Xiaoqian Xu

A Spine X-Ray Image Retrieval System Using Partial Shape Matching

In recent years, there has been a rapid increase in the size and number of medical image collections. Thus, the development of appropriate methods for medical information retrieval is especially important. In a large collection of spine X-ray images, maintained by the National Library of Medicine, vertebral boundary shape has been determined to be relevant to pathology of interest. This paper presents an innovative partial shape matching (PSM) technique using dynamic programming (DP) for the retrieval of spine X-ray images. The improved version of this technique called corner-guided DP is introduced. It uses nine landmark boundary points for DP search and improves matching speed by approximately 10 times compared to traditional DP. The retrieval accuracy and processing speed of the retrieval system based on the new corner-guided PSM method are evaluated and included in this paper.

Contour matching for a fish recognition and migration-monitoring system

Fish migration is being monitored year round to provide valuable information for the study of behavioral responses of fish to environmental variations. However, currently all monitoring is done by human observers. An automatic fish recognition and migration monitoring system is more efficient and can provide more accurate data. Such a system includes automatic fish image acquisition, contour extraction, fish categorization, and data storage. Shape is a very important characteristic and shape analysis and shape matching are studied for fish recognition. Previous work focused on finding critical landmark points on fish shape using curvature function analysis. Fish recognition based on landmark points has shown satisfying results. However, the main difficulty of this approach is that landmark points sometimes cannot be located very accurately. Whole shape matching is used for fish recognition in this paper. Several shape descriptors, such as Fourier descriptors, polygon approximation and line segments, are tested. A power cepstrum technique has been developed in order to improve the categorization speed using contours represented in tangent space with normalized length. Design and integration including image acquisition, contour extraction and fish categorization are discussed in this paper. Fish categorization results based on shape analysis and shape matching are also included.

An automated fish species classification and migration monitoring system

The quantification of abundance, distribution, and movement of fish is critical to ecological and environmental studies of fish communities. To properly manage, regulate, and protect migratory fisheries it is essential to accurately monitor numbers, size, and species of fish at specific fish passages during migratory seasons. Currently, all monitoring is done manually with significant time and financial constraints. An automated fish classification system will simplify data gathering and improve data accuracy. In this research, 22 images of 9 target species were recorded. The contour of each image was extracted to form a closed curve for shape analysis. A new shape analysis algorithm was developed for removing edge noise and redundant data points such as short straight lines. A curvature function analysis was used to locate critical landmark points. The fish contour segments of interest were then extracted based on these landmark points for species classification. By comparing individual contour segments to the curves in the database, accurate pattern matching was achieved.

Using relevance feedback with short-term memory for content-based spine X-ray image retrieval

Managing large medical image databases has become a challenging task as more medical images are produced and stored in digital format. Computer-aided decision support for content-based image retrieval (CBIR) is an essential tool for medical image management. This paper presents a novel hybrid relevance feedback (RF) system for shape-based retrieval of spine X-ray images. A new shape similarity measure that considers both whole shape and partial shape matching is presented. The proposed RF architecture includes separate retrieval and feedback modes to solicit users opinion for refining retrieval results. A unique short-term memory approach is implemented to avoid repeated request for users feedback on the same, already approved, and retrieved relevant images. An automatic weight updating scheme is developed to present the images on which it is best for the user to provide feedback. Incorporating all these unique features, the proposed RF retrieval system is able to reduce the gap between high-level human visual perception and low-level computerized features. Experimental results show overall retrieval accuracy improvement of 22.0% and 17.5% after the second feedback iteration for retrieving spine X-ray images with similar osteophytes severity and type, respectively.

Partial shape matching for CBIR of spine x-ray images

Efficient content-based image retrieval (CBIR) of biomedical images is a challenging problem. Feature representation algorithms used in indexing medical images on the pathology of interest have to address conflicting goals of reducing feature dimensionality while retaining important and often subtle biomedical features. In case of the vertebra, its shape effectively describes various pathologies identified by medical experts as being consistently and reliably found in x-rays in the image collection. A suitable shape method must enable retrieval relevant to the pathology in question. An approach to enabling pathology based retrieval is to use partial shape matching techniques. This paper describes our research in the development of such methods and initial retrieval results and related issues. The research is a part of our ongoing effort in developing CBIR for digitized images of a collection of 17,000 cervical and lumbar spine x-rays taken as a part of the second National Health and Nutrition Examination Survey (NHANES II) at the Lister Hill National Center for Biomedical Communications, an intramural R&D division of the U.S. National Library of Medicine.

