Geraldo Braz Junior

Classification of breast tissues using Moran's index and Geary's coefficient as texture signatures and SVM

Female breast cancer is the major cause of cancer-related deaths in western countries. Efforts in computer vision have been made in order to help improving the diagnostic accuracy by radiologists. In this paper, we present a methodology that uses Morans index and Gearys coefficient measures in breast tissues extracted from mammogram images. These measures are used as input features for a support vector machine classifier with the purpose of distinguishing tissues between normal and abnormal cases as well as classifying them into benign and malignant cancerous cases. The use of both proposed techniques showed to be very promising, since we obtained an accuracy of 96.04% and Az ROC of 0.946 with Gearys coefficient and an accuracy of 99.39% and Az ROC of 1 with Morans index to discriminate tissues in mammograms as normal or abnormal. We also obtained accuracy of 88.31% and Az ROC of 0.804 with Gearys coefficient and accuracy of 87.80% and Az ROC of 0.89 with Morans index to discriminate tissues in mammograms as benign and malignant.

A mass classification using spatial diversity approaches in mammography images for false positive reduction

Abstract Breast cancer is configured as a public health problem that affects mainly women population. One of the main ways of prevention is through screening mammography. The interpretation made by the physician is a repetitive task because of a low contrast image and the examination of several exams. So, computer systems have been proposed to aid detection step and helps physician, with the aim to increase sensitivity at the same time that reduces invasive procedures. Although these systems had improved the sensitivity of the original examination of mammography, they also generate a lot of false positives. This paper presents a methodology for reducing false positives by analyzing the diversity of approaches with improved spatial decomposition. After experiments the results reaches a high level of sensitivity at the same time promote a high rate of reduction of false positives.

AGITO: Virtual Reality Environment for Power Systems Substations Operators Training

This paper presents the architecture and development of a virtual reality environment for powers systems substations operators training. The proposal intents to reduce the training time for new operators and increase the effectiveness of the continuous training of operators. Using the simulation, the operator can interact with a virtual reality interface (immersive or non immersive) viewing the state of the power system captured through the supervisory system and acting through the virtual environment operating the substation, without exposing the system to dangerous situations, avoiding the occurrence of injuries of any kind. Also, the training sessions can be analyzed offline by an instructor.

Lung Nodules Classification in CT Images Using Simpson's Index, Geometrical Measures and One-Class SVM

In this paper, we present the Simpsons Index, a feature used in Spatial Analysis and in Biology, specifically in Ecology to determine the homogeneity or heterogeneity of a certain species. This index will be investigated as a promising feature, since little observation has been done on the application of these features for the analysis of medical images, with three geometrical features, in the characterization of lung nodules as benign or malignant. Using One-Class SVM for classification we obtained sensibility rates of 100%, specificity 100% and accuracy of 100%.

Texture analysis of masses malignant in mammograms images using a combined approach of diversity index and local binary patterns distribution

A methodology to texture analysis of masses in digitized mammography is proposed.Our methodology uses only the texture analysis for recognition.We provide the specialist bigger support to the diagnosis of breast cancer.We cooperate to a more precise diagnosis and support in the medical intervention. A World Health Organization (WHO) report estimates that in 2015, at least 561 thousand women will die of breast cancer. Although breast cancer is considered a disease of the developed world, nearly 50% of the cases and 58% of the deaths occur in the less developed countries. A mammogram is a way of discovering not just the palpable tumors that cause cancer but also other lesions that are not perceived during the physical examination performed by the expert physician or during self-exams; however, it is known that this exam is targeted for women after the age of 40 because age is one of the factors that can cause great variations in sensitivity during the exam. Besides the patients age, the experts experience and the quality of the images obtained during the exam are decisive factors in the detection of breast cancer. This work presents two novelties. The first is the use of Local Binary Patterns (LBPs) to generate a representation of a Region of Interest (ROI) image. Over this representation, we generate other representations using techniques such as image histograms, gray-level co-occurrence matrices (GLCMs) and gray-level run-length matrices (GLRLMs). These representations allow texture analysis through several perspectives. The second novelty uses these representations as input to the application of indexes adapted from ecology (Shannon, McIntosh, Simpson, Gleason and Menhinick) as texture descriptors. Based on this strategy, we analyze mammographic image textures to classify regions of these images as benign or malignant using a Support Vector Machine (SVM). The best result achieved was of 88.31% accuracy, 85% sensitivity, 91.89% specificity, a positive probability ratio of 10.48, a negative probability ratio of 0.16, and an area under the Receiver Operating Characteristic (ROC) curve of 0.88, obtained through the Shannon index. We believe that the proposed method, with some adaptations, may also be used for image texture analysis of several different lesions such as lung nodules, glaucoma and prostates. This belief is based on the achieved results and the methods simplicity.

