Conventional Convolutional Neural Network Research Articles

SkinSwinViT: A Lightweight Transformer-Based Method for Multiclass Skin Lesion Classification with Enhanced Generalization Capabilities

In recent decades, skin cancer has emerged as a significant global health concern, demanding timely detection and effective therapeutic interventions. Automated image classification via computational algorithms holds substantial promise in significantly improving the efficacy of clinical diagnoses. This study is committed to mitigating the challenge of diagnostic accuracy in the classification of multiclass skin lesions. This endeavor is inherently formidable owing to the resemblances among various lesions and the constraints associated with extracting precise global and local image features within diverse dimensional spaces using conventional convolutional neural network methodologies. Consequently, this study introduces the SkinSwinViT methodology for skin lesion classification, a pioneering model grounded in the Swin Transformer framework featuring a global attention mechanism. Leveraging the inherent cross-window attention mechanism within the Swin Transformer architecture, the model adeptly captures local features and interdependencies within skin lesion images while additionally incorporating a global self-attention mechanism to discern overarching features and contextual information effectively. The evaluation of the model’s performance involved the ISIC2018 challenge dataset. Furthermore, data augmentation techniques augmented training dataset size and enhanced model performance. Experimental results highlight the superiority of the SkinSwinViT method, achieving notable metrics of accuracy, recall, precision, specificity, and F1 score at 97.88%, 97.55%, 97.83%, 99.36%, and 97.79%, respectively.

Assessing robustness to adversarial attacks in attention-based networks: Case of EEG-based motor imagery classification

The classification of motor imagery (MI) using Electroencephalography (EEG) plays a pivotal role in facilitating communication for individuals with physical limitations through Brain-Computer Interface (BCI) systems. Recent strides in Attention-Based Networks (ATN) have showcased remarkable performance in EEG signal classification, presenting a promising alternative to conventional Convolutional Neural Networks (CNNs). However, while CNNs have been extensively analyzed for their resilience against adversarial attacks, the susceptibility of ATNs in comparable scenarios remains largely unexplored. This paper aims to fill this gap by investigating the robustness of ATNs in adversarial contexts. We propose a high-performing attention-based deep learning model specifically designed for classifying Motor Imagery (MI) brain signals extracted from EEG data. Subsequently, we conduct a thorough series of experiments to assess various attack strategies targeting ATNs employed in EEG-based BCI tasks. Our analysis utilizes the widely recognized BCI Competition 2a dataset to demonstrate the effectiveness of attention mechanisms in BCI endeavors. Despite achieving commendable classification results in terms of accuracy (87.15%) and kappa score (0.8287), our findings reveal the vulnerability of attention-based models to adversarial manipulations (accuracy: 9.07%, kappa score: -0.21), highlighting the imperative for bolstering the robustness of attention architectures for EEG classification tasks.

Identification of rheumatoid arthritis using bone image features with radial basis function neural networks

AbstractRheumatoid arthritis (RA) leads to the destruction, deformation, and loss of function and causes joint damages to 85% of patients. The detection of bone mineral density from traditional x‐ray images consumes more time and it is observer dependent which decreases the evaluation performance when RA is in its early stage. Therefore, it is necessary to develop an observer‐independent computer‐aided automatic analysis system for evaluating JS narrowing. An efficient RA detection system based on feed‐forward neural network is proposed in this article. Initially the dataset is preprocessed to remove blur, redundant data and nonlinearities in RA images using wiener filter in less computation time. The edge boundaries are detected using nonlinear partial differential equation as the texture features of bones has great impact on the system accuracy. The detection strategy is implemented by optimized gray‐level co‐occurrence matrix based on bone image features. The selected features are optimized by genetic algorithm to improve the classification performance. The images are then classified as inflamed and noninflamed by radial basis function neural network. The effectiveness of the proposed classification approach is verified in terms of accuracy, sensitivity, and specificity and compared with the conventional convolutional neural network, artificial neural network, and support vector machine classifiers. Among the several existing techniques, our method found to be much effective with a maximum accuracy of 98.5% as some novel approaches are proposed and also tends to possess more compatible than the present system. Our computerized rheumatoid arteries detection approach will give more precise and flawless consistency rate.

