Fine-tuned Convolutional Neural Network Research Articles

Multi-culture Sign Language Detection and Recognition Using Fine-tuned Convolutional Neural Network

The speech and hearing-impaired community use sign language as the primary means of communication. It is quite challenging for the general population to interpret or learn sign language completely. A sign language recognition system must be designed and developed to address this communication barrier. Most current sign language recognition systems rely on wearable sensors, keeping the recognition system unaffordable for most individuals. Moreover, the existing vision-based sign recognition frameworks do not consider all of the spatial and temporal information required for accurate recognition. A novel vision-based hybrid deep neural net methodology is proposed in this project for recognizing American Sign Language (ASL) and custom sign gestures. The proposed framework aims to establish a single framework for tracking and extracting multi-semantic properties, such as non-manual components and manual co-articulations. Furthermore, spatial feature extraction from the sign gestures is deployed using a Hybrid Deep Neural Network (HDNN) with atrous convolutions. The temporal and sequential feature extraction is carried out by employing attention-based HDNN. In addition, the distinguished abstract feature extraction is done using modified autoencoders. The discriminative feature extraction for differentiating the sign gestures from unwanted transition gestures is done by leveraging the hybrid attention module. The experimentation of the proposed model has been carried out on the novel multi-signer ASL and custom sign language dataset. The proposed sign language recognition framework with hybrid neural nets, specifically using HDNN, yields better results than other state-of-the-art frameworks. Additionally, a detection module is incorporated using Flask web, allowing manual input of images/signs for real-time recognition.

Deep learning approaches to recover the plasma current density profile from the safety factor based on Grad–Shafranov solutions across multiple tokamaks

Many magnetohydrodynamic stability analyses require generation of a set of equilibria with a fixed safety factor q-profile while varying other plasma parameters. A neural network (NN)-based approach is investigated that facilitates such a process. Both multilayer perceptron (MLP)-based NN and convolutional neural network (CNN) models are trained to map the q-profile to the plasma current density J-profile, and vice versa, while satisfying the Grad–Shafranov radial force balance constraint. When the initial target models are trained, using a database of semi-analytically constructed numerical equilibria, an initial CNN with one convolutional layer is found to perform better than an initial MLP model. In particular, a trained initial CNN model can also predict the q- or J-profile for experimental tokamak equilibria. The performance of both initial target models is further improved by fine-tuning the training database, i.e. by adding realistic experimental equilibria with Gaussian noise. The fine-tuned target models, referred to as fine-tuned MLP and fine-tuned CNN, well reproduce the target q- or J-profile across multiple tokamak devices. As an important application, these NN-based equilibrium profile convertors can be utilized to provide a good initial guess for iterative equilibrium solvers, where the desired input quantity is the safety factor instead of the plasma current density.

Thermal radiation of tongue surface as a human computer interaction diagnostics technique based on image classification with software interface

Backgroundand Purpose: Thermal radiation plays a pivotal role in addressing recognition limitations encountered by traditional Chinese medicine's tongue diagnosis. Modern thermal radiation diagnostic instruments offer enhanced objectivity, and in this study, we strive to further improve the diagnostic process through the integration of Human-Computer Interaction (HCI) principles. Our objective is to introduce the thermal radiation DenseNet ensemble model, synergizing the strengths of Convolutional Neural Networks (CNNs) and HCI principles to advance tongue image recognition in healthcare. This approach aims to enhance classification accuracy, even when faced with limited training data. MethodologyTransfer learning, utilizing fine-tuned CNNs and DenseNet, is employed for precise classification, with a specific focus on distinguishing tooth-marked tongues. A portable hand-held thermal radiation diagnostic instrument, seamlessly integrated with HCI, is developed for both clinical use and research. Experimental resultsMeticulously curated datasets, formed through the annotation of thermal radiation images, confirm the superior performance of the DenseNet architecture in tooth mark and tongue feature recognition. The proposed tooth mark tongue recognition model exhibits a significant accuracy improvement of approximately 25% compared to similar tasks in existing studies. Additionally, an AI health detector tailored for Traditional Chinese Medicine (TCM) thermal radiation image recognition is introduced, showcasing the seamless integration of HCI principles into healthcare applications. This innovative tool not only assesses users' health but also provides personalized health recommendations, illustrating the transformative potential of combining technology and healthcare. Discussion and conclusionIn addressing the limitations of traditional tongue diagnosis, thermal radiation emerges as a key factor enhancing objectivity and accuracy. The thermal radiation DenseNet ensemble model, along with HCI integration through the portable hand-held device, proves effective for improved TCM thermal radiation image identification in healthcare. The fusion of technology and HCI principles underscores the potential to elevate health assessments and offer valuable guidance to users.

