Performance Of Convolutional Neural Networks Research Articles

Convolutional and hybrid neural network for cluster membership

We conduct investigations of the membership of galaxies inside a cluster using a machine-learning Convolutional Neural Network (CNN), and a hybrid of CNN with Multi-Layer Perceptron (MLP) neural network. These networks are trained using optical images of each galaxy in I and V bands as well as using various photometric and morphological parameters of galaxies extracted and measured from these images. The optical images are observed with Subaru Suprime-Cam for 16 clusters at redshift ∼ 0.15–0.3. This dataset enables a wide and faint investigation of cluster membership that is essential for the investigations of dwarf galaxies, but it is often difficult to achieve both simultaneously. We optimize the network structure to directly determine the membership classification, rather than estimating the galaxy redshift first as an intermediate step and then determining the membership from the redshifts. We find: 1. The CNN model can be useful for the application of identifying the cluster membership of galaxies, similar to the MLP model, implying that the CNN model can use information beyond galaxy SED. 2. A simple CNN model appears to predict the cluster membership less accurately than the 17-measure MLP model. 3. For better performance of CNN, the masking of the object is important, as well as scaling based on the photometric calibration. 4. The hybrid model of CNN and MLP combined can produce a better prediction than the single MLP (or CNN) model; however, the improvement is small at best implying that the information extracted by 17 measures for the MLP model is fairly comprehensive in sense of the total information contained in pixels of the image. 5. All models appear to predict the membership more accurately inside a smaller radius closer to the cluster center. 6. Faint galaxies are harder to assign their memberships even if using our models, though the performance is more robust than other photometric methods.Artificial intelligence (AI) can help us effectively combine various conventional methods of finding cluster membership. By making it inherit advantages of each of these conventional methods, The AI membership can achieve overall good performance in various performance statistics simultaneously, such as purity, completeness, F1 score, AUC, and accuracy. The overall good performance of the AI membership is vital to cluster studies in the era of faint and data-intensive galaxy survey s in which the complete spectroscopic observation of all galaxies is out of reach.

Variety classification and identification of jujube based on near-infrared spectroscopy and 1D-CNN

Jujube is an important economic crop in Xinjiang, China, and its related industry provides abundant employment opportunities and contributes positively to local economic development. Due to the high economic value and similar appearance of jujubes from different regions, there is a phenomenon of counterfeit products in the market. Therefore, it is necessary to trace the origin of jujubes to address these issues. In this study, we conducted traceability research on jujubes from four locations: Alaer, Hotan, Ruoqiang, and Zhangye. Near-infrared spectroscopy was used to extract non-destructive data from the jujubes for analysis. The extracted data was transformed into one-dimensional data, and six algorithms, including the Back Propagation Neural Network (BP), Radial Basis Function Neural Network(RBF), Convolutional Neural Networks (CNN), Long Short-Term Memory Network(LSTM), Support Vector Machine (SVM), and Random Forest (RF), were used for classification research on the obtained 4000 sets of one-dimensional data. The results showed that when sufficient data were available, the RBF, LSTM and CNN perform relatively better best with 93.50 %, 94.33 % and 94.25 % accuracy. The RF achieved an accuracy of 92.42 % and demonstrated good detection efficiency. However, the BP and SVM performed relatively poorly, with accuracies of 90.42 % and 80.42 % respectively. Considering the time-consuming nature of data collection for large datasets during the detection process, several sets of one-dimensional data were randomly selected for further experiments, through which 400 sets of data were found to be more prominent with 700 sets of data. The final experimental results demonstrate that, in the traceability detection of a small-scale dataset containing 400 instances, the CNN exhibits the best performance, achieving an accuracy of 86.67 %. The performance of the LSTM is slightly lower than that of the CNN, with an accuracy of 84.12 %. The accuracies of the BP, RBF, SVM, and RF were relatively lower at 82.5 %, 82.5 %, 74.12 %, and 74.18 % respectively. In the experiments with a dataset of 700 instances, the performance of CNN stands out prominently. In comparison to the CNN algorithm on the 400-instance dataset and the second-best LSTM algorithm, the accuracy, precision, recall, and F1-score have exhibited significant improvements by 53.13 %, 98.98 %, 481.43 %, and 140.36 %, respectively. The corresponding values are 0.9043, 0.9083, 0.9052, and 0.906. This study demonstrates the non-destructive traceability of jujube fruits using near-infrared spectroscopy data combined with neural network algorithms. Additionally, it reduces the need for additional preprocessing steps in data handling and achieves promising detection results. This research has positive implications for the traceability of jujube fruits and related crops.

