An efficient deep neural network framework for COVID-19 lung infection segmentation.

Since the outbreak of Coronavirus Disease 2019 (COVID-19) in 2020, it has significantly affected the global health system. The use of deep learning technology to automatically segment pneumonia lesions from Computed Tomography (CT) images can greatly reduce the workload of physicians and expand traditional diagnostic methods. However, there are still some challenges to tackle the task, including obtaining high-quality annotations and subtle differences between classes. In the present study, a novel deep neural network based on Resnet architecture is proposed to automatically segment infected areas from CT images. To reduce the annotation cost, a Vector Quantized Variational AutoEncoder (VQ-VAE) branch is added to reconstruct the input images for purpose of regularizing the shared decoder and the latent maps of the VQ-VAE are utilized to further improve the feature representation. Moreover, a novel proportions loss is presented for mitigating class imbalance and enhance the generalization ability of the model. In addition, a semi-supervised mechanism based on adversarial learning to the network has been proposed, which can utilize the information of the trusted region in unlabeled images to further regularize the network. Extensive experiments on the COVID-SemiSeg are performed to verify the superiority of the proposed method, and the results are in line with expectations.

Improving Generalization via Attribute Selection on Out-of-the-Box Data.

Zero-shot learning (ZSL) aims to recognize unseen objects (test classes) given some other seen objects (training classes) by sharing information of attributes between different objects. Attributes are artificially annotated for objects and treated equally in recent ZSL tasks. However, some inferior attributes with poor predictability or poor discriminability may have negative impacts on the ZSL system performance. This letter first derives a generalization error bound for ZSL tasks. Our theoretical analysis verifies that selecting the subset of key attributes can improve the generalization performance of the original ZSL model, which uses all the attributes. Unfortunately, previous attribute selection methods have been conducted based on the seen data, and their selected attributes have poor generalization capability to the unseen data, which is unavailable in the training stage of ZSL tasks. Inspired by learning from pseudo-relevance feedback, this letter introduces out-of-the-box data-pseudo-data generated by an attribute-guided generative model-to mimic the unseen data. We then present an iterative attribute selection (IAS) strategy that iteratively selects key attributes based on the out-of-the-box data. Since the distribution of the generated out-of-the-box data is similar to that of the test data, the key attributes selected by IAS can be effectively generalized to test data. Extensive experiments demonstrate that IAS can significantly improve existing attribute-based ZSL methods and achieve state-of-the-art performance.

Temporal segmentation of EEG based on functional connectivity network structure.

In the study of brain functional connectivity networks, it is assumed that a network is built from a data window in which activity is stationary. However, brain activity is non-stationary over sufficiently large time periods. Addressing the analysis electroencephalograph (EEG) data, we propose a data segmentation method based on functional connectivity network structure. The goal of segmentation is to ensure that within a window of analysis, there is similar network structure. We designed an intuitive and flexible graph distance measure to quantify the difference in network structure between two analysis windows. This measure is modular: a variety of node importance indices can be plugged into it. We use a reference window versus sliding window comparison approach to detect changes, as indicated by outliers in the distribution of graph distance values. Performance of our segmentation method was tested in simulated EEG data and real EEG data from a drone piloting experiment (using correlation or phase-locking value as the functional connectivity strength metric). We compared our method under various node importance measures and against matrix-based dissimilarity metrics that use singular value decomposition on the connectivity matrix. The results show the graph distance approach worked better than matrix-based approaches; graph distance based on partial node centrality was most sensitive to network structural changes, especially when connectivity matrix values change little. The proposed method provides EEG data segmentation tailored for detecting changes in terms of functional connectivity networks. Our study provides a new perspective on EEG segmentation, one that is based on functional connectivity network structure differences.

The efficacy and safety of eltrombopag in treating TKI-induced thrombocytopenia in patients with chronic myeloid leukemia.

