Distinguished Representations Research Articles

Open-world electrocardiogram classification via domain knowledge-driven contrastive learning

Automatic electrocardiogram (ECG) classification provides valuable auxiliary information for assisting disease diagnosis and has received much attention in research. The success of existing classification models relies on fitting the labeled samples for every ECG type. However, in practice, well-annotated ECG datasets usually cover only limited ECG types. It thus raises an issue: conventional classification models trained with limited ECG types can only identify those ECG types that have already been observed in the training set, but fail to recognize unseen (or unknown) ECG types that exist in the wild and are not included in training data. In this work, we investigate an important problem called open-world ECG classification that can predict fine-grained observed ECG classes and identify unseen classes. Accordingly, we propose a customized method that first incorporates clinical knowledge into contrastive learning by generating “hard negative” samples to guide learning diagnostic ECG features (i.e., distinguishable representations), and then performs multi-hypersphere learning to learn compact ECG representations for classification. The experiment results on 12-lead ECG datasets (CPSC2018, PTB-XL, and Georgia) demonstrate that the proposed method outperforms the state-of-the-art methods. Specifically, our method achieves superior accuracy than the comparative methods on the unseen ECG class and certain seen classes. Overall, the investigated problem (i.e., open-world ECG classification) helps to draw attention to the reliability of automatic ECG diagnosis, and the proposed method is proven effective in tackling the challenges. The code and datasets are released at https://github.com/betterzhou/Open_World_ECG_Classification.

Archimedean distinguished representations and exceptional poles

Archimedean distinguished representations and exceptional poles

SPReCHD: Four-Chamber Semantic Parsing Network for Recognizing Fetal Congenital Heart Disease in Medical Metaverse.

Echocardiography is essential for evaluating cardiac anatomy and function during early recognition and screening for congenital heart disease (CHD), a widespread and complex congenital malformation. However, fetal CHD recognition still faces many difficulties due to instinctive fetal movements, artifacts in ultrasound images, and distinctive fetal cardiac structures. These factors hinder capturing robust and discriminative representations from ultrasound images, resulting in CHD's low prenatal detection rate. Hence, we propose a multi-scale gated axial-transformer network (MSGATNet) to capture fetal four-chamber semantic information. Then, we propose a SPReCHD: four-chamber semantic parsing network for recognizing fetal CHD in the clinical treatment of the medical metaverse, integrating MSGATNet to segment and locate four-chamber arbitrary contours, further capturing distinguished representations for the fetal heart. Comprehensive experiments indicate that our SPReCHD is sufficient in recognizing fetal CHD, achieving a precision of 95.92%, a recall of 94%, an accuracy of 95%, and a F1 score of 94.95% on the test set, dramatically improving the fetal CHD's prenatal detection rate.

WLEDD: Legal judgment prediction with legal feature word subgraph label-embedding and dual-knowledge distillation

Legal judgment prediction(LJP) has achieved remarkable results. However, existing methods still face problems such as difficulties in obtaining key feature words for charges, which impose limitations on the improvement of prediction results. To this end, we propose a legal judgment prediction model with legal feature Word subgraph Label-Embedding and Dual-knowledge Distillation(WLEDD). Compared with traditional methods, our method has two contributions: (1) To mitigate the impact of overly sparse tail class data and high similarity text representations, we capture the critical features related to the charges by fusing LDA and legal feature word subgraphs. Then we encode them as label information to obtain highly distinguished representations of legal documents. (2) To solve the problem of high difficulty in some subtasks in LJP, we perform subtask-oriented compression of models to construct a student model with lower complexity and higher accuracy through dual knowledge distillation. Moreover, we exploit the logical association between the subtasks to constrain the labels of articles by charge prediction results. It greatly reduces the difficulty of article prediction. Experimental results on four datasets show that our approach significantly outperforms the baseline models. Compared with the state-of-art method, the F1 value of WLEDD for charge prediction has increased by an average of 2.57% . For article prediction, the F1 value has increased by an average of 1.09% . In addition, we demonstrate its effectiveness through ablation experiments and analytical experiments.