Automated 3D convolutional neural network architecture design using genetic algorithm for pulmonary nodule classification

Cancer of the lungs is considered one of the primary causes of death among patients globally. Early detection contributes significantly to the success of pulmonary cancer treatment. To aid the pulmonary nodule classification, many models for the analysis of medical image utilizing deep learning have been developed. Convolutional neural network (CNN) recently, has attained remarkable results in various image classification tasks. Nevertheless, the CNNs performance is heavily dependent on their architectures which still heavily reliant on human domain knowledge. This study introduces a cutting-edge approach that leverages genetic algorithms (GAs) to automatically design 3D CNN architectures for differentiation between benign and malignant pulmonary nodules. The suggested algorithm utilizes the dataset of lung nodule analysis 2016 (LUNA16) for evaluation. Notably, our approach achieved exceptional model accuracy, with evaluations on the testing dataset yielding up to 95.977%. Furthermore, the algorithm exhibited high sensitivity, showcasing its robust performance in distinguishing between benign and malignant nodules. Our findings demonstrate the outstanding capabilities of the proposed algorithm and show an outstanding performance and attain a state of art solution in lung nodule classification.

An advanced approach for accurate pneumonia detection using combined deep convolutional neural networks

Pneumonia, a lung infection caused by viral or bacterial agents, poses a significant health risk by affecting one or both lungs in humans. Accurate diagnosis, particularly in pediatric cases, is crucial for timely intervention. chest X-rays (CXRs) are a common and non-invasive diagnostic tool to detect pneumonia-related abnormalities. Nonetheless, the minimal radiation exposure suitable for pediatric diagnosis poses a challenge in accurately detecting pneumonia in children. This work proposes a concatenation model that combines two pre-trained convolutional neural networks (CNNs) depending on the transfer learning (TL) technique and optimizes the training parameters to build a highly accurate model for detecting pediatric pneumonia from CXR images. The concatenated extracted features from the two pre-trained CNNs are passed through a convolutional layer to select more valuable semantic features to reduce the extracted features, which helps reduce the model parameters and execution time. Experimental results demonstrate that the feature concatenation technique, along with optimization of training parameters, surpasses the performance of individual CNNs and several state-of-the-art methods. The proposed method achieves a classification accuracy of 98.5%, precision of 99.5%, sensitivity of 98.4%, and F1 score of 99.1%. The primary objective of the proposed approach is to aid radiologists in achieving accurate pneumonia diagnosis in real-time.

Performance Analysis of Deep Learning and Machine Learning Methods for Music Genre Classification System

Performance Analysis of Deep Learning and Machine Learning Methods for Music Genre Classification System

Generation and classification of patch-based land use and land cover dataset in diverse Indian landscapes: a comparative study of machine learning and deep learning models.

In the context of environmental and social applications, the analysis of land use and land cover (LULC) holds immense significance. The growing accessibility of remote sensing (RS) data has led to the development of LULC benchmark datasets, especially pivotal for intricate image classification tasks. This study addresses the scarcity of such benchmark datasets across diverse settings, with a particular focus on the distinctive landscape of India. The study entails the creation of patch-based datasets, consisting of 4000 labelled images spanning four distinct LULC classes derived from Sentinel-2 satellite imagery. For the subsequent classification task, three traditional machine learning (ML) models and three convolutional neural networks (CNNs) were employed. Despite facing several challenges throughout the process of dataset generation and subsequent classification, the CNN models consistently attained an overall accuracy of 90% or more. Notably, one of the ML models stood out with 96% accuracy, surpassing CNNs in this specific context. The study also conducts a comparative analysis of ML models on existing benchmark datasets, revealing higher prediction accuracy when dealing with fewer LULC classes. Thus, the selection of an appropriate model hinges on the given task, available resources, and the necessary trade-offs between performance and efficiency, particularly crucial in resource-constrained settings. The standardized benchmark dataset contributes valuable insights into the relative performance of deep CNN and ML models in LULC classification, providing a comprehensive understanding of their strengths and weaknesses.