Comparative Analysis of Vision Transformers and Conventional Convolutional Neural Networks in Detecting Referable Diabetic Retinopathy

ObjectiveVision Transformers (ViTs) have shown promising performance in various classification tasks previously dominated by Convolutional Neural Networks (CNNs). However, the performance of ViTs in referable Diabetic Retinopathy (DR) detection is relatively underexplored. In this study, using retinal photographs, we evaluated the comparative performances of ViTs and CNNs on detection of referable DR. DesignRetrospective study ParticipantsA total of 48,269 retinal images from the open-source Kaggle DR detection dataset, the Messidor-1 dataset and the Singapore Epidemiology of Eye Diseases (SEED) study were included. MethodUsing 41,614 retinal photographs from the Kaggle dataset, we developed five CNN (VGG19, ResNet50, InceptionV3, DenseNet201 and EfficientNetV2S) and four ViTs models (VAN_small, CrossViT_small, ViT_small and SWIN_tiny) for the detection of referable DR. We defined the presence of referable DR as eyes with moderate or worse DR. The comparative performance of all nine models was evaluated in the Kaggle internal test dataset (with 1,045 study eyes), and in two external test sets, the SEED study (5,455 study eyes) and the Messidor-1 (1,200 study eyes). Main Outcome MeasuresArea Under Receiver Operative Curve (AUC), Specificity and Sensitvity. ResultsAmong all models, the SWIN transformer displayed the highest area under the curve (AUC) of 95.7% on the internal test set, significantly outperforming the CNN models (all P<0.001). The same observation was confirmed in the external test sets, with the SWIN transformer achieving AUC of 97.3% in SEED and 96.3% in Messidor-1. When specificity level was fixed at 80% for the internal test, the SWIN transformer achieved the highest sensitivity of 94.4%, significantly better than all the CNN models (sensitivity levels ranging between 76.3% and 83.8%; all P <0.001). This trend was also consistently observed in both external test sets. ConclusionOur findings demonstrate that ViTs provide superior performance over CNNs in detecting referable DR from retinal photographs. These results point to the potential of utilizing ViT models to improve and optimize retinal photo-based deep learning for referable DR detection.

Development of A Palm-Vein Recognition System for Identification and Verification Systems using Enhanced Convolutional Neural Network

Biometric authentication systems have gained significant attention in access control applications due to their ability to provide enhanced security and convenience. Among various biometric modalities, palm-vein recognition has emerged as a promising approach, offering high accuracy, reliability, and resistance to forgery. However, existing palm-vein recognition systems often face challenges in implementation costs, computational efficiency, and performance limitations. This research aimed to develop an enhanced palm-vein recognition system for access control applications by optimizing a Convolutional Neural Network (CNN) architecture. A palm-vein dataset comprising 1000 images from 200 LAUTECH students was acquired, with 5 images per individual. The dataset was split into 700 training images and 300 testing images. The acquired images were pre-processed for quality enhancement and region of interest extraction. A Gravitational Search Algorithm (GSA) optimized CNN (GSA-CNN) was then employed to extract deep features from the pre-processed images, which were classified using a SoftMax layer. Experimental results revealed that the CNN technique achieved a specificity, sensitivity, False Positive Rate (FPR), accuracy of 74.60%, 79.89%, 25.40%, 77.67% at 117.52 seconds, respectively. In contrast, the proposed GSA-CNN technique demonstrated superior performance, achieving a specificity, sensitivity, FPR, accuracy of 92.06%, 92.53%, 7.94%, 92.33% at 97.14 seconds, respectively. The GSA-CNN system outperformed the conventional CNN approach in terms of accuracy, specificity, sensitivity, FPR, and processing time, demonstrating its potential for reliable and efficient palm-vein recognition in access control applications. The findings have significant implications for developing robust and secure access control systems, contributing to enhanced privacy and security across various domains.

CDAM-Net: Channel shuffle dual attention based multi-scale CNN for efficient glaucoma detection using fundus images

