Search

Your search keyword '"Bernardo, Lucas Salvador"' showing total 5 results
Start Over Author "Bernardo, Lucas Salvador"
5 results on '"Bernardo, Lucas Salvador"'

1. VGG11 Parkinson’s Disease Detection Based on Voice Attributes

Author

Bernardo, Lucas Salvador, Damaševičius, Robertas, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Guarda, Teresa, editor, Portela, Filipe, editor, and Augusto, Maria Fernanda, editor

Published
2022
Full Text
View/download PDF

2. A Hybrid Two–Stage Squeezenet and Support Vector Machine System for Parkinson’s Disease Detection Based on Handwritten Spiral Patterns

Author

Bernardo Lucas Salvador, Damaševičius Robertas, De Albuquerque Victor Hugo C., and Maskeliūnas Rytis

Subjects
parkinson’s disease, spirography, convolutional neural network, deep learning, Mathematics, QA1-939, Electronic computers. Computer science, QA75.5-76.95
Abstract

Parkinson’s disease (PD) is the second most common neurological disorder in the world. Nowadays, it is estimated that it affects from 2% to 3% of the global population over 65 years old. In clinical environments, a spiral drawing task is performed to help to obtain the disease’s diagnosis. The spiral trajectory differs between people with PD and healthy ones. This paper aims to analyze differences between handmade drawings of PD patients and healthy subjects by applying the SqueezeNet convolutional neural network (CNN) model as a feature extractor, and a support vector machine (SVM) as a classifier. The dataset used for training and testing consists of 514 handwritten draws of Archimedes’ spiral images derived from heterogeneous sources (digital and paper-based), from which 296 correspond to PD patients and 218 to healthy subjects. To extract features using the proposed CNN, a model is trained and 20% of its data is used for testing. Feature extraction results in 512 features, which are used for SVM training and testing, while the performance is compared with that of other machine learning classifiers such as a Gaussian naive Bayes (GNB) classifier (82.61%) and a random forest (RF) (87.38%). The proposed method displays an accuracy of 91.26%, which represents an improvement when compared to pure CNN-based models such as SqueezeNet (85.29%), VGG11 (87.25%), and ResNet (89.22%).

Published
2021
Full Text
View/download PDF

4. PKDIAGNOSTIC: A COMPUTATIONAL TOOL TO ASSIST PAKINSON'S DISEASE DIAGNOSIS. (Inglês)

Author

Bernardo, Lucas Salvador, Albuquerque, Victor Hugo Costa de, Pereira, Clayton Reginaldo, Carvalho, Fernanda Martins Maia, Felix, John Hebert da Silva, and Rebouças Filho, Pedro Pedrosa

Subjects
Processamento de imagens, Doença de Parkinson, Software
Abstract

Made available in DSpace on 2022-06-23T00:04:03Z (GMT). No. of bitstreams: 0 Previous issue date: 2018-06-13 The Parkinson's Disease (PD) consists in a neurodegenerative disorder that affects around 10 million people in the world. This condition is slightly more prevalent in males. PD is characterized by the loss of neurons in a region of the brain, known as Substantia Nigra. The neurons of this region, are responsible for synthesizing the neurotransmitter Dopamine, and a decrease in the production of this substance may cause the motor symptoms that are characteristic of the disease. In order to obtain a definitive diagnosis, the patient's medical history is analysed and the subject is submitted to a series of clinical exams. One of the exams, that take place in the clinical environment, consists of asking the patient to create a series of specific drawings. This work is based in asking the patients to make this drawings using a software developed for this specific purpose. The drawings will then pass through a series of image processing methods in order to "reduce noises and extract the characteristics of each drawing being those 11 metrics that will be stored into a .Excel file. Machine learning techniques such as: OPF, SVM and Naive Bayes, will use the dataset in search for learning about the characteristics for the process of classification of the individuals that were labelled by the classes (Sick and Healthy). Keywords: Machine Learning. Image processing. Software. Parkinson disease. A Doença de Parkinson (DP) consiste em uma disordem neurodegenerativa que afeta atualmente cerca de 10 milhões de pessoas no mundo, sendo homens os mais propícios a apresentarem a doença. A DP é caracterizada pela perda de neurônios na região do cérebro chamada de Substância Negra. Os neurônios dessa região, são os responsáveis pela sintetização do neurotransmissor Dopamina, cuja a redução nessa região pode ocasionar os sintomas motores tão caracteristicos da doença. Para a realização do diagnóstico da DP, é observado o histórico médico do paciente e realiza-se uma série de exames clinicos afim de estabelecer o diagnostico definitivo da doença e sua mensuração nas escalas. Um dos exames a ser realizado no consultório médico, consiste em pedir que o paciente realize uma série de desenhos específicos, onde são observados os desvios cometidos pelo paciente. Estre trabalho, consiste em fazer com que o paciente realize esses desenhos, utilizando o mouse do computador, no software desenvolvido para este trabalho. Os desenhos passam então por uma série de técnicas de processamento de imagem, afim de reduzir ruidos e extrair as caracteristicas de cada desenho, sendo essas, 11 métricas que são armazenadas em arquivos Excel. Técnicas de Aprendizado de Máquina como Floreta de Caminhos Ótimos (Optimum Path Forest - OPF), Support Vector Machines (SVM) e Classificador Bayesiano (Bayes) são aplicados nos conjuntos de dados em busca do aprendizado sobre as características para o processo de classificação dos indivíduos que foram rotulados pelas classes (Pacientes e Saudáveis). Palavras-chave: Aprendizagem de máquina. Processamento de imagem. Software. Doença de Parkinson.

Published
2018

Catalog

Books, media, physical & digital resources
See catalog results