Convolutional Neural Network to Classify Infrared Thermal Images of Fractured Wrists in Pediatrics.

Convolutional neural network (CNN) models were devised and evaluated to classify infrared thermal (IRT) images of pediatric wrist fractures. The images were recorded from 19 participants with a wrist fracture and 21 without a fracture (sprain). The injury diagnosis was by X-ray radiography. For each participant, 299 IRT images of their wrists were recorded. These generated 11,960 images (40 participants × 299 images). For each image, the wrist region of interest (ROI) was selected and fast Fourier transformed (FFT) to obtain a magnitude frequency spectrum. The spectrum was resized to 100 × 100 pixels from its center as this region represented the main frequency components. Image augmentations of rotation, translation and shearing were applied to the 11,960 magnitude frequency spectra to assist with the CNN generalization during training. The CNN had 34 layers associated with convolution, batch normalization, rectified linear unit, maximum pooling and SoftMax and classification. The ratio of images for the training and test was 70:30, respectively. The effects of augmentation and dropout on CNN performance were explored. Wrist fracture identification sensitivity and accuracy of 88% and 76%, respectively, were achieved. The CNN model was able to identify wrist fractures; however, a larger sample size would improve accuracy.

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.

Deep learned features selection algorithm: Removal operation of anomaly feature maps (RO-AFM)

Class/Regression Activation Maps (CAMs/RAMs; AMs) are often embedded into Convolutional Neural Networks (CNNs) for checking activated regions on input images at estimation. CNNs sometime generate unreliable AMs, such as activated regions, are inappropriate. Because AM is calculated by stacking many feature maps generated by the final convolutional layer, when there are Anomaly Feature Maps (AFMs), unreliable AMs can be generated. For example, suppose we have a CNN that evaluates the heart. In this case, the feature maps that focus on regions unrelated to the heart (e.g., shoulders and esophagus) are AFMs. Additionally, we have a hypothesis that the estimation accuracy of CNNs is increased by removing AFMs. However, methods for automatically detecting and removing AFMs have not been sufficiently studied in previous research to improve the performance of CNNs. Therefore, we propose a method named “Removal Operation of Anomaly Feature Maps (RO-AFMs)” to automatically detect and remove AFMs. When applying an RO-AFM to the Global Average Pooling (GAP) feature vectors of a CNN, dimensions of the GAP vector are reduced. Therefore, an RO-AFM is regarded as a deep-feature selection algorithm. From the results of adopting an RO-AFM to a Regression CNN (R-CNN) for estimating pulmonary artery wedge pressure, which is one of the measurement score for representing cardiac anomaly state, improved reliability of AM and estimation accuracy were verified. A comparison of RO-AFM and the existing methods, i.e., Lasso and the Feature Selection Layer (FSL), indicated that RO-AFM performed slightly better on the estimation accuracy. The computation time required for RO-AFM to evaluate all features was 1.833 s on average, confirming that RO-AFM is a lightweight process. Therefore, RO-AFM is useful for constructing a medical CNN that emphasizes explainability (e.g., CNNs for estimating the risk of a disease or a test value from chest X-ray or computed tomography images).

On the use of GNN-based structural information to improve CNN-based semantic image segmentation

Convolutional neural networks (CNNs) are widely used for semantic image segmentation across various fields (medicine, robotics), capturing local pixel dependencies for good results. Nevertheless, CNNs struggle to grasp global contextual representations, sometimes leading to structural inconsistencies. Recent approaches aim to broaden their scope using attention mechanisms or deep models, resulting in heavy-weight architectures. To boost CNN performance in semantic segmentation, we propose using a graph neural network (GNN) as a post-processing step. The GNN conducts node classification on appropriately coarsened graphs encoding class probabilities and structural information related to regions segmented by the CNN. The proposal, applicable to any CNN producing a segmentation map, is evaluated on several CNN architectures, using two public datasets (FASSEG and IBSR), with four graph convolution operators. Results reveal performance improvements, enhancing on average the Hausdorff distance by 24.3% on FASSEG and by 74.0% on IBSR. Furthermore, our approach demonstrates resilience to small training datasets.