ABSTRACTThrombocytopenia is one of the most common hematological adverse reactions in chronic myeloid leukemia (CML) patients receiving tyrosine kinase inhibitors (TKI) therapy, causing life-threatening bleeding cases. However, there are fewer therapeutic drugs for TKI-induced thrombocytopenia. Eltrombopag is a non-peptide thrombopoietin receptor agonist used for the treatment of immune thrombocytopenia， aplastic anemia, and hepatitis C-associated thrombocytopenia. Nevertheless, studies of eltrombopag for TKI-induced thrombocytopenia are still lacking. This study retrospectively analyzed the clinical and test data of 21 CML patients with TKI-related thrombocytopenia. The results demonstrated that the median baseline value of thrombocytopenia in the 21 CML patients was 15.57 × 109/L [2-28 × 109/L]. Following treatment with eltrombopag, 16 patients had a significant increase in their platelet levels. The peak median for platelet increase in effective responders was 145.12 × 109/L (51-460 × 109/L). However, 5 patients failed to respond to eltrombopag. Moreover, 4 of the 21 patients enrolled had adverse reactions, including reversible liver function impairment, palpitation, headache, insomnia, and loss of appetite. Nonetheless, no cases of disease progression, thrombotic events, or myelofibrosis were observed. Hence, eltrombopag may be a useful adjunctive therapy for relieving TKI-related thrombocytopenia in patients with CML.

An outlier detection-based method for artifact removal of few-channel EEGs.

Objective.The electroencephalogram (EEG) is one of the most important brain-imaging tools. The few-channel EEG is more suitable and affordable for practical use as a wearable device. Removing artifacts from collected EEGs is a prerequisite for accurately interpreting brain function and state. Previous studies proposed methods combining signal decomposition with the blind source separation (BSS) algorithms, but most of them used threshold-based criteria for artifact rejection, resulting in a lack of effectiveness in removing specific artifacts and the excessive suppression of brain activities. In this study, we proposed an outlier detection-based method for artifact removal under the few-channel condition.Approach. The underlying components (sources) were extracted using the decomposition-BSS schema. Based on our assumptions that in the feature space, the artifact-related components are dispersed, while the components related to brain activities are closely distributed, the artifact-related components were identified and rejected using one-class support vector machine. The assumptions were validated by visualizing the distribution of clusters of components.Main results. In quantitative analyses with semisimulated data, the proposed method outperformed the threshold-based methods for various artifacts, including muscle artifact, ocular artifact, and power line noise. With a real dataset and an event-related potential dataset, the proposed method demonstrated good performance in real-life situations.Significance. This study provided a fully data-driven and adaptive method for removing various artifacts in a single process without excessive suppression of brain activities.

Complementary Attributes: A New Clue to Zero-Shot Learning.

Zero-shot learning (ZSL) aims to recognize unseen objects using disjoint seen objects via sharing attributes. The generalization performance of ZSL is governed by the attributes, which transfer semantic information from seen classes to unseen classes. To take full advantage of the knowledge transferred by attributes, in this paper, we introduce the notion of the complementary attributes (CAs), as a supplement to the original attributes, to enhance the semantic representation ability. Theoretical analyses demonstrate that CAs can improve the PAC-style generalization bound of the original ZSL model. Since the proposed CA focuses on enhancing the semantic representation, CA can be easily applied to any existing attribute-based ZSL methods, including the label-embedding strategy-based ZSL (LEZSL) and the probability-prediction strategy-based ZSL (PPZSL). In PPZSL, there is a strong assumption that all attributes are independent of each other, which is arguably unrealistic in practice. To solve this problem, a novel rank aggregation (RA) framework is proposed to circumvent the assumption. Extensive experiments on five ZSL benchmark datasets and the large-scale ImageNet dataset demonstrate that the proposed CA and RA can significantly and robustly improve the existing ZSL methods and achieve state-of-the-art performance.

Mental workload classification based on ignored auditory probes and spatial covariance.

