RESUMEN

This paper presents an ensemble of pre-trained models for the accurate classification of endoscopic images associated with Gastrointestinal (GI) diseases and illnesses. In this paper, we propose a weighted average ensemble model called GIT-NET to classify GI-tract diseases. We evaluated the model on a KVASIR v2 dataset with eight classes. When individual models are used for classification, they are often prone to misclassification since they may not be able to learn the characteristics of all the classes adequately. This is due to the fact that each model may learn the characteristics of specific classes more efficiently than the other classes. We propose an ensemble model that leverages the predictions of three pre-trained models, DenseNet201, InceptionV3, and ResNet50 with accuracies of 94.54%, 88.38%, and 90.58%, respectively. The predictions of the base learners are combined using two methods: model averaging and weighted averaging. The performances of the models are evaluated, and the model averaging ensemble has an accuracy of 92.96% whereas the weighted average ensemble has an accuracy of 95.00%. The weighted average ensemble outperforms the model average ensemble and all individual models. The results from the evaluation demonstrate that utilizing an ensemble of base learners can successfully classify features that were incorrectly learned by individual base learners.

RESUMEN

Introduction: Social media platforms such as Facebook, LinkedIn, Twitter, among others have been used as tools for staging protests, opinion polls, campaign strategy, medium of agitation and a place of interest expression especially during elections. Aim: In this work, a Natural Language Processing framework is designed to understand Nigeria 2023 presidential election based on public opinion using Twitter dataset. Methods: Two million tweets with 18 features were collected from Twitter containing public and personal tweets of the three top contestants - Atiku Abubakar, Peter Obi and Bola Tinubu - in the forthcoming 2023 Presidential election. Sentiment analysis was performed on the preprocessed dataset using three machine learning models namely: Long Short-Term Memory (LSTM) Recurrent Neural Network, Bidirectional Encoder Representations from Transformers (BERT) and Linear Support Vector Classifier (LSVC) models. This study spanned ten weeks starting from the candidates' declaration of intent to run for Presidency. Results: The sentiment models gave an accuracy, precision, recall, AUC and f-measure of 88%, 82.7%, 87.2%, 87.6% and 82.9% respectively for LSTM; 94%, 88.5%, 92.5%, 94.7% and 91.7% respectively for BERT and 73%, 81.4%, 76.4%, 81.2% and 79.2% respectively for LSVC. Result also showed that Peter Obi has the highest total impressions the highest positive sentiments, Tinubu has the highest network of active friends while Atiku has the highest number of followers. Conclusion: Sentiment analysis and other Natural Language Understanding tasks can aid in the understanding of the social media space in terms of public opinion mining. We conclude that opinion mining from Twitter can form a general basis for generating insights for election as well as modeling election outcomes.

RESUMEN

Cloud computing is a long-standing dream of computing as a utility, where users can store their data remotely in the cloud to enjoy on-demand services and high-quality applications from a shared pool of configurable computing resources. Thus, the privacy and security of data are of utmost importance to all of its users regardless of the nature of the data being stored. In cloud computing environments, it is especially critical because data is stored in various locations, even around the world, and users do not have any physical access to their sensitive data. Therefore, we need certain data protection techniques to protect the sensitive data that is outsourced over the cloud. In this paper, we conduct a systematic literature review (SLR) to illustrate all the data protection techniques that protect sensitive data outsourced over cloud storage. Therefore, the main objective of this research is to synthesize, classify, and identify important studies in the field of study. Accordingly, an evidence-based approach is used in this study. Preliminary results are based on answers to four research questions. Out of 493 research articles, 52 studies were selected. 52 papers use different data protection techniques, which can be divided into two main categories, namely noncryptographic techniques and cryptographic techniques. Noncryptographic techniques consist of data splitting, data anonymization, and steganographic techniques, whereas cryptographic techniques consist of encryption, searchable encryption, homomorphic encryption, and signcryption. In this work, we compare all of these techniques in terms of data protection accuracy, overhead, and operations on masked data. Finally, we discuss the future research challenges facing the implementation of these techniques.

Asunto(s)

RESUMEN

Blood is key evidence to reconstruct crime scenes in forensic sciences. Blood identification can help to confirm a suspect, and for that reason, several chemical methods are used to reconstruct the crime scene however, these methods can affect subsequent DNA analysis. Therefore, this study presents a non-destructive method for bloodstain identification using Hyperspectral Imaging (HSI, 397-1000 nm range). The proposed method is based on the visualization of heme-components bands in the 500-700 nm spectral range. For experimental and validation purposes, a total of 225 blood (different donors) and non-blood (protein-based ketchup, rust acrylic paint, red acrylic paint, brown acrylic paint, red nail polish, rust nail polish, fake blood, and red ink) samples (HSI cubes, each cube is of size 1000 × 512 × 224, in which 1000 × 512 are the spatial dimensions and 224 spectral bands) were deposited on three substrates (white cotton fabric, white tile, and PVC wall sheet). The samples are imaged for up to three days to include aging. Savitzky Golay filtering has been used to highlight the subtle bands of all samples, particularly the aged ones. Based on the derivative spectrum, important spectral bands were selected to train five different classifiers (SVM, ANN, KNN, Random Forest, and Decision Tree). The comparative analysis reveals that the proposed method outperformed several state-of-the-art methods.

Asunto(s)

RESUMEN

Today, accurate and automated abnormality diagnosis and identification have become of paramount importance as they are involved in many critical and life-saving scenarios. To accomplish such frontiers, we propose three artificial intelligence models through the application of deep learning algorithms to analyze and detect anomalies in human heartbeat signals. The three proposed models include an attention autoencoder that maps input data to a lower-dimensional latent representation with maximum feature retention, and a reconstruction decoder with minimum remodeling loss. The autoencoder has an embedded attention module at the bottleneck to learn the salient activations of the encoded distribution. Additionally, a variational autoencoder (VAE) and a long short-term memory (LSTM) network is designed to learn the Gaussian distribution of the generative reconstruction and time-series sequential data analysis. The three proposed models displayed outstanding ability to detect anomalies on the evaluated five thousand electrocardiogram (ECG5000) signals with 99% accuracy and 99.3% precision score in detecting healthy heartbeats from patients with severe congestive heart failure.

Asunto(s)

RESUMEN

Retinal vessel segmentation (RVS) is a significant source of useful information for monitoring, identification, initial medication, and surgical development of ophthalmic disorders. Most common disorders, i.e., stroke, diabetic retinopathy (DR), and cardiac diseases, often change the normal structure of the retinal vascular network. A lot of research has been committed to building an automatic RVS system. But, it is still an open issue. In this article, a framework is recommended for RVS with fast execution and competing outcomes. An initial binary image is obtained by the application of the MISODATA on the preprocessed image. For vessel structure enhancement, B-COSFIRE filters are utilized along with thresholding to obtain another binary image. These two binary images are combined by logical AND-type operation. Then, it is fused with the enhanced image of B-COSFIRE filters followed by thresholding to obtain the vessel location map (VLM). The methodology is verified on four different datasets: DRIVE, STARE, HRF, and CHASE_DB1, which are publicly accessible for benchmarking and validation. The obtained results are compared with the existing competing methods.

Asunto(s)