Search

The CT angiography (CTA) is a clinically indicated test for the assessment of coronary luminal stenosis that requires centerline extractions. There is currently no centerline extraction algorithm that is automatic, real-time and very accurate. Therefore, we sought to (i) develop a hybrid approach by incorporating fast marching and Runge-Kutta based methods for the extraction of coronary artery centerlines from CTA; (ii) evaluate the accuracy of the present method compared to Van's method by using ground truth centerline as a reference; (iii) evaluate the coronary lumen area of our centerline method in comparison with the intravascular ultrasound (IVUS) as the standard of reference. The proposed method was found to be more computationally efficient, and performed better than the Van's method in terms of overlap measures (i.e., OV: [Formula: see text] vs. [Formula: see text]; OF: [Formula: see text] vs. [Formula: see text]; and OT: [Formula: see text] vs. [Formula: see text], all [Formula: see text]). In comparison with IVUS derived coronary lumen area, the proposed approach was more accurate than the Van's method. This hybrid approach by incorporating fast marching and Runge-Kutta based methods could offer fast and accurate extraction of centerline as well as the lumen area. This method may garner wider clinical potential as a real-time coronary stenosis assessment tool.

This paper presents a new approach to detect and segment liver tumors. The detection and segmentation of liver tumors can be formulized as novelty detection or two-class classification problem. Each voxel is characterized by a rich feature vector, and a classifier using random feature subspace ensemble is trained to classify the voxels. Since Extreme Learning Machine (ELM) has advantages of very fast learning speed and good generalization ability, it is chosen to be the base classifier in the ensemble. Besides, majority voting is incorporated for fusion of classification results from the ensemble of base classifiers. In order to further increase testing accuracy, ELM autoencoder is implemented as a pre-training step. In automatic liver tumor detection, ELM is trained as a one-class classifier with only healthy liver samples, and the performance is compared with two-class ELM. In liver tumor segmentation, a semi-automatic approach is adopted by selecting samples in 3D space to train the classifier. The proposed method is tested and evaluated on a group of patients' CT data and experiment show promising results.

A novel vessel context-based voting is proposed for automatic liver vasculature segmentation in CT images. It is able to conduct full vessel segmentation and recognition of multiple vasculatures effectively. The vessel context describes context information of a voxel related to vessel properties, such as intensity, saliency, direction and connectivity. Voxels are grouped to liver vasculatures hierarchically based on vessel context. They are first grouped locally into vessel branches with the advantage of a vessel junction measurement, and then grouped globally into vasculatures, which is implemented using a multiple feature point voting mechanism. The proposed method has been evaluated on 10 clinical CT datasets. Segmentation of third-order vessel trees from CT images (0.76 × 0.76 × 2.0mm) of the portal venous phase takes less than 3 min on a PC with 2.0 GHz dual core processor and the average segmentation accuracy is up to 98%.

Lots of studies on myocardial infarction (MI) computer assisted diagnosis are based on certain important ECG components which only account for local information. 12-Lead ECG signals which were regarded as hyper-dimensional time-series data were utilized to extract features from global information in this study. Existing feature extraction techniques for classification attempt to classify all the classes included. However sometimes it is more important to better recognize certain specific classes rather than to discriminate all the classes. A feature extraction method based on subjective-classification was proposed to discriminate the specific classes, which the classification priority was given subjectively, and each of the other classes was separated at the same time. The method includes data reduction by principal component analysis (PCA), data normalization by whitening transformation and derivation of projecting vectors for subjective-classification, etc. The data in the analysis were collected from PTB diagnostic ECG database. The results show that the proposed method can obtain a small number of effective features from 12-lead ECGs to better classify classes with priority, and the other classes can be classified at the same time.

In this paper, we consider the extraction of nasopharyngeal carcinoma lesion from MR images as a region segmentation problem. We propose a semi-supervised segmentation approach to segment the lesion in two steps. First, a metric is learned in a supervised fashion, which maximizes the separation between two groups of pixels (tumor or non-tumor) with minimal user interaction. Second, the learned metric is used to complete extraction of tumor region in an unsupervised fashion. Several experiments were conducted to evaluate the performance of similar methods with learned metrics for grouping or classifying pixels to form the tumor region. It is observed that the spectral clustering-based method performs well and the performance is comparable or marginally better than the discriminative SVM-based method.

Efficient scheduling algorithms have been a leading research topic for heterogeneous computing systems. Although duplication-based scheduling algorithms can significantly reduce the total completion time, they are generally accompanied by an exorbitant time complexity. In this paper, we propose a new task duplication-based heuristic scheduling algorithm, LDLS, that can reduce the total completion time and maintains a low time complexity. The scheduling procedure of LDLS is composed of three main phases: In the beginning phase, the maximum number of duplications per level and per task is calculated to prevent excessive duplications from blocking regular tasks. In the next phase, the optimistic cost table (OCT) and ranking of tasks are calculated with reference to PEFT. In the final phase, scheduling is conducted based on the ranking, and the duplication of each task is dynamically determined, enabling the duplicated tasks to effectively reduce the start execution time of its successor tasks. Experiments of algorithms on randomly generated graphs and real-world applications indicate that both the scheduling length and the number of better case occurrences of LDLS are better than others. [ABSTRACT FROM AUTHOR]

In clinical diagnosis of nasopharyngeal carcinoma (NPC) lesion, clinicians are often required to delineate boundaries of NPC on a number of tumor-bearing magnetic resonance images, which is a tedious and time-consuming procedure highly depending on expertise and experience of clinicians. Computer-aided tumor segmentation methods (either contour-based or region-based) are necessary to alleviate clinicians' workload. For contour-based methods, a minimal user interaction to draw an initial contour inside or outside the tumor lesion for further curve evolution to match the tumor boundary is preferred, but parameters within most of these methods require manual adjustment, which is technically burdensome for clinicians without specific knowledge. Therefore, segmentation methods with a minimal user interaction as well as automatic parameters adjustment are often favored in clinical practice. In this paper, two region-based methods with parameters learning are introduced for NPC segmentation. Two hundred fifty-three MRI slices containing NPC lesion are utilized for evaluating the performance of the two methods, as well as being compared with other similar region-based tumor segmentation methods. Experimental results demonstrate the superiority of adopting learning in the two introduced methods. Also, they achieve comparable segmentation performance from a statistical point of view. [ABSTRACT FROM AUTHOR]