A metric learning-based method for biomedical entity linking.

Biomedical entity linking task is the task of mapping mention(s) that occur in a particular textual context to a unique concept or entity in a knowledge base, e.g., the Unified Medical Language System (UMLS). One of the most challenging aspects of the entity linking task is the ambiguity of mentions, i.e., (1) mentions whose surface forms are very similar, but which map to different entities in different contexts, and (2) entities that can be expressed using diverse types of mentions. Recent studies have used BERT-based encoders to encode mentions and entities into distinguishable representations such that their similarity can be measured using distance metrics. However, most real-world biomedical datasets suffer from severe imbalance, i.e., some classes have many instances while others appear only once or are completely absent from the training data. A common way to address this issue is to down-sample the dataset, i.e., to reduce the number instances of the majority classes to make the dataset more balanced. In the context of entity linking, down-sampling reduces the ability of the model to comprehensively learn the representations of mentions in different contexts, which is very important. To tackle this issue, we propose a metric-based learning method that treats a given entity and its mentions as a whole, regardless of the number of mentions in the training set. Specifically, our method uses a triplet loss-based function in conjunction with a clustering technique to learn the representation of mentions and entities. Through evaluations on two challenging biomedical datasets, i.e., MedMentions and BC5CDR, we show that our proposed method is able to address the issue of imbalanced data and to perform competitively with other state-of-the-art models. Moreover, our method significantly reduces computational cost in both training and inference steps. Our source code is publicly available here.

Neural Representations of Post-Decision Accuracy and Reward Expectation in the Caudate Nucleus and Frontal Eye Field.

Performance monitoring that supports ongoing behavioral adjustments is often examined in the context of either choice confidence for perceptual decisions (i.e., "did I get it right?") or reward expectation for reward-based decisions (i.e., "what reward will I receive?"). However, our understanding of how the brain encodes these distinct evaluative signals remains limited because they are easily conflated, particularly in commonly used two-alternative tasks with symmetric rewards for correct choices. Previously we used a motion-discrimination task with asymmetric rewards to identify neural substrates of forming reward-biased perceptual decisions in the caudate nucleus (part of the striatum in the basal ganglia) and the frontal eye field (FEF, in prefrontal cortex). Here we leveraged this task design to partially decouple estimates of accuracy and reward expectation and examine their impacts on subsequent decisions and their representations in those two brain areas. We identified distinguishable representations of these two evaluative signals in individual caudate and FEF neurons, with regional differences in their distribution patterns and time courses. We observed that well-trained monkeys (both sexes) used both evaluative signals, infrequently but consistently, to adjust their subsequent decisions. We found further that these behavioral adjustments had reliable relationships with the neural representations of both evaluative signals in caudate, but not FEF. These results suggest that the cortico-striatal decision network may use diverse evaluative signals to monitor and adjust decision-making behaviors, adding to our understanding of the different roles that the FEF and caudate nucleus play in a diversity of decision-related computations.

Contribution of coincidence detection to speech segregation in noisy environments

This study introduces a biologically-inspired model designed to examine the role of coincidence detection cells in speech segregation tasks. The model consists of three stages: a time-domain cochlear model that generates instantaneous rates of auditory nerve fibers, coincidence detection cells that amplify neural activity synchronously with speech presence, and an optimal spectro-temporal speech presence estimator. A comparative analysis between speech estimation based on the firing rates of auditory nerve fibers and those of coincidence detection cells indicates that the neural representation of coincidence cells significantly reduces noise components, resulting in a more distinguishable representation of speech in noise. The proposed framework demonstrates the potential of brainstem nuclei processing in enhancing auditory skills. Moreover, this approach can be further tested in other sensory systems in general and within the auditory system in particular.

Harmonic Analysis on the Affine Group of the Plane

For any natural number n, the group $$G_n$$ of all invertible affine transformations of n-dimensional Euclidean space has, up to equivalence, just one square-integrable representation and the left regular representation of $$G_n$$ is a multiple of this square-integrable representation. We provide a concrete realization $$\sigma $$ of this distinguished representation in the two-dimensional case. We explicitly decompose the Hilbert space $$L^2(G_2)$$ as a direct sum of left-invariant closed subspaces on each of which the left regular representation acts as a representation equivalent to $$\sigma $$ .