Diabetic Retinopathy Stage Detection Using CNN and Inception V3

The project explores the deployment of Convolutional Neural Networks (CNN) and the Inception V3 model for the automated detection and classification of diabetic retinopathy stages using fundus images. Recognizing diabetic retinopathy as a leading cause of blindness among the working-age population globally, this research aims to streamline the diagnostic process, traditionally reliant on the manual examination by ophthalmologists. Through the utilization of the DRIVE and STARE datasets, the project benchmarks the performance of CNN and Inception V3 models in accurately categorizing the severity of diabetic retinopathy into five distinct stages. The comparison between these models is grounded on parameters such as accuracy, loss, and predicted value, with findings indicating Inception V3's superiority in both performance metrics and diagnostic precision. This advancement could significantly contribute to early and more accessible detection of diabetic retinopathy, thereby mitigating progression towards blindness. Furthermore, the project underscores the potential of deep learning algorithms in enhancing diagnostic methodologies for retinal diseases, paving the way for future explorations in the field of medical imaging and artificial intelligence.

Multi-task learning with self-learning weight for image denoising

BackgroundImage denoising technology removes noise from the corrupted image by utilizing different features between image and noise. Convolutional neural network (CNN)-based algorithms have been the concern of the recent progress on diverse image restoration problems and become an efficient solution in image denoising.ObjectiveAlthough a quite number of existing CNN-based image denoising methods perform well on the simplified additive white Gaussian noise (AWGN) model, their performance often degrades severely on the real-world noisy images which are corrupted by more complicated noise.MethodsIn this paper, we utilized the multi-task learning (MTL) framework to integrate multiple loss functions for collaborative training of CNN. This approach aims to improve the denoising performance of CNNs on real-world images with non-Gaussian noise. Simultaneously, to automatically optimize the weights of individual sub-tasks within the MTL framework, we incorporated a self-learning weight layer into the CNN.ResultsExtensive experiments demonstrate that our approach effectively enhances the denoising performance of CNN-based image denoising algorithms on real-world images. It reduces excessive image smoothing, improves quantitative metrics, and enhances visual quality in the restored images.ConclusionOur method shows the effectiveness of the improved performance of denoising CNNS for real-world image denoising processing.

A convolutional neural network based on an evolutionary algorithm and its application

PM2.5 concentration predictions can provide air pollution control, management, and early warning. However, the PM2.5 data with high-dimensionality, complexity, and dynamics pose a great challenge to achieve optimal prediction results. Convolutional Neural Networks (CNNs) has unique advantages in processing complex data and contributes to state-of-the-art performances. However, designing the architecture and selecting the learning rate for CNN are time-consuming and requires prior knowledge. Evolutionary algorithms, with the advantages of global convergence, ergodicity, robustness and adaptability, are the most commonly used methods to design the optimal framework for CNNs. Therefore, to improve the predictive performance of CNNs, this paper proposes an improved CNN method (EBRO-ICNN) which employs the enhanced battle royale optimization (EBRO) algorithm and proportional-derivative (PD) control. Firstly, the EBRO algorithm with multistrategy collaborative optimization is introduced, and validated by CEC2017 benchmark functions, which demonstrates EBRO strong global exploration capability, fast convergence speed, and low time complexity. Next, PD control is applied to adjust the learning rate of CNN (ICNN) dynamically, which improves the efficiency and stability of the network training process. At the same time, the ICNN model is optimized using the EBRO algorithm, which can reduce the human interference, generate the optimal framework automatically and enhance the prediction accuracy effectively. Finally, a private air quality dataset and two public datasets are utilized to evaluate the performance of the EBRO-ICNN model, considering 3 error evaluation metrics and 7 prediction comparison models. The experimental results demonstrate that the EBRO-ICNN model exhibits great accuracy and stability.

