Networks For Image Classification Research Articles

MultiFusionNet: multilayer multimodal fusion of deep neural networks for chest X-ray image classification

MultiFusionNet: multilayer multimodal fusion of deep neural networks for chest X-ray image classification

Global-local manifold embedding broad graph convolutional network for hyperspectral image classification

Graph convolutional neural networks (GCNs) with domain-specific feature aggregation capabilities have unique advantages in hyperspectral image (HSI) classification. However, current GCN-based approaches frequently encounter the issue of node characteristics being over-smoothed while aggregating in higher-order domains. Furthermore, GCN linear classifiers focus solely on sample separability and ignore the potential manifold information of graph features, resulting in a failure to fully investigate extracted features. To address these problems, we propose a global-local manifold embedding broad graph convolutional network (GLMBG) for HSI classification. In GLMBG, we designed two modules from feature extraction and classification perspectives: The graph convolutional edge feature fusion extractor (GEFF) and the broad classifier of global-local manifold embedding (BGLME). GEFF is designed to learn graph node and local edge features from HSI through GCN and recursive filtering, combining them in a weighted manner to construct fused graph features. BGLME is designed to replace traditional linear classifiers with broad learning classifiers through manifold regularized embedding, fully utilizing the global and local manifold discriminant information of graph node features. The combination of GEFF and BGLME effectively reduces over-smoothing of graph node features while maximizing the utilization of manifold discriminant information, hence improving model feature discriminative ability. Experimental evaluations of three commonly used hyperspectral datasets show that our method surpasses state-of-the-art methods.

Development and validation of a deep learning-based framework for automated lung CT segmentation and acute respiratory distress syndrome prediction: a multicenter cohort study

Acute respiratory distress syndrome (ARDS) is a life-threatening condition with a high incidence and mortality rate in intensive care unit (ICU) admissions. Early identification of patients at high risk for developing ARDS is crucial for timely intervention and improved clinical outcomes. However, the complex pathophysiology of ARDS makes early prediction challenging. This study aimed to develop an artificial intelligence (AI) model for automated lung lesion segmentation and early prediction of ARDS to facilitate timely intervention in the intensive care unit. A total of 928 ICU patients with chest computed tomography (CT) scans were included from November 2018 to November 2021at three centers in China. Patients were divided into a retrospective cohort for model development and internal validation, and three independent cohorts for external validation. A deep learning-based framework using the UNet Transformer (UNETR) model was developed to perform the segmentation of lung lesions and early prediction of ARDS. We employed various data augmentation techniques using the Medical Open Network for AI (MONAI) framework, enhancing the training sample diversity and improving the model's generalization capabilities. The performance of the deep learning-based framework was compared with a Densenet-based image classification network and evaluated in external and prospective validation cohorts. The segmentation performance was assessed using the Dice coefficient (DC), and the prediction performance was assessed using area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The contributions of different features to ARDS prediction were visualized using Shapley Explanation Plots. This study was registered with the China Clinical Trial Registration Centre (ChiCTR2200058700). The segmentation task using the deep learning framework achieved a DC of 0.734±0.137 in the validation set. For the prediction task, the deep learning-based framework achieved AUCs of 0.916 [0.858-0.961], 0.865 [0.774-0.945], 0.901 [0.835-0.955], and 0.876 [0.804-0.936] in the internal validation cohort, external validation cohort I, external validation cohort II, and prospective validation cohort, respectively. It outperformed the Densenet-based image classification network in terms of prediction accuracy. Moreover, the ARDS prediction model identified lung lesion features and clinical parameters such as C-reactive protein, albumin, bilirubin, platelet count, and age as significant contributors to ARDS prediction. The deep learning-based framework using the UNETR model demonstrated high accuracy and robustness in lung lesion segmentation and early ARDS prediction, and had good generalization ability and clinical applicability. This study was supported by grants from the Shanghai Renji Hospital Clinical Research Innovation and Cultivation Fund (RJPY-DZX-008) and Shanghai Science and Technology Development Funds (22YF1423300).