Localizing contour points for indexing an X-ray image retrieval system

Vertebra shape can effectively describe various pathologies found in spine X-ray images. There are some critical regions on the shape contour which help determine whether the shape is pathologic or normal. We selected a subset of 250 segmented vertebra boundaries for study from a collection of 17,000 digitized X-rays of cervical and lumbar spine taken as a part of the second National Health and Nutrition Examination Survey (NHANES II). A board certified expert radiologist marked nine morphometric landmark points on the contour of these cervical and lumbar images. Image indexing could mimic the model used by the radiologists to mark the images, e.g. 6-, 9-, or 10-point, thereby improve the query and retrieval of vertebra shapes from the image database. In this paper, we present a technique to automatically select nine points from the boundary contour. The comparison between two 9-point models using the L/sub 2/ distance and retrieval rank results derived respectively from the 9-point model marked by the expert and the 9-point model selected with our algorithm provides a good measure of how well the two models match.

Area and volume measurements of objects with irregular shapes using multiple silhouettes

Surface area and volume measurements provide important information for agriculture and food-processing applications. A machine vision system that uses a nondestructive method to measure volume and surface area of objects with irregular shapes is presented in this paper. The system first takes a series of silhouettes of the object from different directions by rotating the object at a fixed angular interval. The boundary points of each image are then extracted to construct a silhouette. A three-dimensional wire-frame model of the object can be reconstructed by integrating silhouettes obtained from different view angles. Surface area and volume can then be measured by means of surface fitting and approximation on the wire-frame model. System calibration and surface approximation were two major challenges for the design of this machine vision system. A unique centerline calibration method is introduced in this paper. Surface approximation and calculation are also discussed. Examples of applications in agriculture and food processing using this vision system for surface area measurement are included, and its accuracy is verified.

Relevance feedback for spine X-ray retrieval

Relevance feedback (RF) has been an active research area in content-based image retrieval (CBIR). RF intends to bridge the gap between the low-level image features and the high-level human visual perception by analyzing and employing the feedback information provided by the user. This gap becomes more evident and important in medical image retrieval due to the two distinct facts with regard to medical images: (1) subtle differences between images, even between pathological and non-pathological images; (2) subjective and different diagnosis even among experts. This paper describes a novel linear weight-updating approach for RF applying to spine X-ray image retrieval. The algorithm utilizes both positive and negative examples to gain feedback from the user. Experimental results show that the proposed approach can substantially improve the retrieval performance to better satisfy the individual users preferences.

Partial shape matching of spine X-ray shapes using dynamic programming

The osteophyte shows only on some particular locations on the vertebra. This indicates that other locations on the vertebra shape contain that are not of interest hinder the spine X-ray image retrieval relevance precision, which motivates our research in partial shape matching (PSM). This paper presents PSM methods for matching shapes with variable number of points and with different data point distributions. Dynamic programming (DP) is proposed for matching partial shapes by allowing merging of the data points in the process of PSM. DP is implemented based on two shape representation methods: line segments and multiple open triangles. The performance evaluation, which is based on human relevance judgements of these two shape representations in PSM is also presented.

Using distance transform to solve real-time machine vision inspection problems

This paper describes novel solutions to two challenging real-time inspection tasks in machine vision. The first is fast surface approximation for volume and surface area measurements of irregularly shaped objects; the second is fast intensity gradient correction for surface inspection and evaluation of spherical objects. Both solutions apply a distance transform (DT) based on the distance of each image pixel from the object boundary. We describe both real-time machine vision inspection tasks and discuss their complexity. We show that the new solutions result in significant improvements in both accuracy and efficiency—despite the relative simplicity of the DT approach.