Multigrade brain tumor classification in MRI images using Fine tuned efficientnet

Medical imaging plays a vital role in detecting and treating brain tumors. Malignant or non-malignant brain tissue’s abnormal growth causes long-term brain damage. It is crucial to detect and properly categorize the kind of brain tumor. Specialists normally use MRI to create high-contrast grayscale brain images to segment them. Convolutional neural networks (CNN) driven by deep learning (DL) have transformed computer-assisted testing systems by producing good results in a wide range of medical imaging analytics applications, including tumor diagnosis in the brain. The paper introduces a lightweight fine-tuned Convolutional Neural Network EfficientNet ’ECNN’ to detect brain tumors. In this study, we provide a transfer learning-based measurement strategy for grouping cerebrum growths in three distinct datasets with different classifications, such as meningioma, glioma, and pituitary growth, using fine-tuned EfficientNets. The findings of this research rely on Efficient Nets to classify brain tumors in three different types of datasets utilizing a fine-tuned transfer learning mechanism. With EfficientNetV2S as the system’s foundation, our proposed way of fine-tunned pre-trained EfficientNetV2S model outperformed for all datasets over state of the art methods. The effectiveness of the suggested model has been assessed using performance metrics, and outcomes were compared to those produced using state-of-the-art approaches. The average test accuracy, recall, precision, and sensitivity score are 98.48%, 98%, 98.5%, and 98.71%, respectively.

Improving classification accuracy of fine-tuned CNN models: Impact of hyperparameter optimization

The immense popularity of convolutional neural network (CNN) models has sparked a growing interest in optimizing their hyperparameters. Discovering the ideal values for hyperparameters to achieve optimal CNN training is a complex and time-consuming task, often requiring repetitive numerical experiments. As a result, significant attention is currently being devoted to developing methods aimed at tailoring hyperparameters for specific CNN models and classification tasks. While existing optimization methods often yield favorable image classification results, they do not provide guidance on which hyperparameters are worth optimizing, the appropriate value ranges for those hyperparameters, or whether it is reasonable to use a subset of training data for the optimization process. This work is focused on the optimization of hyperparameters during transfer learning, with the goal of investigating how different optimization methods and hyperparameter selections impact the performance of fine-tuned models. In our experiments, we assessed the importance of various hyperparameters and identified the ranges within which optimal CNN training can be achieved. Additionally, we compared four hyperparameter optimization methods—grid search, random search, Bayesian optimization, and the Asynchronous Successive Halving Algorithm (ASHA). We also explored the feasibility of fine-tuning hyperparameters using a subset of the training data. By optimizing the hyperparameters, we observed an improvement in CNN classification accuracy of up to 6%. Furthermore, we found that achieving a balance in class distribution within the subset of data used for parameter optimization is crucial in establishing the optimal set of hyperparameters for CNN training. The results we obtained demonstrate that hyperparameter optimization is highly dependent on the specific task and dataset at hand.

Melanoma identification and classification model based on fine-tuned convolutional neural network.