MIM-Graph: A multi-sensor network approach for fault diagnosis of HSR Bogie bearings at the IoT edge via mutual information maximization

The Internet of Things (IoT) is crucial in developing next-generation high-speed railways (HSRs). HSR IoT enables intelligent diagnosis of trains using multi-sensor data, which is critical for maintaining high speeds and ensuring passenger safety. Graph neural network (GNN)-based methods have gained popularity in HSR IoT research due to the ability to represent the sensor network as intuitive graphs. However, labeling monitoring data in the HSR scenario takes time and effort. To address this challenge, we propose a semi-supervised graph-level representation learning approach called MIM-Graph, which uses mutual information maximization to learn from a large amount of unlabeled data. First, the multi-sensor data is converted into association graphs based on their spatial topology. The unsupervised encoder is trained using global–local mutual maximization. The teacher–student framework transfers knowledge from the unsupervised encoder learned to the supervised encoder, which is trained using a small amount of labeled data. As a result, the supervised encoder learns distinguishable representations for intelligent diagnosis of HSR. We evaluate the proposed method using CWRU dataset and data from HSR Bogie test platform, and the experimental results demonstrate the effectiveness and superiority of MIM-Graph.

Boosting automatic COVID-19 detection performance with self-supervised learning and batch knowledge ensembling

Problem:Detecting COVID-19 from chest X-ray (CXR) images has become one of the fastest and easiest methods for detecting COVID-19. However, the existing methods usually use supervised transfer learning from natural images as a pretraining process. These methods do not consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Aim:In this paper, we want to design a novel high-accuracy COVID-19 detection method that uses CXR images, which can consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Methods:Our method consists of two phases. One is self-supervised learning-based pertaining; the other is batch knowledge ensembling-based fine-tuning. Self-supervised learning-based pretraining can learn distinguished representations from CXR images without manually annotated labels. On the other hand, batch knowledge ensembling-based fine-tuning can utilize category knowledge of images in a batch according to their visual feature similarities to improve detection performance. Unlike our previous implementation, we introduce batch knowledge ensembling into the fine-tuning phase, reducing the memory used in self-supervised learning and improving COVID-19 detection accuracy. Results:On two public COVID-19 CXR datasets, namely, a large dataset and an unbalanced dataset, our method exhibited promising COVID-19 detection performance. Our method maintains high detection accuracy even when annotated CXR training images are reduced significantly (e.g., using only 10% of the original dataset). In addition, our method is insensitive to changes in hyperparameters. Conclusion:The proposed method outperforms other state-of-the-art COVID-19 detection methods in different settings. Our method can reduce the workloads of healthcare providers and radiologists.

Distinction inside L-packets of SL(n)

If $E/F$ is a quadratic extension $p$-adic fields, we first prove that the $\mathrm{SL}_n(F)$-distinguished representations inside a distinguished unitary L-packet of $\mathrm{SL}_n(E)$ are precisely those admitting a degenerate Whittaker model with respect to a degenerate character of $N(E)/N(F)$. Then we establish a global analogue of this result. For this, let $E/F$ be a quadratic extension of number fields and let $\pi$ be an $\mathrm{SL}_n(\mathbb{A}_F)$-distinguished square integrable automorphic representation of $\mathrm{SL}_n(\mathbb{A}_E)$. Let $(\sigma,d)$ be the unique pair associated to $\pi$, where $\sigma$ is a cuspidal representation of $\mathrm{GL}_r(\mathbb{A}_E)$ with $n=dr$. Using an unfolding argument, we prove that an element of the L-packet of $\pi$ is distinguished with respect to $\mathrm{SL}_n(\mathbb{A}_F)$ if and only if it has a degenerate Whittaker model for a degenerate character $\psi$ of type $r^d:=(r,\dots,r)$ of $N_n(\mathbb{A}_E)$ which is trivial on $N_n(E+\mathbb{A}_F)$, where $N_n$ is the group of unipotent upper triangular matrices of $\mathrm{SL}_n$. As a first application, under the assumptions that $E/F$ splits at infinity and $r$ is odd, we establish a local-global principle for $\mathrm{SL}_n(\mathbb{A}_F)$-distinction inside the L-packet of $\pi$. As a second application we construct examples of distinguished cuspidal automorphic representations $\pi$ of $\mathrm{SL}_n(\mathbb{A}_E)$ such that the period integral vanishes on some canonical copy of $\pi$, and of everywhere locally distinguished representations of $\mathrm{SL}_n(\mathbb{A}_E)$ such that their L-packets do not contain any distinguished representation.