Objective.Estimation of mental workload (MWL) levels by electroencephalography (EEG)-based mental state monitoring systems has been widely explored. Using event-related potentials (ERPs), elicited by ignored auditory probes, minimizes intrusiveness and has shown high performance for estimating MWL level when tested in laboratory situations. However, when facing real-world applications, the characteristics of ERP waveforms, like latency and amplitude, are often affected by noise, which leads to a decrease in classification performance. One approach to mitigating this is using spatial covariance, which is less sensitive to latency and amplitude distortion. In this study, we used ignored auditory probes in a single-stimulus paradigm and tested Riemannian processed covariance-based features for MWL level estimation in a realistic flight-control task.Approach.We recorded EEG data with an eight-channel system from participants while they performed a simulated drone-control task and manipulated MWL levels (high and low) by task difficulty. We compared support vector machine classification performance based on frequency band power features versus features generated via the Riemannian log map operator from spatial covariance matrices. We also compared accuracy of using data segmented as auditory ERPs versus non-ERPs, for which data windows did not overlap with the ERPs.Main results.Classification accuracy of both types of features showed no significant difference between ERPs and non-ERPs. When we ignore auditory stimuli to perform continuous decoding, covariance-based features in the gamma band had area under the receiver-operating-characteristic curve (AUC) of 0.883, which was significantly higher than band power features (AUC = 0.749).Significance.This study demonstrates that Riemannian-processed covariance features are viable for MWL classification under a realistic experimental scenario.

Error Correcting Input and Output Hashing.

Most learning-based hashing algorithms leverage sample-to-sample similarities, such as neighborhood structure, to generate binary codes, which achieve promising results for image retrieval. This type of methods are referred to as instance-level encoding. However, it is nontrivial to define a scalar to represent sample-to-sample similarity encoding the semantic labels and the data structure. To address this issue, in this paper, we seek to use a class-level encoding method, which encodes the class-to-class relationship, to take the semantic information of classes into consideration. Based on these two encodings, we propose a novel framework, error correcting input and output (EC-IO) coding, which does class-level and instance-level encoding under a unified mapping space. Our proposed model contains two major components, which are distribution preservation and error correction. With these two components, our model maps the input feature of samples and the output code of classes into a unified space to encode the intrinsic structure of data and semantic information of classes simultaneously. Under this framework, we present our hashing model, EC-IO hashing (EC-IOH), by approximating the mapping space with the Hamming space. Extensive experiments are conducted to evaluate the retrieval performance, and EC-IOH exhibits superior and competitive performances comparing with popular supervised and unsupervised hashing methods.

A one-dimensional slope detection approach.

This paper extends the scale-invariant edge detector to the one-dimensional slope. It can accurately detect the slope and estimate its parameters. The method has been verified with several mathematical functions, sample sizes, and noise levels. A contrast-invariant operator is proposed to suppress noise. The inter-sample localization and interpolation greatly improve the accuracy. The proposed slope detector is also suitable for real-world signals. In additional to above-mentioned, a threshold formula is developed for the first derivative slope detector, and the upper-bound of the filterable noise level is also explored.

Membership-degree preserving discriminant analysis with applications to face recognition.

In pattern recognition, feature extraction techniques have been widely employed to reduce the dimensionality of high-dimensional data. In this paper, we propose a novel feature extraction algorithm called membership-degree preserving discriminant analysis (MPDA) based on the fisher criterion and fuzzy set theory for face recognition. In the proposed algorithm, the membership degree of each sample to particular classes is firstly calculated by the fuzzy k-nearest neighbor (FKNN) algorithm to characterize the similarity between each sample and class centers, and then the membership degree is incorporated into the definition of the between-class scatter and the within-class scatter. The feature extraction criterion via maximizing the ratio of the between-class scatter to the within-class scatter is applied. Experimental results on the ORL, Yale, and FERET face databases demonstrate the effectiveness of the proposed algorithm.

Similarity preserving principal curve: an optimal 1-d feature extractor for data representation.

This paper discusses the problem of what kind of learning model is suitable for the tasks of feature extraction for data representation and suggests two evaluation criteria for nonlinear feature extractors: reconstruction error minimization and similarity preservation. Based on the suggested evaluation criteria, a new type of principal curve-similarity preserving principal curve (SPPC) is proposed. SPPCs minimize the reconstruction error under the condition that the similarity between similar samples are preserved in the extracted features, thus giving researchers effective and reliable cognition of the inner structure of data sets. The existence and properties of SPPCs are analyzed; a practical learning algorithm is proposed and high dimensional extensions of SPPCs are also discussed. Experimental results show the virtues of SPPCs in preserving inner structures of data sets and discovering manifolds with high nonlinearity.