thorough analysis of the available data.

Content-based indexing and retrieval of uterine cervix images

This work is motivated by the need for visual information management in the growing field of digital libraries and by the increasing information retrieval demands in the domains of medical imaging and telemedicine. We focus on a large database of digitized 35 mm slides of the uterine cervix collected by the National Cancer Institute (NCI), National Institutes of Health (NIH), to study the evolution of lesions related to cervical cancer. As a first step towards this goal, we focus on the problem of intelligently labeling (segmenting) regions of medical interest within the cervigram image. We use statistical tools for the segmentation of three tissue types of interest.

Probabilistic models for generating, modelling and matching image categories

In this paper we present a probabilistic and continuous framework for supervised image category modelling and matching as well as unsupervised clustering of image space into image categories. A generalized GMM-KL framework is described in which each image or image-set (category) is represented as a Gaussian mixture distribution and images (categories) are compared and matched via a probabilistic measure of similarity between distributions. Image-to-category matching is investigated and unsupervised clustering of a random image set into visually coherent image categories is demonstrated.

Co-segmentation of multiple images into multiple regions: Application to mouse brain MRI

Challenging biomedical segmentation problems can be addressed by combining top-down information based on the known anatomy along with bottom-up models of the image data. Anatomical priors can be provided by probabilistic atlases. Nevertheless, in many cases the available atlases are inadequate. We present a novel method for the co-segmentation of multiple images into multiple regions, where only a very few annotated examples exist. The underlying, unknown anatomy is learned throughout an interleaved process, in which the segmentation of a region is supported both by the segmentation of the neighboring regions which share common boundaries and by the segmentation of corresponding regions in the other jointly segmented images. The method is applied to a mouse brain MRI dataset for the segmentation of five anatomical structures. Experimental results demonstrate the segmentation accuracy with respect to the data complexity.

Multidimensional co-segmentation of longitudinal brain MRI ensembles in the presence of a neurodegenerative process

Abstract MRI Segmentation of a pathological brain poses a significant challenge, as the available anatomical priors that provide top‐down information to aid segmentation are inadequate in the presence of abnormalities. This problem is further complicated for longitudinal data capturing impaired brain development or neurodegenerative conditions, since the dynamic of brain atrophies has to be considered as well. For these cases, the absence of compatible annotated training examples renders the commonly used multi‐atlas or machine‐learning approaches impractical. We present a novel segmentation approach that accounts for the lack of labeled data via multi‐region multi‐subject co‐segmentation (MMCoSeg) of longitudinal MRI sequences. The underlying, unknown anatomy is learned throughout an iterative process, in which the segmentation of a region is supported both by the segmentation of the neighboring regions, which share common boundaries, and by the segmentation of corresponding regions, in the other jointly segmented images. A 4D multi‐region atlas that models the spatio‐temporal deformations and can be adapted to different subjects undergoing similar degeneration processes is reconstructed concurrently. An inducible mouse model of p25 accumulation (the CK‐p25 mouse) that displays key pathological hallmarks of Alzheimer disease (AD) is used as a gold‐standard to test the proposed algorithm by providing a conditional control of rapid neurodegeneration. Applying the MMCoSeg to a cohort of CK‐p25 mice and littermate controls yields promising segmentation results that demonstrate high compatibility with expertise manual annotations. An extensive comparative analysis with respect to current well‐established, atlas‐based segmentation methods highlights the advantage of the proposed approach, which provides accurate segmentation of longitudinal brain MRIs in pathological conditions, where only very few annotated examples are available. Graphical abstract Figure. No Caption available.

Atlas of Classifiers for Brain MRI Segmentation

We present a conceptually novel framework for brain tissue segmentation based on an Atlas of Classifiers (AoC). The AoC allows a statistical summary of the annotated datasets taking into account both the imaging data and the corresponding labels. It is therefore more informative than the classical probabilistic atlas and more economical than the popular multi-atlas approaches, which require large memory consumption and high computational complexity for each segmentation. Specifically, we consider an AoC as a spatial map of voxel-wise multinomial logistic regression (LR) functions learned from the labeled data. Upon convergence, the resulting fixed LR weights (a few for each voxel) represent the training dataset, which might be huge. Segmentation of a new image is therefore immediate and only requires the calculation of the LR outputs based on the respective voxel-wise features. Moreover, the AoC construction is independent of the test images, providing the flexibility to train it on the available labeled data and use it for the segmentation of images from different datasets and modalities.