Search

Objective: To describe and analyze the clinical manifestation and pre-DRF of UPJO children with polyps and explore the possible influencing factors of supranormal pre-DRF. Patients and methods: All patients undergoing primary Anderson-Hynes pyeloplasty for UPJO due to polyp were retrospectively reviewed. Patients' characteristics, parameters of ultrasound and dynamic renograms (DR) were recorded in elaborate. Pre-DRF in groups of different age, weight, gender, pain, grade of hydronephrosis, anterio-posterior pelvic diameter (APD), length of kidney and postoperative ultrasonic parameters were compared. Results: A total of 18 UPJO children with polyps were included. Five (27.78%) patients had SFU III grade of hydronephrosis. Seven (38.89%) patients were supranormal pre-DRF. All patients had pre-DRF > 40%. Drainage curve was delayed excretion in 12 (66.67%) patients and T1/2 < 20 min was in 4 (22.22%) patients. Among the 16 patients who underwent preoperative IVP examination, 15 (93.75%) patients had concentration of intrarenal pelvis contrast agent within 10 min. No significant difference in post-APD reduction rate and post-minPT increased was found between supranormal pre-DRF and non-supranormal pre-DRF groups. The supranormal pre-DRF was more likely to occur in young and low-weight children. Conclusion: The preoperative renal function of UPJO patients with polyps was well preserved, and 38.89% of them had supranormal pre-DRF. Patients with supranormal pre-DRF can be managed with the same strategies as those with normal renal function. [ABSTRACT FROM AUTHOR]

Numerous lung nodule candidates can be produced through an automated lung nodule detection system. Classifying these candidates to reduce false positives is an important step in the detection process. The objective during this paper is to predict real nodules from a large number of pulmonary nodule candidates. Facing the challenge of the classification task, we propose a novel 3D convolution neural network (CNN) to reduce false positives in lung nodule detection. The novel 3D CNN includes embedded multiple branches in its structure. Each branch processes a feature map from a layer with different depths. All of these branches are cascaded at their ends; thus, features from different depth layers are combined to predict the categories of candidates. The proposed method obtains a competitive score in lung nodule candidate classification on LUNA16 dataset with an accuracy of 0.9783, a sensitivity of 0.8771, a precision of 0.9426, and a specificity of 0.9925. Moreover, a good performance on the competition performance metric (CPM) is also obtained with a score of 0.830. As a 3D CNN, the proposed model can learn complete and three-dimensional discriminative information about nodules and non-nodules to avoid some misidentification problems caused due to lack of spatial correlation information extracted from traditional methods or 2D networks. As an embedded multi-branch structure, the model is also more effective in recognizing the nodules of various shapes and sizes. As a result, the proposed method gains a competitive score on the false positive reduction in lung nodule detection and can be used as a reference for classifying nodule candidates. [ABSTRACT FROM AUTHOR]

This paper proposes a new decision tree approach for analog circuit fault diagnosis using binary support vector machines (BSVMs) that are trained by using different data sets. To evaluate the performance of those different BSVMs, a new criterion is defined by considering the weighted number of fault-pairs that can be separated by one BSVM. Specifically, the weight of each fault-pair is inversely proportional to the number of BSVMs that can separate it. Simulation results by using two example circuits demonstrate that the proposed approach achieves better performance in terms of accuracy and efficiency than existing approaches. [ABSTRACT FROM AUTHOR]

A novel modified binary bat algorithm (MBBA) with chaos and Doppler effect in echoes is presented in this paper to deal with the fault diagnosis problem of analog circuits with tolerance. The proposed method focuses on further mimicking the bats' behaviors and improving the binary bat algorithm (BBA) in view of biology. The proposed algorithm utilizes the bats' self-adaptive compensation for Doppler effect in echoes and the chaotic optimization method to improve the BBA. The elitist strategy is also used to store all the possible global best solutions, which will provide us with more possible choices in practice. Since single feature extraction method has limitations during fault diagnosis of analog circuits with tolerance, several kinds of feature extraction methods are used together to build a big feature set firstly in our method, and then, the new proposed MBBA is used to select the optimal feature subset simultaneously optimize the classifier parameters for the analog fault diagnosis. Three analog circuits' examples are given to demonstrate the effectiveness and generalization ability of our method, and other methods are also used to do comparisons. The results indicate that combining different kinds of feature extraction method together to optimally select the feature subset simultaneously optimize the classifier parameters by MBBA can obtain better performance in dealing with analog fault diagnosis problems, and our proposed method provides a better tradeoff between fault diagnosis accuracy and total cost, which is superior to other methods. Therefore, it is an effective method for analog fault diagnosis and has broad application prospects in practice. [ABSTRACT FROM AUTHOR]