Glaucoma is a typical eye disorder that induces damage to the optic nerve due to increased intraocular pressure. It ultimately leads to partial or complete blindness without clinical reversal. Hence, it is of utmost importance to screen and detect glaucoma at an early stage. Most of the earlier glaucoma diagnosis methods rely on manual feature engineering, which is time-consuming and requires domain experts. Although recent methods, particularly, convolutional neural networks (CNNs) facilitate learning high-level feature representations from fundus images, they need extensive parameters and pose overfitting issues due to insufficient training samples. Further, conventional CNNs often ignore the minute changes in the lesion region. To overcome these issues, we propose a lightweight multi-scale CNN architecture called as CDAM-Net for effective glaucoma identification from retinal fundus images. Additionally, we introduce an attention module called channel shuffle dual attention (CSDA), comprising of a channel attention block, a spatial attention block and a channel shuffle unit, to focus on important regions in the fundus images, thereby extracting class-specific features. The CDAM-Net mainly consists of multi-scale feature representation (MFR) blocks that enable the extraction of multi-scale features from fundus images. Each MFR block is followed by a CSDA module, which further helps enrich the feature representation. The CDAM-Net is evaluated on a retinal fundus image (RFI) dataset containing 1426 fundus images (837 glaucoma and 589 normal), and the results indicate that CDAM-Net yields promising classification performance compared to existing techniques. Also, ablation studies are carried out to test the effectiveness of each component of the CDAM-Net.

Vein segmentation and visualization of upper and lower extremities using convolution neural network.

The study focused on developing a reliable real-time venous localization, identification, and visualization framework based upon deep learning (DL) self-parametrized Convolution Neural Network (CNN) algorithm for segmentation of the venous map for both lower and upper limb dataset acquired under unconstrained conditions using near-infrared (NIR) imaging setup, specifically to assist vascular surgeons during venipuncture, vascular surgeries, or Chronic Venous Disease (CVD) treatments. A portable image acquisition setup has been designed to collect venous data (upper and lower extremities) from 72 subjects. A manually annotated image dataset was used to train and compare the performance of existing well-known CNN-based architectures such as ResNet and VGGNet with self-parameterized U-Net, improving automated vein segmentation and visualization. Experimental results indicated that self-parameterized U-Net performs better at segmenting the unconstrained dataset in comparison with conventional CNN feature-based learning models, with a Dice score of 0.58 and displaying 96.7 % accuracy for real-time vein visualization, making it appropriate to locate veins in real-time under unconstrained conditions. Self-parameterized U-Net for vein segmentation and visualization has the potential to reduce risks associated with traditional venipuncture or CVD treatments by outperforming conventional CNN architectures, providing vascular assistance, and improving patient care and treatment outcomes.

A modified vision transformer architecture with scratch learning capabilities for effective fire detection

Fire is considered to be one of the major influencing factors that cause fatalities, property damage, and economic, and ecological disruption. To perform the early detection of fire using data from vision sensors and prevent or reduce the damage that fires cause, deep learning models have been widely adopted to overcome the limitations of conventional methods. However, mainstream convolutional neural network (CNN) models have limited generalisation abilities in unseen scenarios and struggle to obtain a good trade-off among the accuracy, inference speed, and model size. Currently, vision transformers (ViT) outperform conventional CNN models; however, they are computationally expensive and required more data for training. They provide a limited performance for small and medium-sized datasets, which are very common in the fire scene classification domain. In this work, we employ a novel ViT architecture by combining shifted patch tokenisation and local self-attention modules for efficient fire scene classification and enable the model to learn from scratch even on small and medium-sized datasets. Furthermore, to make the model suitable for real-time inferencing, we modify the transformer encoder and eventually achieve a reduced number of floating-point operations and a reduced model size. Additionally, in this work, a medium-scale fire dataset is developed that contains complex real-world scenarios. Our model is assessed on three benchmark and a self-created datasets using several evaluation metrics, including a novel cross-corpse evaluation metric, as well as a robustness evaluation metric. The experimental results indicate that our model achieved an overwhelmingly better performance compared to existing methods in terms of the accuracy and model complexity.

Action Recognition of Taekwondo Unit Actions Using Action Images Constructed with Time-Warped Motion Profiles.

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint-motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint-motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, a precision of 0.983, a recall of 0.982, and an F1 score of 0.982. These results underscore this time-warping technique's contribution to enhancing feature representation, as well as the proposed method's scalability and effectiveness in recognizing Taekwondo unit actions.

In-Season Cotton Yield Prediction with Scale-Aware Convolutional Neural Network Models and Unmanned Aerial Vehicle RGB Imagery.

