CT scans are thin cross-sectional, radiographic images that can be obtained at any body level. CT images can describe the soft tissues with better clarity because it is more sensitive to slight differences in attenuation than standard radiography. Image segmentation is the key process to identify body fat in CT images. CT images at different body levels have different structures and hence different grayness histogram. Furthermore, the grayness histogram itself, in one CT image, has multiple peaks. Therefore, three segmentation methods, automatic threshold segmentation, morphological reconstruction segmentation, and potential function clustering segmentation, are used in this paper. Body fat contents and distributions are got according to segmented CT images. Experiment results show the effectiveness and stability of the multi-thresholds image segmentation method based on potential function clustering.
In this paper a novel method for the chromatic analysis of burn scar is proposed. The aim of the algorithm is to evaluate the curative effect and set up the treatment plan pertinently, because the scar color is an impersonal parameter reflects the degree of scar hypertrophy. The method is based on artificial neural network (ANN) by using photoelectrical technique, and composed of three main parts: firstly capture the digital color images of the burn scar using CCD camera, then change the RGB color data of the burn scar into that of HSB color space and emend it using ANN, lastly judge the degree of burn scar hypertrophy by chromatic analysis using ANN again. The experimental results were good conformed to the degrees of scar hypertrophy given by clinical evaluations. It suggests that the chromatic analysis technique of the burn scar is valuable for further study and apply to the clinical engineering.
Support vector machine (SVM) is a new statistical learning method. Compared with the classical machine learning methods, SVM learning discipline is to minimize the structural risk instead of the empirical risk of the classical methods, and it gives better generative performance. Because SVM algorithm is a convex quadratic optimization problem, the local optimal solution is certainly the global optimal one. In this paper a SVM algorithm is applied to detect the micro-calcifications (MCCs) in mammograms for the diagnostics of breast cancer that has not been reported yet. It had been tested with 10 mammograms and the results show that the algorithm can achieve a higher true positive in comparison with artificial neural network (ANN) based on the empirical risk minimization, and is valuable for further study and application in the clinical engineering.
In this paper, a new method of body fat and its distribution testing is proposed based on CT image processing. As it is more sensitive to slight differences in attenuation than standard radiography, CT depicts the soft tissues with better clarity. And body fat has a distinct grayness range compared with its neighboring tissues in a CT image. An effective multi-thresholds image segmentation method based on potential function clustering is used to deal with multiple peaks in the grayness histogram of a CT image. The CT images of abdomens of 14 volunteers with different fatness are processed with the proposed method. Not only can the result of total fat area be got, but also the differentiation of subcutaneous fat from intra-abdominal fat has been identified. The results show the adaptability and stability of the proposed method, which will be a useful tool for diagnosing obesity.
Clustered microcalcifications (MCCs) in mammograms are an important sign in the detection of breast cancer. Nevertheless, it is a complex and difficult task for radiologists to detect the clustered MCCs from the tissue background of mammograms only by naked eyes. This paper presents a prototype of a computer-aided detection system to automatically detect MCCs in mammograms. The detection algorithm mainly comprises three modules. The first module, called the mammogram pre-progressing module, inputs and digitizes mammograms into 8-bit images of size 2048x2048, normalizes the images, manually extracts the breast region from the background. The second module, called the feature extraction module, is achieved by using mixed features consisting of two wavelet features and two gray level statistical features. The wavelet features are generated by a five-level wavelet decomposition and reconstruction algorithm. The gray level statistical features used in this paper are median contrast and normalized gray level value. Finally, the third module, called the MCCs detection module, discovers MCCs in the images by using a classifier. This paper uses a three-layer artificial neural network (ANN) as a classifier to segment MCCs from the processing image. The ANN takes these four features generated in the second module as inputs. The output of the ANN corresponding to the true MCC pixels is then thresholded to segment out the true MCC pixels. One advantage of the designed system is that each module is a separate component that can be individually upgraded to improve the whole system. The algorithm is tested with a series of clinical mammograms. A sensitivity of more than 78% is obtained at a relatively low false-positive (FP) detection of 2.09 per image. The results are compared with the judgement of radiological experts, and they are very encouraging.
Clustered microcalcifications (MCCs) on mammograms are an important early sign of breast cancer. An intelligent computer-aided diagnosis system can be very helpful for radiologist in detecting and diagnosing MCCs earlier than typical screening programs. In this paper, the detection algorithm is able to extract high-frequency signal and remove low-frequency background by exploiting a difference-image technique in which a signal-suppressed image is subtracted from a signal-enhanced image to remove the structured background in a mammogram. The difference image is thresholded to detect these MCCs in mammograms. The algorithm is tested with a series of clinical mammograms. A true positive rate ofmore than 75.5% is obtained at a false-positive (FP) detection of 2.18 per image
Clustered microcalcifications (MCCs) on mammograms are important hints of breast cancer. Nevertheless, it is a complex and difficult task for radiologists to detect the clustered MCCs from the tissue background of mammograms only by naked eyes. This paper presents a method for computer-aided detection of MCCs in digital mammograms. The detection algorithm mainly consists of two different methods. The first one, based on the difference-image technique, recognizes high-frequency signals and very high-frequency noise. The second one is able to extract high-frequency signal by exploiting a wavelet based noise suppression and neural network (NN) classification. In the false-positive reduction step, false signals are separated from MCCs by means of an AND operation on signals from two methods. The algorithm is tested with a series of clinical mammograms. A sensitivity ofmore than 90% is obtained at a relatively low false-positive (FP) detection of 2.18 per image. The results are compared with thejudgement ofradiological experts, and they are very encouraging.