A Dual-Branch Fusion of a Graph Convolutional Network and a Convolutional Neural Network for Hyperspectral Image Classification.

Semi-supervised graph convolutional networks (SSGCNs) have been proven to be effective in hyperspectral image classification (HSIC). However, limited training data and spectral uncertainty restrict the classification performance, and the computational demands of a graph convolution network (GCN) present challenges for real-time applications. To overcome these issues, a dual-branch fusion of a GCN and convolutional neural network (DFGCN) is proposed for HSIC tasks. The GCN branch uses an adaptive multi-scale superpixel segmentation method to build fusion adjacency matrices at various scales, which improves the graph convolution efficiency and node representations. Additionally, a spectral feature enhancement module (SFEM) enhances the transmission of crucial channel information between the two graph convolutions. Meanwhile, the CNN branch uses a convolutional network with an attention mechanism to focus on detailed features of local areas. By combining the multi-scale superpixel features from the GCN branch and the local pixel features from the CNN branch, this method leverages complementary features to fully learn rich spatial-spectral information. Our experimental results demonstrate that the proposed method outperforms existing advanced approaches in terms of classification efficiency and accuracy across three benchmark data sets.

A Review of Transformer Networks in MRI Image Classification

The urgent need to improve the efficiency and precision of analyzing images in the field of magnetic resonance imaging (MRI) has led to the rise of transformer networks, as a groundbreaking solution. This review delves deeply into how transformer networksre used in classifying MRI images summarizing studies to showcase the progress made and challenges faced in this evolving area. With their abilities transformers excel at analyzing MRI images by highlighting both details and functional complexities while effectively capturing both global and local nuances using their advanced handling of long distance connections. Additionally their adaptability allows for processing of input data in sizes along with the capability to break down processing and analysis tasks. By providing researchers with an insight into transformers and their crucial role in enhancing medical imaging methods this research aims to lay the groundwork, for advancements in this critical field.

MonuNet: a high performance deep learning network for Kolkata heritage image classification

Kolkata, renowned as the City of Joy, boasts a rich tapestry of cultural heritage spanning centuries. Despite the significance of its architectural marvels, accessing comprehensive visual documentation of Kolkata's heritage sites remains a challenge. In online searches, limited imagery often fails to provide a detailed understanding of these historical landmarks. To address this gap, this paper introduces MonuNet, a high-performance deep-learning network specifically designed for the classification of heritage images from Kolkata. The development of MonuNet addresses the critical need for efficient and accurate identification of Kolkata's architectural marvels, which are significant tangible cultural heritages. The dataset used to train MonuNet is organized by heritage sites, each category within the dataset represents distinct sites. It includes images from 13 prominent heritage sites in Kolkata. For each of these sites, there are 50 images, making it a structured collection where each category (heritage site) is equally represented. The proposed network utilizes a unique architecture incorporating a Dense channel attention module and a Parallel-spatial channel attention module to capture intricate architectural details and spatial relationships within the images. Experimental evaluations demonstrate the superior performance of MonuNet in classifying Kolkata heritage images with an accuracy of 89%, Precision of 87.77%, and Recall of 86.61%. The successful deployment of MonuNet holds significant implications for cultural preservation, tourism enhancement, and urban planning in Kolkata, aligning with the United Nations Sustainable Development Goals (SDGs) for sustainable city development. By providing a robust tool for the automatic identification and classification of heritage images, MonuNet promises to enrich online repositories with detailed visual documentation, thereby enhancing accessibility to Kolkata's cultural heritage for researchers, tourists, and urban planners alike.Graphical

Optimizing Convolutional Neural Networks for Image Classification on Resource-Constrained Microcontroller Units