Semantic Communities from Graph-Inspired Visual Representations of Cityscapes

The swift development of autonomous vehicles raises the necessity of semantically mapping the environment by producing distinguishable representations to recognise similar areas. To this end, in this article, we present an efficient technique to cut up a robot’s trajectory into semantically consistent communities based on graph-inspired descriptors. This allows an agent to localise itself in future tasks under different environmental circumstances in an urban area. The proposed semantic grouping technique utilizes the Leiden Community Detection Algorithm (LeCDA), which is a novel and efficient method of low computational complexity and exploits semantic and topometric information from the observed scenes. The presented experimentation was carried out on a novel dataset from the city of Xanthi, Greece (dubbed as Gryphonurban urban dataset), which was recorded by RGB-D, IMU and GNSS sensors mounted on a moving vehicle. Our results exhibit the formulation of a semantic map with visually coherent communities and the realisation of an effective localisation mechanism for autonomous vehicles in urban environments.

Functional connectivity profiles of the default mode and visual networks reflect temporal accumulative effects of sustained naturalistic emotional experience

Determining and decoding emotional brain processes under ecologically valid conditions remains a key challenge in affective neuroscience. The current functional Magnetic Resonance Imaging (fMRI) based emotion decoding studies are mainly based on brief and isolated episodes of emotion induction, while sustained emotional experience in naturalistic environments that mirror daily life experiences are scarce. Here we used 12 different 10-minute movie clips as ecologically valid emotion-evoking procedures in n = 52 individuals to explore emotion-specific fMRI functional connectivity (FC) profiles on the whole-brain level at high spatial resolution (432 parcellations including cortical and subcortical structures). Employing machine-learning based decoding and cross validation procedures allowed to investigate FC profiles contributing to classification that can accurately distinguish sustained happiness and sadness and that generalize across subjects, movie clips, and parcellations. Both functional brain network-based and subnetwork-based emotion classification results suggested that emotion manifests as distributed representation of multiple networks, rather than a single functional network or subnetwork. Further, the results showed that the Visual Network (VN) and Default Mode Network (DMN) associated functional networks, especially VN-DMN, exhibited a strong contribution to emotion classification. To further estimate the temporal accumulative effect of naturalistic long-term movie-based video-evoking emotions, we divided the 10-min episode into three stages: early stimulation (1∼200 s), middle stimulation (201∼400 s), and late stimulation (401∼600 s) and examined the emotion classification performance at different stimulation stages. We found that the late stimulation contributes most to the classification (accuracy=85.32%, F1-score=85.62%) compared to early and middle stimulation stages, implying that continuous exposure to emotional stimulation can lead to more intense emotions and further enhance emotion-specific distinguishable representations. The present work demonstrated that sustained happiness and sadness under naturalistic conditions are presented in emotion-specific network profiles and these expressions may play different roles in the generation and modulation of emotions. These findings elucidated the importance of network level adaptations for sustained emotional experiences during naturalistic contexts and open new venues for imaging network level contributions under naturalistic conditions.

