Enmin Song

Characteristic analysis of Otsu threshold and its applications

This paper proves that Otsu threshold is equal to the average of the mean levels of two classes partitioned by this threshold. Therefore, when the within-class variances of two classes are different, the threshold biases toward the class with larger variance. As a result, partial pixels belonging to this class will be misclassified into the other class with smaller variance. To address this problem and based on the analysis of Otsu threshold, this paper proposes an improved Otsu algorithm that constrains the search range of gray levels. Experimental results demonstrate the superiority of new algorithm compared with Otsu method.

Ultrasound intima–media segmentation using Hough transform and dual snake model

Common carotid artery intima-media thickness (IMT), which is usually measured upon ultrasound images, is an important indicator to cardiovascular diseases. This paper proposes a snake model based segmentation method to automatically detect the boundary of intima-media for IMT measurement. In the proposed method, two contours are initialized from line segments generated by Hough transform and then evolved simultaneously by dual snake model for boundary detection. Experimental results show that the proposed method has strong robustness against ultrasound artifacts, gives better results than traditional snake model and dynamic programming based methods, and achieves similar clinical parameters to ground truth data.

Segmentation of Lung Nodules in Computed Tomography Images Using Dynamic Programming and Multidirection Fusion Techniques1

RATIONALE AND OBJECTIVES The aim of this study was to develop a novel algorithm for segmenting lung nodules on three-dimensional (3D) computed tomographic images to improve the performance of computer-aided diagnosis (CAD) systems. MATERIALS AND METHODS The database used in this study consists of two data sets obtained from the Lung Imaging Database Consortium. The first data set, containing 23 nodules (22% irregular nodules, 13% nonsolid nodules, 17% nodules attached to other structures), was used for training. The second data set, containing 64 nodules (37% irregular nodules, 40% nonsolid nodules, 62% nodules attached to other structures), was used for testing. Two key techniques were developed in the segmentation algorithm: (1) a 3D extended dynamic programming model, with a newly defined internal cost function based on the information between adjacent slices, allowing parameters to be adapted to each slice, and (2) a multidirection fusion technique, which makes use of the complementary relationships among different directions to improve the final segmentation accuracy. The performance of this approach was evaluated by the overlap criterion, complemented by the true-positive fraction and the false-positive fraction criteria. RESULTS The mean values of the overlap, true-positive fraction, and false-positive fraction for the first data set achieved using the segmentation scheme were 66%, 75%, and 15%, respectively, and the corresponding values for the second data set were 58%, 71%, and 22%, respectively. CONCLUSION The experimental results indicate that this segmentation scheme can achieve better performance for nodule segmentation than two existing algorithms reported in the literature. The proposed 3D extended dynamic programming model is an effective way to segment sequential images of lung nodules. The proposed multidirection fusion technique is capable of reducing segmentation errors especially for no-nodule and near-end slices, thus resulting in better overall performance.

Age‐dependent changes in skin surface assessed by a novel two‐dimensional image analysis

Background/purpose: Skin microrelief has been studied using various methods and devices. However, the long duration of time needed to process one sample or the expensive equipment hampered the use of those systems for routine diagnosis. Today, the emergence of new software and hardware technologies may allow this issue to be resolved.

Computer aided diagnosis of fatty liver ultrasonic images based on support vector machine

B-scan ultrasound is the primary means for the diagnosis of fatty liver. However, due to use of various ultrasound equipments, poor quality of ultrasonic images and physical differences of patients, fatty liver diagnosis is mainly qualitative, and often depends on the subjective judgment of technicians and doctors. Therefore, computer-aided feature extraction and quantitative analysis of liver B-scan ultrasonic images will help to improve clinical diagnostic accuracy, repeatability and efficiency, and could provide a measure for severity of hepatic steatosis. This paper proposed a novel method of fatty liver diagnosis based on liver B-mode ultrasonic images using support vector machine (SVM). Fatty liver diagnosis was transformed into a pattern recognition problem of liver ultrasound image features. According to the different characteristics of fatty liver and healthy liver, important image features were extracted and selected to distinguish between the two categories. These features could be represented by near-field light-spot density, near-far-field grayscale ratio, grayscale co-occurrence matrix, and neighborhood gray-tone difference matrix (NGTDM). A SVM classifier was modeled and trained using the clinical ultrasound images of both fatty liver and normal liver. It was then exploited to classify normal and fatty livers, achieving a high recognition rate. The diagnostic results are satisfactorily consistent with those made by doctors. This method could be used for computer-aided diagnosis of fatty liver, and help doctors identify the fatty liver ultrasonic images rapidly, objectively and accurately.

Automatic left ventricle segmentation in cardiac MRI using topological stable-state thresholding and region restricted dynamic programming.

