Multiclass Classification and Feature Selection Based on Least Squares Regression with Large Margin.

The corrugated layer structure bismuth has been successfully tailored into negative thermal expansion along c axis by size effect. Pair distribution function and extended X-ray absorption fine structure are combined to reveal the local structural distortion for nanosized bismuth. The comprehensive method to identify the local structure of nanomaterials can benefit the regulating and controlling of thermal expansion in nanodivices.

Ridge Regression (RR) is a classical method that is widely used in multiple regression analysis. However, traditional RR does not take the local geometric structure of data into consideration for discriminative learning and it is sensitive to outliers as it is based on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2}$</tex-math></inline-formula> -norm. To address this problem, this article proposes a novel method called Joint Sparse Locality Preserving Regression (JSLPR) for discriminative learning. JSLPR not only applies <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2,1}$</tex-math></inline-formula> -norm on both loss function and regularization term but also takes the local geometric structure of the data into consideration. The use of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2,1}$</tex-math></inline-formula> -norm can guarantee the robustness to outliers or noises and the joint sparsity for effective feature selection. Taking the local geometric structure into consideration can improve the performance of the feature extraction and selection method when the data lie on a manifold. To solve the optimization problem of JSLPR, an iterative algorithm is proposed and the convergence of the algorithm is also proven. Experiments on four famous face databases are conducted and the results show the merit of the proposed JSLPR on feature extraction and selection.

Unsupervised feature selection via discrete spectral clustering and feature weights

BackgroundIn recent years protein structure prediction methods using local structure information have shown promising improvements. The quality of new fold predictions has risen significantly and in fold recognition incorporation of local structure predictions led to improvements in the accuracy of results.We developed a local structure prediction method to be integrated into either fold recognition or new fold prediction methods. For each local sequence window of a protein sequence the method predicts probability estimates for the sequence to attain particular local structures from a set of predefined local structure candidates.The first step is to define a set of local structure representatives based on clustering recurrent local structures. In the second step a discriminative model is trained to predict the local structure representative given local sequence information.ResultsThe step of clustering local structures yields an average RMSD quantization error of 1.19 Å for 27 structural representatives (for a fragment length of 7 residues). In the prediction step the area under the ROC curve for detection of the 27 classes ranges from 0.68 to 0.88.ConclusionThe described method yields probability estimates for local protein structure candidates, giving signals for all kinds of local structure. These local structure predictions can be incorporated either into fold recognition algorithms to improve alignment quality and the overall prediction accuracy or into new fold prediction methods.

Identifying local useful information for attribute graph anomaly detection

The objective of an effective human-robot collaborative (HRC) task is the maximization of human-robot competencies while ensuring the user’s convenience. In photometrically challenging and unstructured HRC environments, data obtained from vision sensors often tend to get degraded due to illumination irregularities and spatio-temporal complexities. To extract useful and discriminative features from the data under such situations, locality-sensitive methods like locality preserving projections (LPPs) become quite useful as it captures the local geometric structure of the high-dimensional data. In LPP, the local structural information is encoded in the form of weight values between two samples in the higher-dimensional Euclidean space. The weight values are learned in a regular and continuous manner which only depends on the spatial distribution of the data. Moreover, because of its weight dependency solely on the Euclidean distance, improper weight values can occur frequently, as the Euclidean distance is susceptible to noise, outliers, and different types of geometrical transformations. This paper proposes an adaptive weight learning method to be utilized in the weight computation of LPP, which allows it to adaptively select and extract more discriminative features from the higher-dimensional input data while preserving the intrinsic structural information of the data. Additionally, to alleviate the issues with spatial dependency, the concept of bilateral filtering that incorporates the range weights from the feature space along with the similarity weight in the Euclidean space has been utilized here. This paper proposes an augmented version of adaptive spatial-feature kernel-guided bilateral filtering inspired LPP which addresses two of these basic and fundamental issues of the conventional LPP.