In the pursuit of sustainable agriculture, efficient water management remains crucial, with growers relying on advanced techniques for informed decision-making. Cotton yield prediction, a critical aspect of agricultural planning, benefits from cutting-edge technologies. However, traditional methods often struggle to capture the nuanced complexities of crop health and growth. This study introduces a novel approach to cotton yield prediction, leveraging the synergy between Unmanned Aerial Vehicles (UAVs) and scale-aware convolutional neural networks (CNNs). The proposed model seeks to harness the spatiotemporal dynamics inherent in high-resolution UAV imagery to improve the accuracy of the cotton yield prediction. The CNN component adeptly extracts spatial features from UAV-derived imagery, capturing intricate details related to crop health and growth, modeling temporal dependencies, and facilitating the recognition of trends and patterns over time. Research experiments were carried out in a cotton field at the USDA-ARS Cropping Systems Research Laboratory (CSRL) in Lubbock, Texas, with three replications evaluating four irrigation treatments (rainfed, full irrigation, percent deficit of full irrigation, and time delay of full irrigation) on cotton yield. The prediction revealed that the proposed CNN regression models outperformed conventional CNN models, such as AlexNet, CNN-3D, CNN-LSTM, ResNet. The proposed CNN model showed state-of-art performance at different image scales, with the R2 exceeding 0.9. At the cotton row level, the mean absolute error (MAE) and mean absolute percentage error (MAPE) were 3.08 pounds per row and 7.76%, respectively. At the cotton grid level, the MAE and MAPE were 0.05 pounds and 10%, respectively. This shows the proposed model's adaptability to the dynamic interplay between spatial and temporal factors that affect cotton yield. The authors conclude that integrating UAV-derived imagery and CNN regression models is a potent strategy for advancing precision agriculture, providing growers with a powerful tool to optimize cultivation practices and enhance overall cotton productivity.

A Deep Learning-Based Diabetes Diagnosis Model on PIMA Image Dataset

Deep learning is a highly useful technique for the early identification of diabetes mellitus, according to the study done by numerous authors over the past few decades. By using pre-processing techniques on the dataset to get rid of various anomalies like over-fitting, under-fitting, redundancy, missing values, and non-significant features to make it more efficient for analysis, it is possible to increase the effectiveness of deep learning algorithms for diagnosing the disease. This work addresses the global problem of diabetes by exploring a revolutionary deep-learning method for early identification. Conventional convolutional neural network (CNN) models have drawbacks when used with numerical medical datasets, like this study's PIMA Indians Diabetes Database. The article suggests a technique for transforming numerical data into visual representations depending on feature relevance to get over this obstacle. This conversion makes it possible to use strong CNN models for diabetes early diagnosis. Classifying the created diabetic images after feeding them into CNN architectures that have already been trained on VGG16 and ResNet50. The promising outcomes with an accuracy of 97.19% demonstrate the possibility of the suggested strategy for improving diabetes detection and validating the effectiveness of diabetes imaging in obtaining an early diagnosis.

Fruit Quality Analysis and Disease Detection using Deep Learning Techniques

Defective fruits are the main reason for worldwide financial catastrophes in agricultural production. It affects both the dependability and quality of the fruits. Post-harvest, quality checking requires a significant amount of time and labor-intensive skill. Automatically identifying fruit quality enables saving time and labor during harvest. Various algorithms have been created using machine learning and image processing methods to detect and categorize fruit quality. A system using Convolutional Neural Networks (CNN) and transfer learning techniques has been developed to enhance the fruit classification process. Two methods are suggested for predicting fruit freshness. One tailored CNN architecture is proposed by modifying the network's parameters to suit the dataset. The second technique utilizes the pre-trained VGG model with transfer learning to assess the freshness of the fruit. The proposed models are capable of differentiating between fresh and spoiled fruit by analyzing the input photos. This research used 70 diverse types of fruit, such as apples, bananas, oranges, and others. The first CNN obtains 98.55% accuracy for heterogeneous fruit dataset while second VGG16 achieves 99.05% accuracy on similar dataset. The plant village global dataset is utilized with various fruit categories. As a result, The VGG16 provides higher accuracy than conventional CNN and other deep learning algorithms.