RATIONALE AND OBJECTIVES Segmentation of the left ventricle (LV) is very important in the assessment of cardiac functional parameters. The aim of this study is to develop a novel and robust algorithm which can improve the accuracy of automatic LV segmentation on short-axis cardiac magnetic resonance images (MRI). MATERIALS AND METHODS The database used in this study consists of 45 cases obtained from the Sunnybrook Health Sciences Centre. The 45 cases contain 12 ischemic heart failures, 12 non-ischemic heart failures, 12 LV hypertrophies, and 9 normal cases. Three key techniques are developed in this segmentation algorithm: 1) topological stable-state thresholding method is proposed to refine the endocardial contour, 2) an edge map with non-maxima gradient suppression approach, and 3) a region-restricted technique that is proposed to improve the dynamic programming to derive the epicardial boundary. RESULTS The validation experiments were performed on a pool of data sets of 45 cases. For both endo- and epicardial contours of our results, percentage of good contours is about 91%, the average perpendicular distance is about 2 mm, and the overlapping dice metric is about 0.91. The regression and determination coefficient for the experts and our proposed method on the ejection fraction is 1.05 and 0.9048, respectively; they are 0.98 and 0.8221 for LV mass. CONCLUSIONS An automatic method using topological stable-state thresholding and region restricted dynamic programming has been proposed to segment left ventricle in short-axis cardiac MRI. Evaluation results indicate that the proposed segmentation method can improve the accuracy and robust of left ventricle segmentation. The proposed segmentation approach shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

Quaternion-Based Impulse Noise Removal From Color Video Sequences

In this paper, a new quaternion vector filter for removal of random impulse noise in color video sequences is presented. First, luminance distances and chromaticity differences that are represented in quaternion form are combined together to measure color distances between color pixels. Then, based on this new color distance mechanism, the samples along horizontal, vertical, and diagonal directions in current frame and the samples of adjacent frames on motion trajectory are used to detect whether each pixel is noisy or not. By analyzing the spatiotemporal order-statistic information about these directional samples, the video pixels are classified into noise free and noisy. Finally, 3-D weighted vector median filtering is performed on the pixels that are judged as noisy, and the other pixels remain unchanged. The experimental results show that the proposed algorithm significantly outperforms other state-of-the-art video denoising methods in terms of both objective measure and visual evaluation.

Semi-supervised multi-class Adaboost by exploiting unlabeled data

Research highlights? We propose a semi-supervised learning method by using the multi-class boosting. ? It handles K-class classification without reducing into multiple two-class problems. ? The classification accuracy of base classifier requires only 1/K or better. ? Higher classification accuracy is achieved by exploiting the unlabeled data. Semi-supervised learning has attracted much attention in pattern recognition and machine learning. Most semi-supervised learning algorithms are proposed for binary classification, and then extended to multi-class cases by using approaches such as one-against-the-rest. In this work, we propose a semi-supervised learning method by using the multi-class boosting, which can directly classify the multi-class data and achieve high classification accuracy by exploiting the unlabeled data. There are two distinct features in our proposed semi-supervised learning approach: (1) handling multi-class cases directly without reducing them to multiple two-class problems, and (2) the classification accuracy of each base classifier requiring only at least 1/K or better than 1/K (K is the number of classes). Experimental results show that the proposed method is effective based on the testing of 21 UCI benchmark data sets.

Color impulsive noise removal based on quaternion representation and directional vector order-statistics

A new method for detecting and suppressing impulsive noise in color images is presented in this paper. The proposed method is a type of switching vector filters, where the impulse detection is based on the order-statistic information about the color samples in the horizontal, vertical, and diagonal directions. The new solution first uses quaternion-based representation of color differences and median deviation-based techniques to search for the edge direction with the maximum number of similar pixels, and then utilizes the samples aligning with this edge direction to judge whether the current pixel is noisy or not and control the switching between identity (no filtering) and vector median filtering actions. Extensive experimental comparisons exhibit the validity of the proposed approach by showing significant performance improvements over other well-known color image filtering techniques.

Quaternion-based color image filtering for impulsive noise suppression

It is difficult to precisely detect all impulsive noise in color images due to the nonstationarity caused by edges and fine details. For many pixels, we can not absolutely classify them as noisy or noise-free, but can only describe them using the likelihood that they are corrupted by impulsive noise. Based on this consideration, we present a new filtering solution to removing impulsive noise from color images. The proposed method first utilizes the unit transforms of quaternions to represent the chromaticity difference of two color pixels, and then divides the image into noise-free and possible noisy pixels. Finally it performs adaptive weighted vector median filtering operations on only the possible noisy pixels to suppress noise. The new weighting mechanism is based on a joint spatial/quaternion-chromaticity criterion, which ensures that pixels with different contamination likelihoods have different contributions to the filters output. The extensive simulation results indicate that the proposed method significantly outperforms some other well-known multichannel filtering techniques.