Feature extraction plays an important role in the field of fault diagnosis of analog circuits. How to effectively extract fault features is crucial to diagnostic accuracy. The components tolerance and circuit nonlinearities of analog circuits can cause some part overlapping of primal signal among different component faults in time domain and frequency domain. Currently, the existing method aims at wavelet features, statistical property features, conventional frequency features and conventional time-domain features. There is no decoupling ability for the feature extraction methods mentioned above. To solve the problem, a new fault features extraction method is proposed. The diagnostic results are compared with those from other methods. Firstly, it is proposed to use the statistical property features of transformed signals by the fractional Fourier transform in the optimal fractional order domain as fault features, such as range, mean, standard deviation, skewness, kurtosis, entropy, median, the third central moment, and centroid. And then, KPCA is used to reduce the dimensionality of candidate features so as to obtain the optimal features. Next, normalization is applied to rescale input features. Finally, extracted features are trained by SVM to diagnose faulty components in analog circuits. The simulation results show that compared with traditional methods, the proposed method is quite efficient to improve diagnostic accuracy. [ABSTRACT FROM AUTHOR]

A set of novel methods are proposed to optimize analog test point selection in this paper. We have introduced chaos and elitist strategy into the binary bat algorithm so as to increase its global search mobility and effectiveness for robust global optimization. In the present study, nine well known chaotic maps are introduced and used to construct the chaotic binary bat algorithm (CBBA) respectively. As a result, we have developed nine different CBBAs and for the first time applied them to deal with the analog test point selection problem. The attractiveness of our proposed CBBA mainly lies in two aspects: the utilization of chaotic maps to tune the BBA parameter and the application of elitist strategy to store all the possible global best solutions. These improvements obviously enhance the performance of BBA and make our new algorithms (CBBAs) get all the best solutions (usually more than one) easily and effectively, which will give us more possible choices in practice. Analog circuits' examples and a group of statistical experiments are given to demonstrate the feasibility and effectiveness of the proposed algorithms. The other reported algorithms are also used to do the comparison. The results indicate that the proposed algorithms have excellent performance in finding the optimum test point sets. Therefore, they are good solutions and applicable to actual circuits and engineering practice. [ABSTRACT FROM AUTHOR]

A new method to select an optimum test point set in analog fault diagnosis is proposed in this paper. As the probability density of the circuit output approximately satisfies the normal distribution, an accurate way for determining the fault ambiguity gap is used to calculate the isolation probability of the faults. The proposed fault-pair isolation table derived from the mean and standard deviation values of node voltage can exactly represent the fault-pair isolation capability of the test points. The special test points that can uniquely isolate some particular fault pairs are selected first. This step can help to save the total cost of the computation time and even find the final solution directly. After removing the isolated fault pairs (rows) and the selected test points (columns), the size of the fault-pair isolation table could reduce dramatically. If more optimum test points are needed, the normalized fault-pair isolation probability values in the table are used to select the right test point that has the largest fault-pair isolation capability among all the candidate test points. Analog circuits' examples and the statistical experiments are given to demonstrate the feasibility and effectiveness of the proposed algorithm. The other reported algorithms are also used to do the comparison. The results indicate that the proposed algorithm has excellent performance in minimizing the size of the test point set. Therefore, it is a good solution and applicable to actual circuits and engineering practice. [ABSTRACT FROM AUTHOR]

A new efficient method to select an optimum or near optimum set of test points for fault dictionary techniques in analog fault diagnosis is proposed. This is done by constructing a fault-isolated table firstly according to the integer-coded fault dictionary, and picking the special test points with particular fault isolate abilities out from the candidate test points set. This step can help us save the total cost of computation and even find the global minimum test-point set directly. After this step, the selected test points might isolate most of the defined faults together, and these isolated faults (rows) and the chosen test points (columns) are deleted from the fault dictionary, the dimension of which will reduce quite a lot. And then the inclusive and exclusive approach are combined together to choose other optimum test points from the candidate test points set to solve the problem. Three analog circuits' examples and the statistical experiments are given to demonstrate the feasibility and effectiveness of the proposed algorithm, the existing classical algorithms are also used to do the comparison. The results indicate that the proposed method has a better tradeoff between the solution accuracy and the computational cost, and it is superior to other methods in its computational efficiency and quality of final solution. Therefore, it is a good solution to minimize the size of the test-point set, and practical for medium and large scale systems. [ABSTRACT FROM AUTHOR]