Comparison of Conventional CNN Sequential API and Functional API for Microalgae Identification

Microalgae is widely known for its application in producing biodiesel and other health supplements. However, microalgae are also the leading cause of harmful algae blooms that may affect consumers and sea wildlife. The current microalgae identification method requires professionals, resources, budgets, technologies, and time. Therefore, a novel approach to identifying microalgae has been produced by implementing deep learning, specifically the Convolutional Neural Network (CNN). Due to the blooming of research in the deep learning field for microalgae identification, this research aims to compare application programming interface (API) use and study its effects on the accuracy and loss of a model. Using a light microscope, the microalgae images' datasets are self-collected from the AlBio laboratory at the Malaysia-Japan International Institute of Technology, Universiti Teknologi Malaysia. The microalgae species were Acutodesmus obliquus, Monoraphidium sp, Spirullina sp, Tetradesmus deserticola, and Desmodesmus perforatus. The architecture used to identify the microalgae in this research was the conventional CNN with different APIs, functional and sequential. The functional API resulted in 0.85 accuracies and a loss score of 3.77. On the other hand, the sequential scored 0.89 and a loss of 0.32. This study concluded that the sequential API was better than the functional API for a linear convolutional neural network. However, further improvement to the model could be applied by applying better hyperparameters and parameters to prevent underfitting and improve the model’s accuracy.

An adaptive fully convolutional network for bearing fault diagnosis under noisy environments.

Intelligent diagnostic algorithms based on convolutional neural networks (CNNs) have shown great potential in diagnosing various conditions. However, accurately and robustly diagnosing faults in noisy situations remains challenging. This study presents an adaptive fully convolutional network (AFCN) for identifying bearing defects in noisy environments. First, we use a novel large kernel convolution method for high-frequency noise reduction and wide-area temporal feature extraction. By utilizing a sequence of stacked residual adaptive convolution blocks, the AFCN achieves a selective emphasis on significant features and adaptive adjustment of feature weights at various convolution scales. The experimental results have shown that the AFCN achieves a diagnostic accuracy of over 90% for the faults in the CWRU dataset under the -8 dB noise and over 77% for the PU dataset in the case of -6 dB noise. The comparison results with five advanced baseline models have demonstrated the superiority of the AFCN in feature extraction, noise immunity, and robustness to the noise environment. The AFCN provides a better adaption to noise interference than conventional CNNs and other advanced adaptive networks.

An Efficient Segmentation and Classification of Brain Tumor Detection using Deep Learning

In medical diagnostics, the identification of brain tumors is a crucial endeavour that frequently necessitates the careful review of intricate imaging data. With the development of deep learning methods, especially convolutional neural networks (CNNs), automated brain tumor detection has become a viable way to help radiologists diagnose patients accurately and quickly. This study provides an extensive overview of current developments in deep learning approaches for brain tumor identification. The suggested techniques usually start with preprocessing medical images, like CT or MRI scans, and then use CNN architectures to extract features. Different CNN architectures, such as U-Net and conventional CNNs, have been used to extract discriminative characteristics from brain pictures. The performance of brain tumor detection systems has also been improved using transfer learning techniques, which make use of pre-trained models on sizable datasets. These results are especially encouraging when there is a shortage of training data. Additionally, research has been done on using ensemble learning approaches to strengthen the generality and durability of brain tumor detection models. By combining several base models into one prediction, these methods improve overall performance and lower the likelihood of overfitting. Furthermore, the incorporation of sophisticated regularisation methods such batch normalisation and dropout has enhanced the deep learning models' capacity for generalisation in brain tumor identification tasks. Moreover, the use of deep learning models in practical clinical environments demands the resolution of issues of uncertainty estimation and model interpretability

Image and video analysis using graph neural network for Internet of Medical Things and computer vision applications

AbstractGraph neural networks (GNNs) have revolutionised the processing of information by facilitating the transmission of messages between graph nodes. Graph neural networks operate on graph‐structured data, which makes them suitable for a wide variety of computer vision problems, such as link prediction, node classification, and graph classification. The authors explore in depth the applications of GNNs in computer vision, including their design considerations, architectural challenges, applications, and implementation concerns. While conventional convolutional neural networks (CNNs) excel at object recognition in images and videos, GNN architectures offer a novel method for addressing various image and video comprehension challenges. A novel deep neural network‐based model for image and video analysis is proposed, which combines a neural network with fully connected layers on a graph. The proposed architecture extracts highly discriminative information from images and videos by leveraging the graph structure. Also, the investigation focuses on the enhancement of underlying connection network estimation using cutting‐edge graph learning algorithms. Experimental results on real‐world datasets demonstrate that the proposed GNN model is preferable to existing state‐of‐the‐art methods. It obtains a remarkable 96.63% accuracy on the ImageNet dataset, outperforming heuristic approaches, artificial neural networks, and conventional CNN techniques. From the results, we can see that GNNs are a potent instrument for graph data analysis and pave the way for machines to achieve human‐level visual intuition.