Breakthroughs in skin cancer diagnostics have resulted from recent image recognition and Artificial Intelligence (AI) technology advancements. There has been growing recognition that skin cancer can be lethal to humans. For instance, melanoma is the most unpredictable and terrible form of skin cancer. This paper aims to support Internet of Medical Things (IoMT) applications by developing a robust image classification model for the early detection of melanoma, a deadly skin cancer. It presents a novel approach to melanoma detection using a Convolutional Neural Network (CNN)-based method that employs image classification techniques based on Deep Learning (DL). We analyze dermatoscopic images from publicly available datasets, including DermIS, DermQuest, DermIS&Quest, and ISIC2019. Our model applies convolutional and pooling layers to extract meaningful features, followed by fully connected layers for classification. The proposed CNN model achieves high accuracy demonstrates the model's effectiveness in distinguishing between malignant and benign skin lesions. We developed deep features and used transfer learning to improve the categorization accuracy of medical images. Soft-max classification layer and support vector machine have been used to assess the classification performance of deep features. The proposed model's efficacy is rigorously evaluated using benchmark datasets: DermIS, DermQuest, and ISIC2019, having 621, 1233, and 25000 images, respectively. Its performance is compared to current best practices showing an average of 5% improved detection accuracy in DermIS, 6% improvement in DermQuest, and 0.81% in ISIC2019 datasets. Our study showcases the potential of CNN in melanoma detection, contributing to early diagnosis and improved patient outcomes. The developed model proves its capability to aid dermatologists in accurate decision-making, paving the way for enhanced skin cancer diagnosis.

Image-Based Subtype Classification for Glioblastoma Using Deep Learning: Prognostic Significance and Biologic Relevance.

To apply deep learning algorithms to histopathology images, construct image-based subtypes independent of known clinical and molecular classifications for glioblastoma, and produce novel insights into molecular and immune characteristics of the glioblastoma tumor microenvironment. Using whole-slide hematoxylin and eosin images from 214 patients with glioblastoma in The Cancer Genome Atlas (TCGA), a fine-tuned convolutional neural network model extracted deep learning features. Biclustering was used to identify subtypes and image feature modules. Prognostic value of image subtypes was assessed via Cox regression on survival outcomes and validated with 189 samples from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) data set. Morphological, molecular, and immune characteristics of glioblastoma image subtypes were analyzed. Four distinct subtypes and modules (imClust1-4) were identified for the TCGA patients with glioblastoma on the basis of the image feature data. The glioblastoma image subtypes were significantly associated with overall survival (OS; P = .028) and progression-free survival (P = .003). Apparent association was also observed for disease-specific survival (P = .096). imClust2 had the best prognosis for all three survival end points (eg, after 25 months, imClust2 had >7% surviving patients than the other subtypes). Examination of OS in the external validation using the unseen CPTAC data set showed consistent patterns. Multivariable Cox analyses confirmed that the image subtypes carry unique prognostic information independent of known clinical and molecular predictors. Molecular and immune profiling revealed distinct immune compositions of the tumor microenvironment in different image subtypes and may provide biologic explanations for the patterns in patients' outcomes. Our image-based subtype classification on the basis of deep learning models is a novel tool to refine risk stratification in cancers. The image subtypes detected for glioblastoma represent a promising prognostic biomarker with distinct molecular and immune characteristics and may facilitate developing novel, individualized immunotherapies for glioblastoma.

Fine-Tuned Convolutional Neural Networks for Feature Extraction and Classification of Scanned Document Images using Semi-Automatic Labelling Approach

Fine-Tuned Convolutional Neural Networks for Feature Extraction and Classification of Scanned Document Images using Semi-Automatic Labelling Approach

A systematic scoping review of the analysis of COVID-19 disease using chest X-ray images with deep learning models

<p>The significance of chest X-ray data in screening patients for COVID-19 has been recognised by medical experts. Deep learning (DL) technologies, particularly artificial intelligence (AI) algorithms, have emerged as efficient classifiers for diagnosing disease through the inspection of chest X-rays. Medical professionals may use deep learning skills to effectively allocate resources and prioritise patients, ensuring that people in critical need of medical attention receive it on time. In reviewed papers, chest X-ray images datasets are used in order to investigate if trained convolutional neural networks (CNNs) can be utilized to accurately classify COVID-19 cases. The study is made more fascinating by the availability of many kinds of new DL models designed specifically for this specific purpose. As the findings illustrate the efficacy of fine-tuned pretrained CNNs for COVID-19 identification using chest X-ray data, the usage of AI-based approaches for COVID-19 identification using chest X-ray data should see substantial growth, giving a more quick and cost-effective approach. The combination of CNN technology and the diagnostic capacity of chest X-ray imaging offers a lot of promise in the fight against COVID-19. Ultimately, the goal is to reduce the strain on healthcare resources and improve patient outcomes by providing medical practitioners with dependable technologies, such as those based on the artificial intelligence (AI), that can aid in real-time monitoring, rapid diagnosis, and patient triage. These advancements enable more effective use of healthcare resources, which benefits patients.</p>