Running machine learning algorithms for image classification locally on small, cheap, and low-power microcontroller units (MCUs) has advantages in terms of bandwidth, inference time, energy, reliability, and privacy for different applications. Therefore, TinyML focuses on deploying neural networks on MCUs with random access memory sizes between 2 KB and 512 KB and read-only memory storage capacities between 32 KB and 2 MB. Models designed for high-end devices are usually ported to MCUs using model scaling factors provided by the model architecture’s designers. However, our analysis shows that this naive approach of substantially scaling down convolutional neural networks (CNNs) for image classification using such default scaling factors results in suboptimal performance. Consequently, in this paper we present a systematic strategy for efficiently scaling down CNN model architectures to run on MCUs. Moreover, we present our CNN Analyzer, a dashboard-based tool for determining optimal CNN model architecture scaling factors for the downscaling strategy by gaining layer-wise insights into the model architecture scaling factors that drive model size, peak memory, and inference time. Using our strategy, we were able to introduce additional new model architecture scaling factors for MobileNet v1, MobileNet v2, MobileNet v3, and ShuffleNet v2 and to optimize these model architectures. Our best model variation outperforms the MobileNet v1 version provided in the MLPerf Tiny Benchmark on the Visual Wake Words image classification task, reducing the model size by 20.5% while increasing the accuracy by 4.0%.

Effects of precise cardio sounds on the success rate of phonocardiography.

This work investigates whether inclusion of the low-frequency components of heart sounds can increase the accuracy, sensitivity and specificity of diagnosis of cardiovascular disorders. We standardized the measurement method to minimize changes in signal characteristics. We used the Continuous Wavelet Transform to analyze changing frequency characteristics over time and to allocate frequencies appropriately between the low-frequency and audible frequency bands. We used a Convolutional Neural Network (CNN) and deep-learning (DL) for image classification, and a CNN equipped with long short-term memory to enable sequential feature extraction. The accuracy of the learning model was validated using the PhysioNet 2016 CinC dataset, then we used our collected dataset to show that incorporating low-frequency components in the dataset increased the DL model's accuracy by 2% and sensitivity by 4%. Furthermore, the LSTM layer was 0.8% more accurate than the dense layer.

FSNet: A dual-domain network for few-shot image classification

Few-shot learning is a challenging task, that aims to learn and identify novel classes from a limited number of unseen labeled samples. Previous work has focused primarily on extracting features solely in the spatial domain of images. However, the compressed representation in the frequency domain which contains rich pattern information is a powerful tool in the field of signal processing. Combining the frequency and spatial domains to obtain richer information can effectively alleviate the overfitting problem. In this paper, we propose a dual-domain combined model called Frequency Space Net (FSNet), which preprocesses input images simultaneously in both the spatial and frequency domains, extracts spatial and frequency information through two feature extractors, and fuses them to a composite feature for image classification tasks. We start from a different view of frequency analysis, linking conventional average pooling to Discrete Cosine Transformation (DCT). We generalize the compression of the attention mechanism in the frequency domain. Consequently, we propose a novel Frequency Channel Spatial (FCS) attention mechanism. Extensive experiments demonstrate that frequency and spatial information are complementary in few-shot image classification, improving the performance of the model. Our method outperforms state-of-the-art approaches on miniImageNet and CUB.

Inference latency prediction for CNNs on heterogeneous mobile devices and ML frameworks

Due to the proliferation of inference tasks on mobile devices, state-of-the-art neural architectures are typically designed using Neural Architecture Search (NAS) to achieve good tradeoffs between machine learning accuracy and inference latency. While measuring inference latency of a huge set of candidate architectures during NAS is not feasible, latency prediction for mobile devices is challenging, because of hardware heterogeneity, optimizations applied by machine learning frameworks, and diversity of neural architectures. Motivated by these challenges, we first quantitatively assess the characteristics of neural architectures (specifically, convolutional neural networks for image classification), ML frameworks, and mobile devices that have significant effects on inference latency. Based on this assessment, we propose an operation-wise framework which addresses these challenges by developing operation-wise latency predictors and achieves high accuracy in end-to-end latency predictions, as shown by our comprehensive evaluations on multiple mobile devices using multicore CPUs and GPUs. To illustrate that our approach does not require expensive data collection, we also show that accurate predictions can be achieved on real-world neural architectures using only small amounts of profiling data.