A method for intelligent identification of faults in seismic using an attention-based ES-UNet network with model re-training learning

Accurate fault identification provides an important basis for well location deployment, oil and gas resource development. However, obtaining a large number of fault samples through manual labeling or interpretation is difficult. Conventional convolutional neural networks have certain challenges in finely characterizing low-order faults and fault continuity. This paper studies a 3D intelligent fault identification method that integrates deep learning with attention mechanism. Based on 3D-UNet network, the method constructs a dual attention fault identification model with an encoder-decoder structure, which focuses on enhancing low-order fault information and extracting key features. An adaptive hybrid loss function is proposed to address the imbalance problem of fault proportions in seismic amplitude data. The function introduces an adjustment coefficient to adaptively enhance the attention to high-order or low-order faults according to the actual situation of the fault, improving the accuracy and continuity of fault identification. Furthermore, a model re-training learning optimization method is presented to capture the missed fault information in the first training and repair the connectivity of low-order faults. Applying this method to the synthetic data set, the accuracy can be improved to 0.9732, and the loss can be converged to 0.0982. The results of shale reservoirs in the North Sea Basin and Kerry-3D indicate that the method can reduce the situation of fault missing and misidentification, effectively improve fault identification accuracy, and enhance low-order fault identification by obtaining more contextual fault features.

SWPanGAN: A hybrid generative adversarial network for pansharpening

AbstractPansharpening is a vital technique in remote sensing that combines a low‐resolution multi‐spectral image with its corresponding panchromatic image to obtain a high‐resolution multi‐spectral image. Despite its potential benefits, the challenge lies in extracting features from the source images and eliminating artefacts in the fused images. In response to the challenge, a hybrid generative adversarial network‐based model, termed SWPanGAN, is proposed. For better feature extraction, the conventional convolution neural network is replaced with a Swin transformer in the generator, which provides the generator with the ability to model long‐range dependencies. Additionally, to suppress artefacts, a wavelet‐based discriminator is proposed for effectively distinguishing the frequency discrepancy. With these modifications, both the generator and discriminator networks of SWPanGAN are enhanced. Extensive experiments illustrate that our SWPanGAN can generate high‐quality pansharpening images and surpass other state‐of‐the‐art methods.

DeepCG: A cell graph model for predicting prognosis in lung adenocarcinoma.

Lung cancer is the first leading cause of cancer-related death in the United States, with lung adenocarcinoma as the major subtype accounting for 40% of all cases. To improve patient survival, image-based prognostic models were developed due to the ready availability of pathological images at diagnosis. However, the application of these models is hampered by two main challenges: the lack of publicly available image datasets with high-quality survival information and the poor interpretability of conventional convolutional neural network models. Here, we integrated matched transcriptomic and H&E staining data from TCGA (The Cancer Genome Atlas) to develop an image-based prognostic model, termed Deep-learning based Cell Graph (DeepCG) model. Instead of survival data, we used a gene signature to predict patient prognostic risks, which was then used as labels for training DeepCG. Importantly, by employing graph structures to capture cell patterns, DeepCG can provide cell-level interpretation, which was more biologically relevant than previous region-level insights. We validated the prognostic values of DeepCG in independent datasets and demonstrated its ability to identify prognostically informative cells in images.

Transfer-Learning-Based Human Activity Recognition Using Antenna Array

Due to its low cost and privacy protection, Channel-State-Information (CSI)-based activity detection has gained interest recently. However, to achieve high accuracy, which is challenging in practice, a significant number of training samples are required. To address the issues of the small sample size and cross-scenario in neural network training, this paper proposes a WiFi human activity-recognition system based on transfer learning using an antenna array: Wi-AR. First, the Intel5300 network card collects CSI signal measurements through an antenna array and processes them with a low-pass filter to reduce noise. Then, a threshold-based sliding window method is applied to extract the signal of independent activities, which is further transformed into time–frequency diagrams. Finally, the produced diagrams are used as input to a pretrained ResNet18 to recognize human activities. The proposed Wi-AR was evaluated using a dataset collected in three different room layouts. The testing results showed that the suggested Wi-AR recognizes human activities with a consistent accuracy of about 94%, outperforming the other conventional convolutional neural network approach.