Determination and classification of fetal sex on ultrasound images with deep learning

Today, various prenatal diagnostic methods are used to determine the sex of the fetus. All of these medical methods require intervention by a specialist. The sensitivity of fetal ultrasonography (USG) scanning, which is the most commonly used diagnostic method, is variable and depends on the experience of the sonographer. In this study, an automatic, objective and reliable determination of fetal sex was aimed at using deep transfer learning techniques on USG images. For the study, a dataset containing 4400 fetal USG images, of which sexes were labeled by a gynecologist expert in the field, was created. In the first step, images were classified with fine-tuned convolutional neural networks. Following this classification, the fine-tuned DenseNet201 (ft-DenseNet201) network, which gave the most successful result with an accuracy of 0.9627, was used as the feature extractor network in the second step. Obtained features were classified by Logistic Regression (LR), Linear Support Vector Machine (LSVM), K-Nearest Neighbor (KNN), Decision Tree, Random Forest and AdaBoost algorithms. Among the 10 different classifiers used in the application, ft-DenseNet201 + LSVM (0.9782), ft-DenseNet201 + KNN (0.9727) and ft-DenseNet201 + LR (0.9718) algorithms gave very high accuracy values. This study can be evaluated as an automatic, objective, reliable and new medical method in determination of fetus sex; and can be used as an auxiliary system for specialists and patients by being integrated with USG devices.

Malware image classification: comparative analysis of a fine-tuned CNN and pre-trained models

A crucial part is played by malware detection and classification in ensuring the safety and security of computer systems. In this work, a comprehensive study has been presented for the classification of harmful or malware images that uses a Convolutional Neural Network (CNN) which has been finely tuned and its performance has been compared with five pre-trained models: ResNet50, InceptionResNetV2, VGG16, Xception and InceptionV3. The suggested CNN framework has been trained using the dataset MalImg_9010, consisting of 9,376 grayscale images resized to 128 × 128 pixels. The models have been evaluated based on their F1 score, recall, precision, and accuracy. The experiments that were conducted demonstrate that the fine-tuned CNN model achieves an impressive 0.965 as the F1 score and a 95.57% accuracy. Furthermore, the comparison with pre-trained models reveals the dominance of the presented framework concerning the F1 score and accuracy. The output of the conducted simulation suggests that the fine-tuned CNN approach shows promise for accurate malware image classification. Additionally, the paper discusses potential improvements, such as increasing the number of training epochs and incorporating larger and more diverse malware datasets, including RGB images and a broader range of malware families. The current research article gives valuable observations on various models’ effectiveness for classifying malware images and highlights the future scopes for research incorporating this domain.

Mobile app-based tomato disease identification with fine-tuned convolutional neural networks

Rapid detection and diagnosis of plant diseases is crucial for healthy crop growth and maintaining harvest quality. Many diseases, such as bacterial spot, late blight, and leaf mould, can affect tomato cultivation. In this paper, a mobile app-based system for intelligent disease detection on tomato leaves is developed using convolutional neural networks (CNNs). The study uses transfer learning approaches to fine-tune well-known CNN architectures like AlexNet, ResNet-50, SqueezeNet-1.1, VGG19, and DenseNet-121 on a dataset made up of images of tomato leaf from the PlantVillage dataset as well as self-captured images. Over 95% accuracy is attained by all trained CNN models, with DenseNet-121 leading the pack with a stellar 99.85%. Consequently, DenseNet-121 is proposed for use with a user-friendly mobile application to detect disease in tomato leaves. A robust user interface in English and two important Indian languages makes this application more accessible to a wider user base.

