Fetal congenital heart disease (FHD) is a common and serious congenital malformation in children. In Asia, FHD birth defect rates have reached as high as 9.3‰. For the early detection of birth defects and mortality, echocardiography remains the most effective method for screening fetal heart malformations. However, standard echocardiograms of the fetal heart, especially four-chamber view images, are difficult to obtain. In addition, the pathophysiological changes in fetal hearts during different pregnancy periods lead to ever-changing twodimensional fetal heart structures and hemodynamics, and it requires extensive professional knowledge to recognize and judge disease development. Thus, research on the automatic screening for FHD is necessary. In this paper, we proposed a new model named DGACNN that shows the best performance in recognizing FHD, achieving a rate of 85%. The motivation for this network is to deal with the problem that there are insufficient training datasets to train a robust model. There are many unlabeled video slices, but they are tough and time-consuming to annotate. Thus, how to use these unannotated video slices to improve the DGACNN capability for recognizing FHD, in terms of both recognition accuracy and robustness, is very meaningful for FHD screening. The architecture of DGACNN comprises two parts, that is, DANomaly and GACNN (Wgan-GP and CNN). DANomaly, similar to the ALOCC network, but incorporates cycle adversarial learning to train an end-to-end one-class classification (OCC) network that is more robust and has a higher accuracy than ALOCC in screening video slices. For the GACNN architecture, we use FCH(four chamber heart) video slices at around the endsystole, as screened by DANomaly, to train a WGAN-GP for the purpose of obtaining ideal low-level features that can robustly improve the FHD recognition accuracy. A few annotated video slices, as screened by DANomaly, can also be used for data augmentation so as to improve the FHD recognition further. The experiments show that the DGACNN outperforms other state-of-the-art networks by 1%–20% in recognizing FHD.A comparison experiment shows that the proposed network already outperforms the performance of expert cardiologists in recognizing FHD, reaching 84% in a test. Thus, the proposed architecture has high potential for helping cardiologists complete early FHD screenings.

A clustering algorithm named “Clustering by fast search and find of density peaks” is for finding the centers of clusters quickly. Its accuracy excessively depended on the threshold, and no efficient way was given to select its suitable value, i.e., the value was suggested be estimated on the basis of empirical experience. A new way is proposed to automatically extract the optimal value of threshold by using the potential entropy of data field from the original dataset. For any dataset to be clustered, the threshold can be calculated from the dataset objectively instead of empirical estimation. The results of comparative experiments have shown the algorithm with the threshold from data field can get better clustering results than with the threshold from empirical experience.

In this paper, data field is proposed to group data objects via simulating their mutual interactions and opposite movements for hierarchical clustering. Enlightened by the field in physical space, data field to simulate nuclear field is presented to illuminate the interaction between objects in data space. In the data field, the self-organized process of equipotential lines on many data objects discovers their hierarchical clustering-characteristics. During the clustering process, a random sample is first generated to optimize the impact factor. The masses of data objects are then estimated to select core data object with nonzero masses. Taking the core data objects as the initial clusters, the clusters are iteratively merged hierarchy by hierarchy with good performance. The results of a case study show that the data field is capable of hierarchical clustering on objects varying size, shape or granularity without user-specified parameters, as well as considering the object features inside the clusters and removing the outliers from noisy data. The comparisons illustrate that the data field clustering performs better than K-means, BIRCH, CURE, and CHAMELEON.

The sequential recommendation (also known as the next-item recommendation), which aims to predict the following item to recommend in a session according to users’ historical behavior, plays a critical role in improving session-based recommender systems. Most of the existing deep learning-based approaches utilize the recurrent neural network architecture or self-attention to model the sequential patterns and temporal influence among a user's historical behavior and learn the user's preference at a specific time. However, these methods have two main drawbacks. First, they focus on modeling users’ dynamic states from a user-centric perspective and always neglect the dynamics of items over time. Second, most of them deal with only the first-order user-item interactions and do not consider the high-order connectivity between users and items, which has recently been proved helpful for the sequential recommendation. To address the above problems, in this article, we attempt to model user-item interactions by a bipartite graph structure and propose a new recommendation approach based on a Position-enhanced and Time-aware Graph Convolutional Network (PTGCN) for the sequential recommendation. PTGCN models the sequential patterns and temporal dynamics between user-item interactions by defining a position-enhanced and time-aware graph convolution operation and learning the dynamic representations of users and items simultaneously on the bipartite graph with a self-attention aggregator. Also, it realizes the high-order connectivity between users and items by stacking multi-layer graph convolutions. To demonstrate the effectiveness of PTGCN, we carried out a comprehensive evaluation of PTGCN on three real-world datasets of different sizes compared with a few competitive baselines. Experimental results indicate that PTGCN outperforms several state-of-the-art sequential recommendation models in terms of two commonly-used evaluation metrics for ranking. In particular, it can make a better trade-off between recommendation performance and model training efficiency, which holds great potential for online session-based recommendation scenarios in the future.

Schizophrenia is a kind of serious mental illness. Due to the lack of an objective physiological data supporting and a unified data analysis method, doctors can only rely on the subjective experience of the data to distinguish normal people and patients, which easily lead to misdiagnosis. In recent years, functional Near-Infrared Spectroscopy (fNIRS) has been widely used in clinical diagnosis, it can get the hemoglobin concentration through the variation of optical intensity. Firstly, the prefrontal brain networks were constructed based on oxy-Hb signals from 52-channel fNIRS data of schizophrenia and healthy controls. Then, Complex Brain Network Analysis (CBNA) was used to extract features from the prefrontal brain networks. Finally, a classier based on Support Vector Machine (SVM) is designed and trained to discriminate schizophrenia from healthy controls. We recruited a sample which contains 34 healthy controls and 42 schizophrenia patients to do the one-back memory task. The hemoglobin response was measured in the prefrontal cortex during the task using a 52-channel fNIRS system. The experimental results indicate that the proposed method can achieve a satisfactory classification with the accuracy of 85.5%, 92.8% for schizophrenia samples and 76.5% for healthy controls. Also, our results suggested that fNIRS has the potential capacity to be an effective objective biomarker for the diagnosis of schizophrenia. Our results suggested that, using the appropriate classification method, fNIRS has the potential capacity to be an effective objective biomarker for the diagnosis of schizophrenia.