Cnn-trans model: A parallel dual-branch network for fundus image classification

The existence of fundus diseases not only endangers people’s vision, but also brings serious economic burden to the society. Fundus images are an objective and standard basis for the diagnosis of fundus diseases. With the continuous advancement of computer science, deep learning methods dominated by convolutional neural networks (CNN) have been widely used in fundus image classification. However, the current CNN-based fundus image classification research still has a lot of room for improvement: CNN cannot effectively avoid the interference of repeated background information and has limited ability to model the whole world. In response to the above findings, this paper proposes the CNN-Trans model. The CNN-Trans model is a parallel dual-branch network, which is the two branches of CNN-LSTM and Vision Transform (ViT). The CNN-LSTM branch uses Xception after transfer learning. As the original feature extractor, LSTM is responsible for dealing with the gradient disappearance problem in neural network iterations before the classification head, and then introduces a new type of lightweight attention mechanism between Xception and LSTM: Coordinate Attention, so as to emphasize the key information related to classification and suppress the less useful repeated background information; while the self-attention mechanism in the ViT branch is not limited by local interactions, it can establish long-distance dependence on the target and extract global features. Finally, the concatenation (Concat) operation is used to fuse the features of the two branches. The local features extracted by the CNN-LSTM branch and the global features extracted by the ViT branch form complementary advantages. After feature fusion, more comprehensive image feature information is sent to the to the classification layer. Finally, after a large number of experimental tests and comparisons, the results show that: the CNN-Trans model achieved an accuracy of 80.68% on the fundus image classification task, and the CNN-Trans model has a classification that is comparable to the state-of-the-art methods. performance..

Detection of Alzheimer's disease using Otsu thresholding with tunicate swarm algorithm and deep belief network.

Introduction: Alzheimer's Disease (AD) is a degenerative brain disorder characterized by cognitive and memory dysfunctions. The early detection of AD is necessary to reduce the mortality rate through slowing down its progression. The prevention and detection of AD is the emerging research topic for many researchers. The structural Magnetic Resonance Imaging (sMRI) is an extensively used imaging technique in detection of AD, because it efficiently reflects the brain variations. Methods: Machine learning and deep learning models are widely applied on sMRI images for AD detection to accelerate the diagnosis process and to assist clinicians for timely treatment. In this article, an effective automated framework is implemented for early detection of AD. At first, the Region of Interest (RoI) is segmented from the acquired sMRI images by employing Otsu thresholding method with Tunicate Swarm Algorithm (TSA). The TSA finds the optimal segmentation threshold value for Otsu thresholding method. Then, the vectors are extracted from the RoI by applying Local Binary Pattern (LBP) and Local Directional Pattern variance (LDPv) descriptors. At last, the extracted vectors are passed to Deep Belief Networks (DBN) for image classification. Results and Discussion: The proposed framework achieves supreme classification accuracy of 99.80% and 99.92% on the Alzheimer's Disease Neuroimaging Initiative (ADNI) and Australian Imaging, Biomarker and Lifestyle flagship work of ageing (AIBL) datasets, which is higher than the conventional detection models.

Structural prior-driven feature extraction with gradient-momentum combined optimization for convolutional neural network image classification

Recent image classification efforts have achieved certain success by incorporating prior information such as labels and logical rules to learn discriminative features. However, these methods overlook the variability of features, resulting in feature inconsistency and fluctuations in model parameter updates, which further contribute to decreased image classification accuracy and model instability. To address this issue, this paper proposes a novel method combining structural prior-driven feature extraction with gradient-momentum (SPGM), from the perspectives of consistent feature learning and precise parameter updates, to enhance the accuracy and stability of image classification. Specifically, SPGM leverages a structural prior-driven feature extraction (SPFE) approach to calculate gradients of multi-level features and original images to construct structural information, which is then transformed into prior knowledge to drive the network to learn features consistent with the original images. Additionally, an optimization strategy integrating gradients and momentum (GMO) is introduced, dynamically adjusting the direction and step size of parameter updates based on the angle and norm of the sum of gradients and momentum, enabling precise model parameter updates. Extensive experiments on CIFAR10 and CIFAR100 datasets demonstrate that the SPGM method significantly reduces the top-1 error rate in image classification, enhances the classification performance, and outperforms state-of-the-art methods.