Using network-based information to facilitate information spreading is an essential task for spreading dynamics in complex networks. Focusing on degree correlated networks, we propose a preferential contact strategy based on the local network structure and local informed density to promote the information spreading. During the spreading process, an informed node will preferentially select a contact target among its neighbors, basing on their degrees or local informed densities. By extensively implementing numerical simulations in synthetic and empirical networks, we find that when only consider the local structure information, the convergence time of information spreading will be remarkably reduced if low-degree neighbors are favored as contact targets. Meanwhile, the minimum convergence time depends non-monotonically on degree-degree correlation, and a moderate correlation coefficient results in the most efficient information spreading. Incorporating the local informed density information into contact strategy, the convergence time of information spreading can be further reduced, and be minimized by an moderately preferential selection.

Over the years, many dimensionality reduction algorithms have been proposed for learning the structure of high dimensional data by linearly or non-linearly transforming it into a low-dimensional space. Some techniques can keep the local structure of data, while the others try to preserve the global structure. In this paper, we propose a linear dimensionality reduction technique that characterizes the local and global properties of data by firstly applying k-means algorithm on original data, and then finding the projection by simultaneously globally maximizing the between-cluster scatter matrix and locally minimizing the within-cluster scatter matrix, which actually keeps both local and global structure of data. Low complexity and structure preserving are two main advantages of the proposed technique. The experiments on both artificial and real data sets show the effectiveness and novelty of proposed algorithm in visualization and classification tasks.

In the Fishervoice (FSH) based framework, the mean supervectors of the speaker models are divided into several subvectors by mixture index. However, this division strategy cannot capture local acoustic class structure information among similar acoustic classes or discriminative information between different acoustic classes. In order to verify whether or not local structure information can help improve system performance, we develop five different speaker supervector segmentation methods. Experiments on NIST SRE08 prove that clustering similar acoustic classes together improves the system performance. In particular, the proposed method of equal size clustering achieves 5.1% relative decrease on EER compared to FSH1.