Permeability estimation for deformable porous media with convolutional neural network

Purpose Accurately evaluating fluid flow behaviors and determining permeability for deforming porous media is time-consuming and remains challenging. This paper aims to propose a novel machine-learning method for the rapid estimation of permeability of porous media at different deformation stages constrained by hydro-mechanical coupling analysis. Design/methodology/approach A convolutional neural network (CNN) is proposed in this paper, which is guided by the results of finite element coupling analysis of equilibrium equation for mechanical deformation and Boltzmann equation for fluid dynamics during the hydro-mechanical coupling process [denoted as Finite element lattice Boltzmann model (FELBM) in this paper]. The FELBM ensures the Lattice Boltzmann analysis of coupled fluid flow with an unstructured mesh, which varies with the corresponding nodal displacement resulting from mechanical deformation. It provides reliable label data for permeability estimation at different stages using CNN. Findings The proposed CNN can rapidly and accurately estimate the permeability of deformable porous media, significantly reducing processing time. The application studies demonstrate high accuracy in predicting the permeability of deformable porous media for both the test and validation sets. The corresponding correlation coefficients (R2) is 0.93 for the validation set, and the R2 for the test set A and test set B are 0.93 and 0.94, respectively. Originality/value This study proposes an innovative approach with the CNN to rapidly estimate permeability in porous media under dynamic deformations, guided by FELBM coupling analysis. The fast and accurate performance of CNN underscores its promising potential for future applications.

Evaluation of the effects of K-Means Clustered-Based Weight Quantization on a Keras Library Based Convolutional Neural Network for Hand Written Digit Image Recognition

A lot of Convolutional Neural Networks (CNNs) have been implemented using FPGAs for the past years. Subsequently, memory saving features were added to the CNN through weight quantization using K-means clustering. A future goal on an ASIC design, involving CNN and weight quantization working together in one chip, can give way to an automated procedure of memory-saving CNN design. In this paper an evaluation was done on the effect of quantizing the weights of a Keras library-based CNN using K means clustering. Various values of K in K-means clustering were tested to see its effects on the CNN accuracy performance. This paper presents first the design approach of a Keras library based Convolutional Neural Network (CNN) for hand-written digit images. It then presents a hardware model design of K-Means clustering algorithm using VHDL. The performance of CNN for image recognition was then tested for various levels of weight quantization using K-means clustering algorithm. Simulation results showed a compression of weights as high as 60% resulted to less than 1% reduction in CNN’s accuracy. The findings in this paper will serve as guide in determining the relevant values of K i.e. the compression ratio, for future ASIC design on this topic.

Unlocking Machine Learning Algorithms for Bambooshoots.AI: Revolutionizing Agricultural Applications with Computer Science

This research initiated a comprehensive investigation to determine the most suitable Machine Learning (ML) algorithms for bamboshoots.AI application. Convolutional neural networks (CNNs) and support vector machines (SVMs) were cited as two essential ML techniques in a comprehensive review of ten papers. To further assess these two algorithms in the context of the Bambooshoots.ai application, an experimental analysis was also carried out. The CNN model had been trained over 25 epochs with 8 batches of data per epoch, demonstrating a consistent increase in accuracy, and reached 97.94% by the end of training. In contrast, the SVM model provided an accuracy of approximately 57.14%. The experimental results indicated that the CNN model had been better at classifying bamboo shoot images, thus making it the preferred choice for the BambooShoots.AI application. The significant difference in accuracy between the two models, as well as the consistent performance of CNNs in image-based tasks, informed this decision. The results of this research were thrilling because they showed that machine learning could be used to help understand and manage bamboo shoots. This could have been a big help to farmers and could have led to new ways of doing things in agriculture.

Denoising Seismic Waveforms Using a Wavelet-Transform-Based Machine-Learning Method

ABSTRACT Seismic waveform data recorded at stations can be thought of as a superposition of the signal from a source of interest and noise from other sources. Frequency-based filtering methods for waveform denoising do not result in desired outcomes when the targeted signal and noise occupy similar frequency bands. Recently, denoising techniques based on deep-learning convolutional neural networks (CNNs), in which a recorded waveform is decomposed into signal and noise components, have led to improved results. These CNN methods, which use short-time Fourier transform representations of the time series, provide signal and noise masks for the input waveform. These masks are used to create denoised signal and designaled noise waveforms, respectively. However, advancements in the field of image denoising have shown the benefits of incorporating discrete wavelet transforms (DWTs) into CNN architectures to create multilevel wavelet CNN (MWCNN) models. The MWCNN model preserves the details of the input due to the good time–frequency localization of the DWT. Here, we use a data set of over 382,000 constructed seismograms recorded by the University of Utah Seismograph Stations network to compare the performance of CNN and MWCNN-based denoising models. Evaluation of both models on constructed test data shows that the MWCNN model outperforms the CNN model in the ability to recover the ground-truth signal component in terms of both waveform similarity and preservation of amplitude information. Model evaluation of real-world data shows that both the CNN and MWCNN models outperform standard band-pass filtering (BPF; average improvement in signal-to-noise ratio of 9.6 and 19.7 dB, respectively, with respect to BPF). Evaluation of continuous data suggests the MWCNN denoiser can improve both signal detection capabilities and phase arrival time estimates.