Multimodal Quanvolutional and Convolutional Neural Networks for Multi-Class Image Classification

By utilizing hybrid quantum–classical neural networks (HNNs), this research aims to enhance the efficiency of image classification tasks. HNNs allow us to utilize quantum computing to solve machine learning problems, which can be highly power-efficient and provide significant computation speedup compared to classical operations. This is particularly relevant in sustainable applications where reducing computational resources and energy consumption is crucial. This study explores the feasibility of a novel architecture by leveraging quantum devices as the first layer of the neural network, which proved to be useful for scaling HNNs’ training process. Understanding the role of quanvolutional operations and how they interact with classical neural networks can lead to optimized model architectures that are more efficient and effective for image classification tasks. This research investigates the performance of HNNs across different datasets, including CIFAR100 and Satellite Images of Hurricane Damage by evaluating the performance of HNNs on these datasets in comparison with the performance of reference classical models. By evaluating the scalability of HNNs on diverse datasets, the study provides insights into their applicability across various real-world scenarios, which is essential for building sustainable machine learning solutions that can adapt to different environments. Leveraging transfer learning techniques with pre-trained models such as ResNet, EfficientNet, and VGG16 demonstrates the potential for HNNs to benefit from existing knowledge in classical neural networks. This approach can significantly reduce the computational cost of training HNNs from scratch while still achieving competitive performance. The feasibility study conducted in this research assesses the practicality and viability of deploying HNNs for real-world image classification tasks. By comparing the performance of HNNs with classical reference models like ResNet, EfficientNet, and VGG-16, this study provides evidence of the potential advantages of HNNs in certain scenarios. Overall, the findings of this research contribute to advancing sustainable applications of machine learning by proposing novel techniques, optimizing model architectures, and demonstrating the feasibility of adopting HNNs for real-world image classification problems. These insights can inform the development of more efficient and environmentally friendly machine learning solutions.

Automatic diagnosis of pediatric high myopia via Attention-based Patch Residual Shrinkage network

Fundus images can be obtained non-invasively and be adopted to monitor the follow-up on various fundus diseases, such as high myopia. Therefore, the use of fundus images for the early screening of eye diseases has principal clinical significance. However, AI-based medical research continues to face two main challenges, lack of prior-knowledge guidance and complex fundus information. In this paper, we propose a pediatric high myopia classification model, the Attention-based Patch Residual Shrinkage network (APRSnet), which facilitates early clinical diagnosis by evaluating the importance of different image features. A collection of 2492 high-resolution fundus images of children, including 768 images of high myopia and 1724 images of non-high myopia, is fed to APRSnet as the training dataset. To better understand how different features impact the classification, APRSnet is tested on multiple feature-enhanced fusion datasets. The result shows that, by removing irrelevant information in fundus images, APRSnet achieves an accuracy of 0.959 and an F1 score of 0.946 and outperforms all classic image classification networks we compared with. We also testify that, in fundus images, gradient information benefits severity classification the most since it helps with model convergence more than luminance and texture information.

Efficiency in constraint: A comparative analysis review of FCN and DeepLab models on small-scale datasets

Semantic segmentation, crucial in computer vision, differentiates objects within images and finds applications in autonomous vehicles, medical imaging, and assistive technology. It typically employs neural networks for pixel-wise image classification. Key advancements in this field are attributed to Fully Convolutional Networks (FCN) and DeepLab models, known for their high performance with extensive datasets. However, the challenge arises when these models are applied to smaller datasets. Our research presents a concise overview of seminal work in machine learning and semantic segmentation, followed by an exploration of FCN and DeepLab architectures. The study primarily focuses on evaluating these models efficacy on a smaller dataset. Results, summarized in tables and figures, indicate that FCN-16 outperforms others in limited-data scenarios, while DeepLab shows reduced effectiveness. This finding is significant for applications with constrained data resources, providing a direction for future research in semantic segmentation under such conditions.

Enhanced Gradient for Differentiable Architecture Search.

