Multimodel Feature Reinforcement Framework Using Moore-Penrose Inverse for Big Data Analysis

Fast Transfer Learning Method Using Random Layer Freezing and Feature Refinement Strategy

Recently, Moore-Penrose inverse (MPI)-based parameter fine-tuning of fully connected (FC) layers in pretrained deep convolutional neural networks (DCNNs) has emerged within the inductive transfer learning (ITL) paradigm. However, this approach has not gained significant traction in practical applications due to its stringent computational requirements. This work addresses this issue through a novel fast retraining strategy that enhances applicability of the MPI-based ITL. Specifically, during each retraining epoch, a random layer freezing protocol is utilized to manage the number of layers undergoing feature refinement. Additionally, this work incorporates an MPI-based approach for refining the trainable parameters of FC layers under batch processing, contributing to expedited convergence. Extensive experiments on several ImageNet pretrained benchmark DCNNs demonstrate that the proposed ITL achieves competitive performance with excellent convergence speed compared to conventional ITL methods. For instance, the proposed strategy converges nearly 1.5 times faster than retraining the ImageNet pretrained ResNet-50 using stochastic gradient descent with momentum (SGDM).

Multimodal Moore-Penrose Inverse-Based Recomputation Framework for Big Data Analysis

Most multilayer Moore-Penrose inverse (MPI)-based neural networks, such as deep random vector functional link (RVFL), are structured with two separate stages: unsupervised feature encoding and supervised pattern classification. Once the unsupervised learning is finished, the latent encoding is fixed without supervised fine-tuning. However, in complex tasks such as handling the ImageNet dataset, there are often many more clues that can be directly encoded, while unsupervised learning, by definition, cannot know exactly what is useful for a certain task. There is a need to retrain the latent space representations in the supervised pattern classification stage to learn some clues that unsupervised learning has not yet been learned. In particular, the residual error in the output layer is pulled back to each hidden layer, and the parameters of the hidden layers are recalculated with MPI for more robust representations. In this article, a recomputation-based multilayer network using Moore-Penrose inverse (RML-MP) is developed. A sparse RML-MP (SRML-MP) model to boost the performance of RML-MP is then proposed. The experimental results with varying training samples (from 3k to 1.8 million) show that the proposed models provide higher Top-1 testing accuracy than most representation learning algorithms. For reproducibility, the source codes are available at https://github.com/W1AE/Retraining.

Dynamic Moore-Penrose Inversion With Unknown Derivatives: Gradient Neural Network Approach

Finding dynamic Moore-Penrose inverses (DMPIs) in real-time is a challenging problem due to the time-varying nature of the inverse. Traditional numerical methods for static Moore-Penrose inverse are not efficient for calculating DMPIs and are restricted by serial processing. The current state-of-the-art method for finding DMPIs is called the zeroing neural network (ZNN) method, which requires that the time derivative of the associated matrix is available all the time during the solution process. However, in practice, the time derivative of the associated dynamic matrix may not be available in a real-time manner or be subject to noises caused by differentiators. In this article, we propose a novel gradient-based neural network (GNN) method for computing DMPIs, which does not need the time derivative of the associated dynamic matrix. In particular, the neural state matrix of the proposed GNN converges to the theoretical DMPI in finite time. The finite-time convergence is kept by simply setting a large parameter when there are additive noises in the implementation of the GNN model. Simulation results demonstrate the efficacy and superiority of the proposed GNN method.

