A5056219055- Mosquito-borne diseases and control
- Gene expression and cancer classification
- Machine Learning and Data Classification
- Viral Infections and Vectors
- Machine Learning in Bioinformatics
- Evolutionary Algorithms and Applications
- AI in cancer detection
- Hepatitis C virus research
- Model Reduction and Neural Networks
- Tryptophan and brain disorders
- Image Processing Techniques and Applications
- Genomics and Rare Diseases
- Epilepsy research and treatment
- Renal and related cancers
- Malaria Research and Control
- Neural Networks and Applications
- COVID-19 and Mental Health
Istituto delle Scienze Neurologiche di Bologna
2024
Texas A&M University
2019-2021
Mitchell Institute
2021
Instituto Federal de Educação, Ciência e Tecnologia de Pernambuco
2018
Dengue has become one of the most important worldwide arthropod-borne diseases. phenotypes are based on laboratorial and clinical exams, which known to be inaccurate. Objective: We present a machine learning approach for prediction dengue fever severity solely human genome data. Methods: One hundred two Brazilian patients controls were genotyped 322 innate immunity single nucleotide polymorphisms (SNPs). Our model uses support vector algorithm find optimal loci classification subset then an...
Physics-informed neural networks (PINN) have recently emerged as a promising application of deep learning in wide range engineering and scientific problems based on partial differential equation (PDE) models. However, evidence shows that PINN training by gradient descent displays pathologies often prevent convergence when solving PDEs with irregular solutions. In this paper, we propose the use particle swarm optimization (PSO) approach to train PINNs. The resulting PSO-PINN algorithm not...
Dengue is considered one of the most challenging public health threats in world. Infection may be clinically asymptomatic but can result severe forms. The indoleamine 2,3 dioxygenase (IDO) gene encodes first enzymes kynurenine pathway. This study aimed to verify association between G2431A IDO1 single nucleotide polymorphism (SNP) (rs3739319) and dengue fever development. We included 299 dengue-infected individuals 96 dengue-free controls. collected clinical diagnostic test data divided...
For most diseases, building large databases of labeled genetic data is an expensive and time-demanding task. To address this, we introduce Generative Adversarial Networks (gGAN), a semi-supervised approach based on innovative GAN architecture to create synthetic sets starting with small amount unlabeled data. Our goal mechanism able increase the sample size generalize learning over different populations while keeping awareness quality its own predictions. The proposed model achieved...
Dengue has become one of the most important worldwide arthropodborne diseases around world. Here, hundred and two Brazilian dengue virus (DENV) III patients controls were genotyped for 322 innate immunity gene loci. All biological data (including age, sex genome background) analyzed using Machine Learning techniques to discriminate tendency severe phenotype development. Our current approach produces median values accuracy greater than 86%, with sensitivity specificity over 98% 51%,...
For most diseases, building large databases of labeled genetic data is an expensive and time-demanding task. To address this, we introduce Generative Adversarial Networks (gGAN), a semi-supervised approach based on innovative GAN architecture to create synthetic sets starting with small amount unlabeled data. Our goal determine the propensity new individual develop severe form illness from their profile alone. The proposed model achieved satisfactory results using real different datasets...