Crow Way: An Optimization Technique for generating the Weight and Bias in Deep CNN

Fine-tuning Convolutional Neural Networks (CNNs) weights and biases are essential for solving difficult machine learning problems. We provide a hybrid optimizations strategy that combines the benefits of Crowd Search (CSA) and Brainstorm Optimization (BSO). CSA-BSO optimizes CNNs. CSA-BSO optimizes CNN weights. Our fitness measures determine each crow's location and speed throughout the CSA phase. The BSO phase mixes optimal replies with random ones to create new solutions. The best new ideas move on. After a certain number of iterations, we alternate between the CSA and BSO phases and pick the best solution as the CNN's final state. We design and test the CSA-BSO approach on many CNN architectures and datasets. Experimentally, the hybrid algorithm outperforms CSA and BSO in convergence time and solution quality. CSA and BSO collaborate to efficiently employ feasible regions and investigate a bigger search field, boosting optimizations. Stable and adaptable, CSA-BSO supports many CNN architectures and data formats. This study's main contribution is the CSA-BSO, a unique hybrid optimizations approach for convolutional neural network weight and bias optimizations. It beats CSA and BSO. The recommended strategy improves CNN picture classification, object recognition, and NLP.

Fine-tuning U-net for medical image segmentation based on activation function, optimizer and pooling layer

<span lang="EN-US">U-net convolutional neural network (CNN) is a famous architecture developed to deal with medical images. Fine-tuning CNNs is a common technique used to enhance their performance by selecting the building blocks which can provide the ultimate results. This paper introduces a method for tuning U-net architecture to improve its performance in medical image segmentation. The experiment is conducted using an x-ray image segmentation approach. The performance of U-net CNN in lung x-ray image segmentation is studied with different activation functions, optimizers, and pooling-bottleneck-layers. The analysis focuses on creating a method that can be applied for tuning U-net, like CNNs. It also provides the best activation function, optimizer, and pooling layer to enhance U-net CNN’s performance on x-ray image segmentation. The findings of this research showed that a U-net architecture worked supremely when we used the LeakyReLU activation function and average pooling layer as well as RMSProb optimizer. The U-net model accuracy is raised from 89.59 to 93.81% when trained and tested with lung x-ray images and uses the LeakyReLU activation function, average pooling layer, and RMSProb optimizer. The fine-tuned model also enhanced accuracy results with three other datasets.</span>

Classification of dental implant systems using cloud-based deep learning algorithm: an experimental study.

This study aimed to evaluate the accuracy and clinical usability of implant system classification using automated machine learning on a Google Cloud platform. Four dental implant systems were selected: Osstem TSIII, Osstem USII, Biomet 3i Os-seotite External, and Dentsply Sirona Xive. A total of 4,800 periapical radiographs (1,200 for each implant system) were collected and labeled based on electronic medical records. Regions of interest were manually cropped to 400×800 pixels, and all images were uploaded to Google Cloud storage. Approximately 80% of the images were used for training, 10% for validation, and 10% for testing. Google automated machine learning (AutoML) Vision automatically executed a neural architecture search technology to apply an appropriate algorithm to the uploaded data. A single-label image classification model was trained using AutoML. The performance of the mod-el was evaluated in terms of accuracy, precision, recall, specificity, and F1 score. The accuracy, precision, recall, specificity, and F1 score of the AutoML Vision model were 0.981, 0.963, 0.961, 0.985, and 0.962, respectively. Osstem TSIII had an accuracy of 100%. Osstem USII and 3i Osseotite External were most often confused in the confusion matrix. Deep learning-based AutoML on a cloud platform showed high accuracy in the classification of dental implant systems as a fine-tuned convolutional neural network. Higher-quality images from various implant systems will be required to improve the performance and clinical usability of the model.

Recognition of flotation froth conditions with k-shot learning and convolutional neural networks

In this study, previous work on k-shot learning in flotation froth image analysis using small sets of froth images, is extended. As before, image synthesis is used to augment these data sets, but in addition, fine-tuning of convolutional neural networks, as well as smaller data sets with as few as 10 samples per class are considered. Two convolutional neural network models, namely AlexNet and GoogLeNet, were compared and the latter demonstrated better performance generally. Both performed better than a traditional approach based on the use of image features derived from local binary patterns. Fine-tuning of the convolutional neural networks markedly improved their performance compared to pure transfer learning without fine-tuning. Moreover, a case study with platinum froth images showed that as few as only 10–30 real images per class are needed to achieve reasonably good classification performance, i.e. the ability to recognize four different classes of froths with an accuracy of 63%–84%, if AlexNet or GoogLeNet with fine-tuning is used. Using 100 real images per class with a fine-tuned GoogLeNet resulted in predictive accuracy approaching 92%. These results are comparable to those achievable with traditional approaches to feature extraction based on training data sets up to two orders of magnitude larger.