Deep fusion of gray level co-occurrence matrices for lung nodule classification

Lung cancer is a serious threat to human health, with millions dying because of its late diagnosis. The computerized tomography (CT) scan of the chest is an efficient method for early detection and classification of lung nodules. The requirement for high accuracy in analyzing CT scan images is a significant challenge in detecting and classifying lung cancer. In this paper, a new deep fusion structure based on the long short-term memory (LSTM) has been introduced, which is applied to the texture features computed from lung nodules through new volumetric grey-level-co-occurrence-matrices (GLCMs), classifying the nodules into benign, malignant, and ambiguous. Also, an improved Otsu segmentation method combined with the water strider optimization algorithm (WSA) is proposed to detect the lung nodules. WSA-Otsu thresholding can overcome the fixed thresholds and time requirement restrictions in previous thresholding methods. Extended experiments are used to assess this fusion structure by considering 2D-GLCM based on 2D-slices and approximating the proposed 3D-GLCM computations based on volumetric 2.5D-GLCMs. The proposed methods are trained and assessed through the LIDC-IDRI dataset. The accuracy, sensitivity, and specificity obtained for 2D-GLCM fusion are 94.4%, 91.6%, and 95.8%, respectively. For 2.5D-GLCM fusion, the accuracy, sensitivity, and specificity are 97.33%, 96%, and 98%, respectively. For 3D-GLCM, the accuracy, sensitivity, and specificity of the proposed fusion structure reached 98.7%, 98%, and 99%, respectively, outperforming most state-of-the-art counterparts. The results and analysis also indicate that the WSA-Otsu method requires a shorter execution time and yields a more accurate thresholding process.

Hybrid Deep Feature Generation for Appropriate Face Mask Use Detection

Mask usage is one of the most important precautions to limit the spread of COVID-19. Therefore, hygiene rules enforce the correct use of face coverings. Automated mask usage classification might be used to improve compliance monitoring. This study deals with the problem of inappropriate mask use. To address that problem, 2075 face mask usage images were collected. The individual images were labeled as either mask, no masked, or improper mask. Based on these labels, the following three cases were created: Case 1: mask versus no mask versus improper mask, Case 2: mask versus no mask + improper mask, and Case 3: mask versus no mask. This data was used to train and test a hybrid deep feature-based masked face classification model. The presented method comprises of three primary stages: (i) pre-trained ResNet101 and DenseNet201 were used as feature generators; each of these generators extracted 1000 features from an image; (ii) the most discriminative features were selected using an improved RelieF selector; and (iii) the chosen features were used to train and test a support vector machine classifier. That resulting model attained 95.95%, 97.49%, and 100.0% classification accuracy rates on Case 1, Case 2, and Case 3, respectively. Having achieved these high accuracy values indicates that the proposed model is fit for a practical trial to detect appropriate face mask use in real time.

Automatic COVID-19 Detection Using Exemplar Hybrid Deep Features with X-ray Images

COVID-19 and pneumonia detection using medical images is a topic of immense interest in medical and healthcare research. Various advanced medical imaging and machine learning techniques have been presented to detect these respiratory disorders accurately. In this work, we have proposed a novel COVID-19 detection system using an exemplar and hybrid fused deep feature generator with X-ray images. The proposed Exemplar COVID-19FclNet9 comprises three basic steps: exemplar deep feature generation, iterative feature selection and classification. The novelty of this work is the feature extraction using three pre-trained convolutional neural networks (CNNs) in the presented feature extraction phase. The common aspects of these pre-trained CNNs are that they have three fully connected layers, and these networks are AlexNet, VGG16 and VGG19. The fully connected layer of these networks is used to generate deep features using an exemplar structure, and a nine-feature generation method is obtained. The loss values of these feature extractors are computed, and the best three extractors are selected. The features of the top three fully connected features are merged. An iterative selector is used to select the most informative features. The chosen features are classified using a support vector machine (SVM) classifier. The proposed COVID-19FclNet9 applied nine deep feature extraction methods by using three deep networks together. The most appropriate deep feature generation model selection and iterative feature selection have been employed to utilise their advantages together. By using these techniques, the image classification ability of the used three deep networks has been improved. The presented model is developed using four X-ray image corpora (DB1, DB2, DB3 and DB4) with two, three and four classes. The proposed Exemplar COVID-19FclNet9 achieved a classification accuracy of 97.60%, 89.96%, 98.84% and 99.64% using the SVM classifier with 10-fold cross-validation for four datasets, respectively. Our developed Exemplar COVID-19FclNet9 model has achieved high classification accuracy for all four databases and may be deployed for clinical application.