Application of Convolutional Neural Networks for Early Detection and Classification of Alzheimer's disease from MRI Images

This study investigates the application of convolutional neural networks (CNNs) and traditional machine learning algorithms for the early detection and classification of Alzheimer's disease (AD) using brain Magnetic Resonance Imaging (MRI) data. We compare the performance of CNNs with Support Vector Machines (SVM), Random Forest (RF), and Gradient Boosting Machines (GBM) on a dataset comprising MRI images from AD patients and healthy controls. Results show that CNNs achieved the highest accuracy (90.2%) and area under the receiver operating characteristic curve (AUC-ROC) of 0.95, outperforming SVM, RF, and GBM. The CNN model also exhibited high sensitivity (87.5%) and specificity (92.6%) in distinguishing between AD patients and healthy controls. These findings highlight the effectiveness of CNN-based approaches in leveraging raw MRI images for accurate and early detection of AD.

Diagnostic effectiveness of deep learning-based MRI in predicting multiple sclerosis: A meta-analysis.

The brain and spinal cord, constituting the central nervous system (CNS), could be impacted by an inflammatory disease known as multiple sclerosis (MS). The convolutional neural networks (CNN), a machine learning method, can detect lesions early by learning patterns on brain magnetic resonance image (MRI). We performed this study to investigate the diagnostic performance of CNN based MRI in the identification, classification, and segmentation of MS lesions. PubMed, Web of Science, Embase, the Cochrane Library, CINAHL, and Google Scholar were used to retrieve papers reporting the use of CNN based MRI in MS diagnosis. The accuracy, the specificity, the sensitivity, and the Dice Similarity Coefficient (DSC) were evaluated in this study. In total, 2174 studies were identified and only 15 articles met the inclusion criteria. The 2D-3D CNN presented a high accuracy (98.81, 95% CI: 98.50-99.13), sensitivity (98.76, 95% CI: 98.42-99.10), and specificity (98.67, 95% CI: 98.22-99.12) in the identification of MS lesions. Regarding classification, the overall accuracy rate was significantly high (91.38, 95% CI: 83.23-99.54). A DSC rate of 63.78 (95% CI: 58.29-69.27) showed that 2D-3D CNN-based MRI performed highly in the segmentation of MS lesions. Sensitivity analysis showed that the results are consistent, indicating that this study is robust. This metanalysis revealed that 2D-3D CNN based MRI is an automated system that has high diagnostic performance and can promptly and effectively predict the disease.

Comparison of Activation Functions in Convolutional Neural Network for Poisson Noisy Image Classification

Deep learning, specifically the Convolutional Neural Network (CNN), has been a significant technology tool for image processing and human health. CNNs, which mimic the working principles of the human brain, can learn robust representations of images. However, CNNs are susceptible to noise interference, which can impact classification performance. Choosing the right activation function can improve CNNs performance and accuracy. This research aims to test the accuracy of CNN with ResNet50, VGG16, and GoogleNet architectures combined with several activation functions such as ReLU, Leaky ReLU, Sigmoid, and Tanh in the classification of images that experience Poisson noise. Poisson noise is applied to each test data to evaluate CNN accuracy. The data used in this study consists of three scenarios of different numbers of classes, namely 3 classes, 5 classes, and 10 classes. The results showed that combining ResNet50 with the ReLU activation function produced the best performance in class recognition in each scenario of the number of classes experiencing Poisson noise interference. The model achieved 97% accuracy for 3-class data, 95% for 5-class data, and 90% for 10-class data. These results show that using ResNet50 with the ReLU activation function can provide excellent resistance to Poisson noise in image processing. It was found that as the number of classes increases, the accuracy of image recognition tends to decrease. This shows that the more complex the image classification task is with a larger number of classes, the more difficult it is for CNNs to distinguish between different classes. Doi: 10.28991/ESJ-2024-08-02-014 Full Text: PDF