A compact CNN model for automated detection of COVID-19 using thorax x-ray images

COVID-19 is an epidemic, causing an enormous death toll. The mutational changing of an RNA virus is causing diagnostic complexities. RT-PCR and Rapid Tests are used for the diagnosis, but unfortunately, these methods are ineffective in diagnosing all strains of COVID-19. There is an utmost need to develop a diagnostic procedure for timely identification. In the proposed work, we come up with a lightweight algorithm based on deep learning to develop a rapid detection system for COVID-19 with thorax chest x-ray (CXR) images. This research aims to develop a fine-tuned convolutional neural network (CNN) model using improved EfficientNetB5. Design is based on compound scaling and trained on the best possible feature extraction algorithm. The low convergence rate of the proposed work can be easily deployed into limited computational resources. It will be helpful for the rapid triaging of victims. 2-fold cross-validation further improves the performance. The algorithm proposed is trained, validated, and testing is performed in the form of internal and external validation on a self-collected and compiled a real-time dataset of CXR. The training dataset is relatively extensive compared to the existing ones. The performance of the proposed technique is measured, validated, and compared with other state-of-the-art pre-trained models. The proposed methodology gives remarkable accuracy (99.5%) and recall (99.5%) for biclassification. The external validation using two different test dataset also give exceptional predictions. The visual depiction of predictions is represented by Grad-CAM maps, presenting the extracted features of the predicted results.

Anomaly Detection of High-Frequency Sensing Data in Transportation Infrastructure Monitoring System Based on Fine-Tuned Model

Anomaly detection has been widely studied in previous studies during recent decades; however, there are still some challenges for high-frequency sensing data. The most challenging task is to deal with the large volume of data in an extremely short time. In previous studies, it has been proved that converting the data into pictures can improve the speed of anomaly detection. However, the training of the image recognition algorithms, such as deep learning models, still needs a long time. Fortunately, the fine-tuned Convolutional Neural Networks (CNN) give us opportunities to detect anomalies in high-frequency data quickly. Thus, a four-stage model is proposed for anomaly detection in high-frequency data. Using a real-world dataset, one designed CNN, four widely used fine-tuned CNN, and two popular machine learning methods are compared by the confusion matrix. Moreover, three ensemble methods are proposed to improve the accuracy of the detection. The results show that majority of voting can improve the overall accuracy by 2.09%. Significantly, the accuracy of minor and outlier class can be increased by ensemble learning which is a challenge in practice.

Jaywalking detection and localization in street scene videos using fine-tuned convolutional neural networks

Jaywalking detection and localization in street scene videos using fine-tuned convolutional neural networks

Refined Automatic Brain Tumor Classification Using Hybrid Convolutional Neural Networks for MRI Scans.

Refined hybrid convolutional neural networks are proposed in this work for classifying brain tumor classes based on MRI scans. A dataset of 2880 T1-weighted contrast-enhanced MRI brain scans are used. The dataset contains three main classes of brain tumors: gliomas, meningiomas, and pituitary tumors, as well as a class of no tumors. Firstly, two pre-trained, fine-tuned convolutional neural networks, GoogleNet and AlexNet, were used for classification process, with validation and classification accuracy being 91.5% and 90.21%, respectively. Then, to improving the performance of the fine-tuning AlexNet, two hybrid networks (AlexNet-SVM and AlexNet-KNN) were applied. These hybrid networks achieved 96.9% and 98.6% validation and accuracy, respectively. Thus, the hybrid network AlexNet-KNN was shown to be able to apply the classification process of the present data with high accuracy. After exporting these networks, a selected dataset was employed for testing process, yielding accuracies of 88%, 85%, 95%, and 97% for the fine-tuned GoogleNet, the fine-tuned AlexNet, AlexNet-SVM, and AlexNet-KNN, respectively. The proposed system would help for automatic detection and classification of the brain tumor from the MRI scans and safe the time for the clinical diagnosis.