Texture based on geostatistic for glaucoma diagnosis from fundus eye image

Glaucoma is an ocular disorder that can permanently damage patient vision. Initially, it reduces the visual field, and may cause blindness. Effective methods for early detection is crucial for avoiding significant damages of the patient vision. The use of CAD (Computer-Aided Detection) and CADx (Computer-Aided Diagnosis) systems has contributed to increase the chances of detection and precise diagnoses, assisting experts’ decision making on treatment regarding glaucoma. This paper proposes a method that analyzes the texture of the optical disk image region to diagnose glaucoma. Such analysis is done using the Local Binary Pattern (LBP) to represent the optic disk region, and geostatistical functions to describe texture patterns. The obtained texture features are used for classification based on Support Vector Machine. The proposed method presented as best results a sensitivity of 95%, accuracy of 91% and specificity of 88% in the diagnosis of glaucoma. The method has proved to be promising in assisting glaucoma diagnosis.

Visualization of Power Systems Based on Panoramic Augmented Environments

Interactive and contextualized applications have been aimed to support professionals in the field of engineering in order to deal with the difficult of understanding technical diagrams related to power systems when there is a vast amount of information represented on them. Augmented Reality (AR) and Static Panoramic Augmented Environments have been considered promising approaches to build solutions in this field. This paper presents an application that uses Panoramic Augmented Environments to extends the way information is shown to power systems operators supporting data interpretation, monitoring and manipulation. This application is connected with a real power system database and uses images from substations of CHESF, a Brazilian power systems company.

COMPARISON OF SUPPORT VECTOR MACHINES AND BAYESIAN NEURAL NETWORKS PERFORMANCE FOR BREAST TISSUES USING GEOSTATISTICAL FUNCTIONS IN MAMMOGRAPHIC IMAGES

Female breast cancer is a major cause of deaths in occidental countries. Computer-aided Detection (CAD) systems can aid radiologists to increase diagnostic accuracy. In this work, we present a comparison between two classifiers applied to the separation of normal and abnormal breast tissues from mammograms. The purpose of the comparison is to select the best prediction technique to be part of a CAD system. Each region of interest is classified through a Support Vector Machine (SVM) and a Bayesian Neural Network (BNN) as normal or abnormal region. SVM is a machine-learning method, based on the principle of structural risk minimization, which shows good performance when applied to data outside the training set. A Bayesian Neural Network is a classifier that joins traditional neural networks theory and Bayesian inference. We use a set of measures obtained by the application of the semivariogram, semimadogram, covariogram, and correlogram functions to the characterization of breast tissue as normal or abnormal. The results show that SVM presents best performance for the classification of breast tissues in mammographic images. The tests indicate that SVM has more generalization power than the BNN classifier. BNN has a sensibility of 76.19% and a specificity of 79.31%, while SVM presents a sensibility of 74.07% and a specificity of 98.77%. The accuracy rate for tests is 78.70% and 92.59% for BNN and SVM, respectively.

Classification of breast tissues in mammogram images using ripley's K function and support vector machine

Female breast cancer is a major cause of death in western countries. Several computer techniques have been developed to aid radiologists to improve their performance in the detection and diagnosis of breast abnormalities. In Point Pattern Analysis, there is a statistic known as Ripleys K function that is frequently applied to Spatial Analysis in Ecology, like mapping specimens of plants. This paper proposes a new way in applying Ripleys K function in order to distinguish Mass and Non-Mass tissues from mammogram images. The features of each image are obtained through the calculate of that function. Then, the samples gotten are classified through a Support Vector Machine (SVM) as Mass or Non-Mass tissues. SVM is a machinelearning method, based on the principle of structural risk minimization, which performs well when applied to data outside the training set. Another way of computing Ripleys K function, using concentric rings instead of a circle, is also examined. The best result achieved was 94.25% of accuracy, 94.59% of sensitvity and 94.00% of specificity.

Sign Language Recognition Based on 3D Convolutional Neural Networks.

The inclusion of disabled people is still a recurring problem throughout the world. For the hearing impaired, the barrier imposed by the sign language spoken by a small part of the population imposes limitations that interfere in the quality of life of these people. The popularization or even automation of sign language recognition can take their lives to a higher level. Understanding the importance of sign language recognition for the hearing impaired we propose a 3D CNN architecture for the recognition of 64 classes of gestures from Argentinian Sign Language (LSA64). We demonstrate the efficiency of the method when compared to traditional methods based on hand-crafted features and that its results outperform most deep learning-based work reaching 93.9% of accuracy.