目的 基于计算机的胸腔X线影像疾病检测和分类目前存在误诊率高，准确率低的问题。本文在视觉Transformer（vision Transformer，ViT）预训练模型的基础上，通过迁移学习方法，实现胸腔X线影像辅助诊断，提高诊断准确率和效率。方法 选用带有卷积神经网络（convolutional neural network，CNN）的ViT模型，其在超大规模自然图像数据集中进行了预训练；通过微调模型结构，使用预训练的ViT模型参数初始化主干网络，并迁移至胸腔X线影像数据集中再次训练，实现疾病多标签分类。结果 在IU X-Ray数据集中对ViT迁移学习前、后模型平均AUC（area under ROC curve）得分进行对比分析实验。结果表明，预训练ViT模型平均AUC得分为0.774，与不使用迁移学习相比提升了0.208。并针对模型结构和数据预处理进行了消融实验，对ViT中的注意力机制进行可视化，进一步验证了模型有效性。最后使用Chest X-Ray14和CheXpert数据集训练微调后的ViT模型，平均AUC得分为0.839和0.806，与对比方法相比分别有0.014～0.031的提升。结论 与其他方法相比，ViT模型胸腔X线影像的多标签分类精确度更高，且迁移学习可以在降低训练成本的同时提升ViT模型的分类性能和泛化性。消融实验与模型可视化表明，包含CNN结构的ViT模型能重点关注有意义的区域，高效获取胸腔X线影像的视觉特征。;Objective The chest X-ray-relevant screening and diagnostic method is essential for radiology nowadays. Most of chest X-ray images interpretation is still restricted by clinical experience and challenged for misdiagnose and missed diagnoses. To detect and identify one or more potential diseases in images automatically，it is beneficial for improving diagnostic efficiency and accuracy using computer-based technique. Compared to natural images，multiple lesions are challenged to be detected and distinguished accurately in a single image because abnormal areas have a small proportion and complex representations in chest X-ray images. Current convolutional neural network（CNN）based deep learning models have been widely used in the context of medical imaging. The structure of the CNN convolution kernel has sensitive to local detail information，and it is possible to extract richer image features. However，the convolution kernel cannot be used to get global information，and the features-extracted are restricted of redundant information like its relevance of background， muscles，and bones. The model’s performance in multi-label classification tasks are affected to a certain extent. At present，the vision Transformer（ViT）model has achieved its priorities in computer vision-related tasks. The ViT can be used to capture information simultaneously and effectively for multiple regions of the entire image. However，it is required to use large-scale dataset training to achieve good performance. Due to some factors like patient privacy and manual annotate costs，the size of the chest X-ray image data set has been limited. To reduce the model's dependence on data scale and improve the performance of multi-label classification，we develop the CNN-based ViT pre-training model in terms of the transfer learning method for diagnosis-assisted of chest X-ray image and multi-label classification. Method The CNN-based ViT model is pre-trained on a huge scale ground truth dataset，and it is used to obtain the initial parameters of the model. The model structure is fine-tuned according to the features of chest X-ray dataset. A 1×1 convolution layer is used to convert the chest X-ray images channels between 1 to 3. The number of output nodes of the linear layer in the classifier is balanced from 1 000 to the number of chest X-ray classification labels，and the Sigmoid is used as an activation function. The parameters of the backbone network are initialized in terms of the pre-trained ViT model parameters，and it is trained in the chest X-ray dataset after that to complete multi-label classification. The experiment is configured of Python3. 7 and PyTorch1. 8 to construct the model and RTX3090 GPU for training. Stochastic gradient descent（SGD）optimizer，binary cross-entropy（BCE）loss function，an initial learning rate of 1E-3，the cosine annealing learning rate decay are used. For training，each image is scaled to a size of 512×512 pixels，and a 224×224 pixels area and it is then cropped in random as the model input，and data augmentation is performed randomly by some of the flipping，perspective transformation， shearing，translation，zooming，and changing brightness. For testing，the chest X-ray image is scaled to 256×256 pixels and center crop a 224×224 area to input the trained model. Result The experiment is performed on the IU X-Ray，which is a small-scale chest X-ray dataset. This model is evaluated in quantitative using the average of area under ROC curve （AUC）scores across all classification labels. The results show that the average AUC score of the pre-trained ViT model is 0. 774. The accuracy and training efficiency of the non-pre-trained ViT model is dropped significantly. The average AUC score is reached to 0. 566 only，which is 0. 208 lower. In addition，the attention mechanism heat map is generated based on the ViT model，which can strengthen the interpretability of the model. A series of ablation experiments are carried out for data augmentation，model structure，and batch size design. The fine-tuned ViT model is trained on the Chest-Ray14 and CheXpert dataset as well. The average AUC score is reached to 0. 839 and 0. 806，which is optimized by 0. 014 and 0. 031. Conclusion A pre-trained ViT model is used for the multi-label classification of chest X-ray images via transfer learning. The experimental results illustrate that the ViT has its stronger multi-label classification performance in chest Xray images，and its attention mechanism is beneficial for lesions precision-focused like the interior of the chest cavity and the heart. Transfer learning is potential to improve the classification performance and model generalization of the ViT in small-scale datasets，and the training cost is reduced greatly. Ablation experiments demonstrate that the incorporated model of CNN and Transformer has its priority beyond single-structure model. Data enhancement and the batch size cutting can improve the performance of the model，but smaller scale of batch is still interlinked to longer training span. To improve the model's ability，we predict that future research direction can be focused on the extraction for complex disease and highlevel semantic information，such as their small lesions，disease location，and severity.

Improved multiclass feature selection via list combination

Fault diagnosis plays a vital role in ensuring the operation safety of blast furnaces and improving the quality of molten iron in the ironmaking and steelmaking industry. The blast furnace ironmaking process (BFIP) is intrinsically nonlinear. To address the nonlinearity issue of BFIP, a novel fault diagnosis approach that combines the randomized method, local structure information, and Fisher discriminant analysis is proposed in this paper. Using a randomized feature map, the process data is first mapped onto a randomized explicit low‐dimensional feature space. Compared to kernel methods, explicit low‐dimensional random Fourier features considerably reduce the computational cost, particularly for real‐time fault diagnosis for a large training dataset or a large‐scale process. Additionally, the local structure information contained in the randomized low‐dimensional feature space is extracted. The fault diagnosis performance is improved through the exploration of the local structure of random Fourier features. Finally, the blast furnace iron‐marking process state is determined using Bayesian inference. Case studies on a real‐world BFIP are carried out to demonstrate the superior performance of the proposed method in comparison with other related methods.