In recent years, neural architecture search (NAS) methods have been proposed for the automatic generation of task-oriented network architecture in image classification. However, the architectures obtained by existing NAS approaches are optimized only for classification performance and do not adapt to devices with limited computational resources. To address this challenge, we propose a neural network architecture search algorithm aiming to simultaneously improve the network performance and reduce the network complexity. The proposed framework automatically builds the network architecture at two stages: block-level search and network-level search. At the stage of block-level search, a gradient-based relaxation method is proposed, using an enhanced gradient to design high-performance and low-complexity blocks. At the stage of network-level search, an evolutionary multiobjective algorithm is utilized to complete the automatic design from blocks to the target network. The experimental results demonstrate that our method outperforms all evaluated hand-crafted networks in image classification, with an error rate of 3.18% on Canadian Institute for Advanced Research (CIFAR10) and an error rate of 19.16% on CIFAR100, both at network parameter size less than 1 M. Obviously, compared with other NAS methods, our method offers a tremendous reduction in designed network architecture parameters.

Feature Consistency-Based Prototype Network for Open-Set Hyperspectral Image Classification.

Hyperspectral image (HSI) classification methods have made great progress in recent years. However, most of these methods are rooted in the closed-set assumption that the class distribution in the training and testing stages is consistent, which cannot handle the unknown class in open-world scenes. In this work, we propose a feature consistency-based prototype network (FCPN) for open-set HSI classification, which is composed of three steps. First, a three-layer convolutional network is designed to extract the discriminative features, where a contrastive clustering module is introduced to enhance the discrimination. Then, the extracted features are used to construct a scalable prototype set. Finally, a prototype-guided open-set module (POSM) is proposed to identify the known samples and unknown samples. Extensive experiments reveal that our method achieves remarkable classification performance over other state-of-the-art classification techniques.

Enhancing fire safety through IoT-enabled flame detection systems: A cost-effective and scalable approach

The Internet of Things (IoT), which connects and automates numerous systems and gadgets, has completely changed how we live and work. One such application of IoT technology is in fire detection systems, which can help prevent and mitigate the devastating effects of fires on different types of facilities. The research presents a n IoT architecture for a fire detection system using small, low-cost cameras to collect surveillance feeds from large buildings. The data is uploaded to the cloud, where a Machine Learning algorithm detects fires in digital images. The proposed architecture consists of cameras, cloud, and clients, using an inexpensive camera for surveillance feeds and a convolutional neural network for image classification based on large datasets. However, the architecture's cloud component processes surveillance feeds and runs a Machine Learning (ML) model, utilizing computing resources for real-time data processing and continuous training for improved accuracy. Clients can subscribe to the data from the cloud and receive alerts in real-time when the ML model detects a fire in the surveillance feeds. There are significant benefits in comparing the proposed design to conventional fire detection systems. First and foremost, it is economical since the cameras used are compact, affordable, and simple to install around the building without the need for elaborate wiring or infrastructure. Secondly, it is scalable, as the cloud provides the necessary computing resources and storage capacity to handle large amounts of data, making it possible to monitor large structures with many cameras.

Multi-branch CNN and grouping cascade attention for medical image classification

Visual Transformers(ViT) have made remarkable achievements in the field of medical image analysis. However, ViT-based methods have poor classification results on some small-scale medical image classification datasets. Meanwhile, many ViT-based models sacrifice computational cost for superior performance, which is a great challenge in practical clinical applications. In this paper, we propose an efficient medical image classification network based on an alternating mixture of CNN and Transformer tandem, which is called Eff-CTNet. Specifically, the existing ViT-based method still mainly relies on multi-head self-attention (MHSA). Among them, the attention maps of MHSA are highly similar, which leads to computational redundancy. Therefore, we propose a group cascade attention (GCA) module to split the feature maps, which are provided to different attention heads to further improves the diversity of attention and reduce the computational cost. In addition, we propose an efficient CNN (EC) module to enhance the ability of the model and extract the local detail information in medical images. Finally, we connect them and design an efficient hybrid medical image classification network, namely Eff-CTNet. Extensive experimental results show that our Eff-CTNet achieves advanced classification performance with less computational cost on three public medical image classification datasets.