We propose a method for extracting the left ventricular (LV) contours from left ventriculograms by means of a neural edge detector (NED) in order to extract the contours which accord with those traced by a cardiologist. The NED is a supervised edge detector based on a modified multilayer neural network, and is trained by use of a modified back-propagation algorithm. The NED can acquire the function of a desired edge detector through training with a set of input images and the desired edges obtained from the contours traced by a cardiologist. The proposed contour-extraction method consists of 1) detection of "subjective edges" by use of the NED; 2) extraction of rough contours by use of low-pass filtering and edge enhancement; and 3) a contour-tracing method based on the contour candidates synthesized from the edges detected by the NED and the rough contours. Through experiments, it was shown that the proposed method was able to extract the contours in agreement with those traced by an experienced cardiologist, i.e., we achieved an average contour error of 6.2% for left ventriculograms at end-diastole and an average difference between the ejection fractions obtained from the manually traced contours and those obtained from the computer-extracted contours of 4.1%.

Extraction of left ventricular contours from left ventriculograms by means of a neural edge detector

Today, iris recognition, fingerprint recognition, face recognition, voice recognition and other biometric technology are experiencing rapid development. This paper addresses a new biometric technology - skeleton recognition for blood vessel of optic fundus. The green grayscale ocular fundus image is utilized. The skeleton feature of blood vessel of optic fundus is extracted at first. After filtering treatment and extracting feature, vector curve of blood vessels is obtained. Shape curve matching is later carried out by means of reference point matching. The recognition for blood vessel optic fundus has been demonstrated in this paper to possess high recognition rate, low recognition rejection rate as well as good universality, exclusiveness and stability. With more and more progress made in extracting technology, the optic fundus blood vessel recognition is to become an effective biometric technology.

The blood vessel recognition of ocular fundus

New approach to automatic contour detection from image sequences: an application to ventriculographic images

Detection of the left ventricular (LV) endocardial (inner) and epicardial (outer) boundaries in cardiac images, provided by fast computer tomography (cine CT), magnetic resonance (MR), or ultrasound (echocardiography), is addressed. The automatic detection of the LV boundaries is difficult due to background noise, poor contrast, and often unclear differentiation of the tissue characteristics of the ventricles, papillary muscles, and surrounding tissues. An approach to the automatic ventricular boundary detection that employs set-theoretic techniques, and is based on incorporating a priori knowledge of the heart geometry, its brightness, spatial structure, and temporal dynamics into the boundaries detection algorithm is presented. Available knowledge is interpreted as constraint sets in the functional space, and the consistent boundaries are considered to belong to the intersection of all the introduced sets, thus satisfying the a priori information. An algorithm is also suggested for the simultaneous detection of the endocardial and epicardial boundaries of the LV. The procedure is demonstrated using cine CT images of the human heart. >

Constrained detection of left ventricular boundaries from cine CT images of human hearts

Today, iris recognition, fingerprint recognition, face recognition, voice recognition and other biometric technology are experiencing rapid development. This paper addresses a new biometric technology-the identification and recognition based on point of blood vessel skeleton for ocular fundus. The image for green gray scale of ocular fundus is utilized. The cross point of skeleton shape of blood vessel for ocular fundus using contrast-limited adaptive histogram equalization is extracted at first. After filtering treatment and extracting shape, shape curve of blood vessels is obtained. The cross point of shape for curve matching is later carried out by means of cross point matching. The recognition based on shape for blood vessel of ocular fundus has been demonstrated in this paper to possess high Identification and recognition rate, low rejection recognition rate as well as good universality, exclusiveness and stability. With more and more progress made in extracting technology, the recognition for blood vessel of optic fundus is to become an effective biometric technology.

The identification and recognition based on point for blood vessel of ocular fundus

In this paper we present an algorithm for enhancement of angiogra hic images using morphological opening filter. Opening ora gray-scale image by a gray-scale structuring element(SE) can generate a background image. To remove the background image , we can subtract it from the original image. Usually, there are a large number of computations involved in gray-scale opening operation. We present a skipped SE algorithm which can significantly improve the processing speed, and yet gives excellent enliancement for the angiographic images.

http://ieeexplore.ieee.org/iel4/1015/7498/00683908.pdf

A Fast Morphological Filter For Enhancement Of Angiographic Images

A dynamic spider model is presented to automatically extract the left ventricular border in a sequence of cineangiograms. The spider legs correspond to the medical involution chords of a confined region which resembles the shape of the left ventricle and encompasses its deformations in expansion and contraction of the heart. The recognition task is to find a congruence transformation of the spider by adaptively minimizing a deformity measure. The left ventricular border is then determined via the dynamic programming within the confined region of the registered spider. The method has been tested on cineangiograms of 30 subjects with reasonable success. >

Automated left ventricular border determination in cineangiograms based on a dynamic spider model

A cost effective solution for a cineangiogram archive is proposed based on a hybrid method of lossless image compression. Compared with several other approaches, the authors solution achieves the highest average compression ratio of 3.9:1, and yet the shortest decoding time. Decompression contains two steps: (1) from a compressed cineangiogram file on a disk to half decoded data in computer memory; (2) from the data in memory to fully decoded images in a display frame buffer. For an image sequence of 512/spl times/512 resolution, step-1 takes less than 200 ms/frame, and step-2 less than 28 ms/frame on a Sparc-10 workstation. Thus after step-1 loading, cine looping with either DSA or image mode can reach to 25 fps, suitable for clinical applications. >

http://ieeexplore.ieee.org/iel2/3299/9921/00470198.pdf

A cost effective solution for digital cardiac angiogram archiving using lossless image compression

Dynamic spider models were used for automated left ventricular border detection in biplane angiograms. The model consists of a set of medial involution chords of a confined region which encompasses the contraction area of a left ventricular silhouette. To identify a left ventricular silhouette is to find a congruence transformation so that a deformity measure between the mapped model and an image is minimized. The left ventricular border can then be determined within the aligned confined region of the model. The current algorithm is simpler, faster, and more robust than previously reported solutions. Clinical evaluation in 50 patients has shown distinct improvements. The accuracy rate in model registration was 93% versus 80%, and in border determination 74% versus 70%. >

Improved automatic left ventricular border detection in cineangiograms using dynamic spider models

To improve the computational speed of the rolling ball image enhancement algorithm, the authors introduce a new skipped structuring element (SE) and present an algorithm to decompose this and other kinds of skipped structuring elements. To evaluate the performance of the enhancement algorithm with different SEs, they designed a set of synthetic test images which mimic the real angiogram blood vessels and background. A performance index for image quality measurement of the enhanced coronary angiograms has been applied to all test images. The experimental results led to the conclusion that the new approach, which preserves most of the enhancement capability of using the complete structuring element, improves the computational speed by at least one order of magnitude. >

N. Fan

Papers

A Fast Morphological Filter For Enhancement Of Angiographic Images

Automated left ventricular border determination in cineangiograms based on a dynamic spider model

A cost effective solution for digital cardiac angiogram archiving using lossless image compression

Improved automatic left ventricular border detection in cineangiograms using dynamic spider models

Morphological enhancement of coronary angiograms using decomposition of skipped structuring element