Projection-preserving block-diagonal low-rank representation for subspace clustering

In this paper, a novel model named projection-preserving block-diagonal low-rank representation (PBDIR) is proposed and can obtain a more distinguishable representation matrix for clustering. PBDIR acquires a more advantageous representation by extracting the essential features. Specifically, we introduce a projection matrix to our model to learn a new feature space that can capture more significant features. Therefore, our model learns a more robust representation, which can reduce noise interference. Meanwhile, we introduce a block-diagonal regularization to ensure that the obtained representation matrix involves a k-block diagonal, where k denotes the number of clusters. This term brings more benefits for clustering tasks. Experimental results on real datasets show that our model can significantly improve the clustering performance and the proposed approach is robust against Gaussian noise, Multiplicative noise, and Salt-and-Pepper noise.

CCR and CAR Algebras are Connected Via a Path of Cuntz–Toeplitz Algebras

For q in {mathbb {R}}, |q| < 1 we consider the universal enveloping C^*-algebra of a *-algebra of q-canonical commutation relations (q-CCR), which is generated by a_1, ldots , a_n subject to the relations ai∗aj=δij1+qajai∗.\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\begin{aligned} a_i^* a_j = \\delta _{ij} 1 + q a_j a_i^* . \\end{aligned}$$\\end{document}It has a distinguished representation pi _F called the Fock representation, which is believed to be faithful. In this article we denote the image of the universal enveloping C^*-algebra of q-CCR in the Fock representation by beth _{n,q}. The question whether C^*-isomorphism beth _{n,q} simeq beth _{n,0} holds has been considered in the literature and proved for |q| < 0.44. In this article we show that beth _{n,q} simeq beth _{n,0} for |q| < 1.

A contrastive consistency semi-supervised left atrium segmentation model.

Accurate segmentation for the left atrium (LA) is a key process of clinical diagnosis and therapy for atrial fibrillation. In clinical, the semantic-level segmentation of LA consumes much time and labor. Although supervised deep learning methods can somewhat solve this problem, a high-efficient deep learning model requires abundant labeled data that is hard to acquire. Therefore, the research on automatic LA segmentation of leveraging unlabeled data is highly required. In this paper, we propose a semi-supervised LA segmentation framework including a segmentation model and a classification model. The segmentation model takes volumes from both labeled and unlabeled data as input and generates predictions of LAs. And then, a classification model maps these predictions to class-vectors for each input. Afterward, to leverage the class information, we construct a contrastive consistency loss function based on these class-vectors, so that the model can enlarge the discrepancy of the inter-class and compact the similarity of the intra-class for learning more distinguishable representation. Moreover, we set the class-vectors from the labeled data as references to the class-vectors from the unlabeled data to relieve the influence of the unreliable prediction for the unlabeled data. At last, we evaluate our semi-supervised LA segmentation framework on a public LA dataset using four universal metrics and compare it with recent state-of-the-art models. The proposed model achieves the best performance on all metrics with a Dice Score of 89.81%, Jaccard of 81.64%, 95% Hausdorff distance of 7.15mm, and Average Surface Distance of 1.82mm. The outstanding performance of the proposed framework shows that it may have a significant contribution to assisting the therapy of patients with atrial fibrillation. Code is available at: https://github.com/PerceptionComputingLab/SCC.

SPCS: a spatial pyramid convolutional shuffle module for YOLO to detect occluded object

In crowded scenes, one of the most important issues is that heavily overlapped objects are hardly distinguished from each other since most of their pixels are shared and the visible pixels of the occluded objects, which are used to represent their features, are limited. In this paper, a spatial pyramid convolutional shuffle (SPCS) module is proposed to extract refined information from the limited visible pixels of the occluded objects and generate distinguishable representations for the heavily overlapped objects. We adopt four convolutional kernels with different sizes and dilation rates at each location in the pyramid features and adjacently recombine their fused outputs spatially using a pixel shuffle module. In this way, four distinguishable instance predictions corresponding different convolutional kernels can be produced for each location in the pyramid feature. In addition, multiple convolutional operations with different kernel sizes and dilation rates at the same location can generate refined information for the corresponding regions, which is helpful to extract features for the occluded objects from their limited visible pixels. Extensive experimental results demonstrate that SPCS module can effectively boost the performance in crowded human detection. YOLO detector with SPCS module achieves 94.11% AP, 41.75% MR, 97.75% Recall on CrowdHuman, 93.04% AP, and 98.45% Recall on WiderPerson, which are the best compared with previous state-of-the-art models.