Anomaly detection (AD) aims to distinguish the data points that are inconsistent with the overall pattern of the data. Recently, unsupervised anomaly detection methods have aroused huge attention. Among these methods, feature representation (FR) plays an important role, which can directly affect the performance of anomaly detection. Sparse representation (SR) can be regarded as one of matrix factorization (MF) methods, which is a powerful tool for FR. However, there are some limitations in the original SR. On the one hand, it just learns the shallow feature representations, which leads to the poor performance for anomaly detection. On the other hand, the local geometry structure information of data is ignored. To address these shortcomings, a graph regularized deep sparse representation (GRDSR) approach is proposed for unsupervised anomaly detection in this work. In GRDSR, a deep representation framework is first designed by extending the single layer MF to a multilayer MF for extracting hierarchical structure from the original data. Next, a graph regularization term is introduced to capture the intrinsic local geometric structure information of the original data during the process of FR, making the deep features preserve the neighborhood relationship well. Then, a L1-norm-based sparsity constraint is added to enhance the discriminant ability of the deep features. Finally, a reconstruction error is applied to distinguish anomalies. In order to demonstrate the effectiveness of the proposed approach, we conduct extensive experiments on ten datasets. Compared with the state-of-the-art methods, the proposed approach can achieve the best performance.

Prediction of the stage of cancer plays an important role in planning the course of treatment and has been largely reliant on imaging tools which do not capture molecular events that cause cancer progression. Gene-expression data-based analyses are able to identify these events, allowing RNA-sequence and microarray cancer data to be used for cancer analyses. Breast cancer is the most common cancer worldwide, and is classified into four stages - stages 1, 2, 3, and 4 [2]. While machine learning models have previously been explored to perform stage classification with limited success, multi-class stage classification has not had significant progress. There is a need for improved multi-class classification models, such as by investigating deep learning models. Gene-expression-based cancer data is characterised by the small size of available datasets, class imbalance, and high dimensionality. Class balancing methods must be applied to the dataset. Since all the genes are not necessary for stage prediction, retaining only the necessary genes can improve classification accuracy. The breast cancer samples are to be classified into 4 classes of stages 1 to 4. Invasive ductal carcinoma breast cancer samples are obtained from The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) datasets and combined. Two class balancing techniques are explored, synthetic minority oversampling technique (SMOTE) and SMOTE followed by random undersampling. A hybrid feature selection pipeline is proposed, with three pipelines explored involving combinations of filter and embedded feature selection methods: Pipeline 1 - minimum-redundancy maximum-relevancy (mRMR) and correlation feature selection (CFS), Pipeline 2 - mRMR, mutual information (MI) and CFS, and Pipeline 3 - mRMR and support vector machine-recursive feature elimination (SVM-RFE). The classification is done using deep learning models, namely deep neural network, convolutional neural network, recurrent neural network, a modified deep neural network, and an AutoKeras generated model. Classification performance post class-balancing and various feature selection techniques show marked improvement over classification prior to feature selection. The best multiclass classification was found to be by a deep neural network post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with a Cohen-Kappa score of 0.303 and a classification accuracy of 53.1%. For binary classification into early and late-stage cancer, the best performance is obtained by a modified deep neural network (DNN) post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with an accuracy of 81.0% and a Cohen-Kappa score (CKS) of 0.280. This pipeline also showed improved multiclass classification performance on neuroblastoma cancer data, with a best area under the receiver operating characteristic (auROC) curve score of 0.872, as compared to 0.71 obtained in previous work, an improvement of 22.81%. The results and analysis reveal that feature selection techniques play a vital role in gene-expression data-based classification, and the proposed hybrid feature selection pipeline improves classification performance. Multi-class classification is possible using deep learning models, though further improvement particularly in late-stage classification is necessary and should be explored further.

TC-Net: A joint learning framework based on CNN and vision transformer for multi-lesion medical images segmentation