Generalized Whittaker Quotients of Schwartz Functions onG-Spaces

AbstractLet $G$ be a reductive group over a local field $F$ of characteristic zero. Let $X$ be a $G$-space. In this paper we study the existence of generalized Whittaker quotients for the space of Schwartz functions on $X$, considered as a representation of $G$. We show that the set of nilpotent elements of the dual space to the Lie algebra such that the corresponding generalized Whittaker quotient does not vanish contains the nilpotent part of the image of the moment map and lies in the closure of this image. This generalizes recent results of Prasad and Sakellaridis. Applying our theorems to symmetric pairs $(G,H)$ we show that there exists an infinite-dimensional $H$-distinguished representation of $G$ if and only if the real reductive group corresponding to the pair $(G,H)$ is non-compact. For quasi-split $G$ we also extend to the Archimedean case the theorem of Prasad stating that there exists a generic $H$-distinguished representation of $G$ if and only if the real reductive group corresponding to the pair $(G,H)$ is quasi-split. In the non-Archimedean case our result also gives rather sharp bounds on the wave-front sets of distinguished representations. Finally, we deduce a corollary on vanishing of period integrals of automorphic forms with certain Whittaker supports. This corollary, when combined with the restrictions on the Whittaker support of cuspidal automorphic representations proven by Gomez–Gourevitch–Sahi, implies many of the vanishing results on periods of automorphic forms proved by Ash–Ginzburg–Rallis.

AIR-Net: A novel multi-task learning method with auxiliary image reconstruction for predicting EGFR mutation status on CT images of NSCLC patients

Automated and accurate EGFR mutation status prediction using computed tomography (CT) imagery is of great value for tailoring optimal treatments to non-small cell lung cancer (NSCLC) patients. However, existing deep learning based methods usually adopt a single task learning strategy to design and train EGFR mutation status prediction models with limited training data, which may be insufficient to learn distinguishable representations for promoting prediction performance. In this paper, a novel multi-task learning method named AIR-Net is proposed to precisely predict EGFR mutation status on CT images. First, an auxiliary image reconstruction task is effectively integrated with EGFR mutation status prediction, aiming at providing extra supervision at the training phase. Particularly, we adequately employ multi-level information in a shared encoder to generate more comprehensive representations of tumors. Second, a powerful feature consistency loss is further introduced to constrain semantic consistency of original and reconstructed images, which contributes to enhanced image reconstruction and offers more effective regularization to AIR-Net during training. Performance analysis of AIR-Net indicates that auxiliary image reconstruction plays an essential role in identifying EGFR mutation status. Furthermore, extensive experimental results demonstrate that our method achieves favorable performance against other competitive prediction methods. All the results executed in this study suggest that the effectiveness and superiority of AIR-Net in precisely predicting EGFR mutation status of NSCLC.

Seizure prediction in scalp EEG based channel attention dual-input convolutional neural network

Epilepsy is one of the most common brain diseases, and seizures usually occur randomly. Accurately predicting seizures enable doctors and patients to carry out medical prevention timely. In seizure prediction studies, single-domain information input (time domain or frequency domain, Etc.) neglects some parts’ information from signals. In this paper, we propose a novel deep learning framework named channel attention dual-input convolutional neural network (CADCNN) to obtain the signal’s useful information fully. The spatial–temporal features extracted by short-time Fourier transform (STFT) are fed to the CADCNN, and the raw EEG signals are fed for further feature extraction. With the fusion of two inputs from different domains and the combination of channel attention, CADCNN can learn faithful and distinguishable representations of EEG signals and boost the temporal, spectrum, and spatial information utilization capability. We evaluate the proposed method using the Boston Children’s Hospital-MIT scalp EEG public datasets. Compared with other state-of-the-art methods, the sensitivity, false prediction rate, specificity, and AUC of our proposed method reach 97.1%, 0.029h, 95.6%, and 0.917, respectively, presenting better performance and higher prediction accuracy.