Cicero Ferreira Fernandes Costa Filho

Automatic identification of mycobacterium tuberculosis with conventional light microscopy

This article presents an automatic identification method of mycobacterium tuberculosis with conventional microscopy images based on Red and Green color channels using global adaptive threshold segmentation. Differing from fluorescence microscopy, in the conventional microscopy the bacilli are not easily distinguished from the background. The key to the bacilli segmentation method employed in this work is the use of Red minus Green (R-G) images from RGB color format. In this image, the bacilli appear as white regions on a dark background. Some artifacts are present in the (R-G) segmented image. To remove them we used morphological, color and size filters. The best sensitivity achieved was about 76.65%. The main contribution of this work was the proposal of the first automatic identification method of tuberculosis bacilli for conventional light microscopy.

Iris segmentation exploring color spaces

This paper describes a new method for iris segmentation using HSI and RGB color spaces. The outer and inner boundaries of the iris are extracted using the k-means unsupervised clusterization method. For the outer boundary detection the best results is obtained using as input variables of the clusterization method the red and green components of the RGB space. The final outer boundary is detected through the application of a modified version of the Hough Transform. For the inner boundary detection the best result is obtained using as input variables of the clusterization method the hue component of the HSI space. The method was tested with images of section 1 and 2 of the UBIRIS image database. For the section 1 the overall percent accuracy achieved was 97.6%. For section 2 the overall percent accuracy achieved was 93.7%. Some examples of the iris segmentation are provided in the results. A simple method for iris extraction associated with successful results obtained with noise iris images is the main contribution of this paper.

Evaluation of autofocus functions of conventional sputum smear microscopy for tuberculosis

This article presents a systematic analysis of focus functions in conventional sputum smear microscopy for tuberculosis. This is the first step in the development of automatic microscopy. Nine autofocus functions are analyzed in a set of 1200 images with varying degrees of content density. These functions were evaluated using quantitative procedures. The main accomplishment of this work was to show that an autofocus function based on variance measures produced the best results for tuberculosis images.

Automatic identification of tuberculosis mycobacterium

Introduction According to the Global TB control report of 2013, “Tuberculosis (TB) remains a major global health problem. In 2012, an estimated 8.6 million people developed TB and 1.3 million died from the disease. Two main sputum smear microscopy techniques are used for TB diagnosis: Fluorescence microscopy and conventional microscopy. Fluorescence microscopy is a more expensive diagnostic method because of the high costs of the microscopy unit and its maintenance. Therefore, conventional microscopy is more appropriate for use in developing countries. Methods This paper presents a new method for detecting tuberculosis bacillus in conventional sputum smear microscopy. The method consists of two main steps, bacillus segmentation and post-processing. In the first step, the scalar selection technique was used to select input variables for the segmentation classifiers from four color spaces. Thirty features were used, including the subtractions of the color components of different color spaces. In the post-processing step, three filters were used to separate bacilli from artifact: a size filter, a geometric filter and a Rule-based filter that uses the components of the RGB color space. Results In bacillus identification, an overall sensitivity of 96.80% and an error rate of 3.38% were obtained. An image database with 120-sputum-smear microscopy slices of 12 patients with objects marked as bacillus, agglomerated bacillus and artifact was generated and is now available online. Conclusions The best results were obtained with a support vector machine in bacillus segmentation associated with the application of the three post-processing filters.

Using Constraint Satisfaction Problem approach to solve human resource allocation problems in cooperative health services

In developing countries, the increasing utilization of health services, due to a great life expectancy, is followed by a reduction in incomes from the public health system and from private insurance companies, to the payment of medical procedures. Beyond this scenery, it is mandatory an effective hospital cost control though the utilization of planning tools. This work is intended to contribute to the reduction of hospital costs, proposing a new tool for planning human resources utilization in hospital plants. Specifically, it is proposed a new tool for human resources allocation in health units. The solution to the allocation problem uses the CSP technique (Constraint Satisfaction Problem) associated with the backtracking search algorithm. With the objective of enhancing the backtracking search algorithm performance a new heuristics is proposed. Through some simulations the performance of the proposed heuristics is compared to the other heuristics previously published in literature: remaining minimum values, forward checking and grade heuristics. Another important contribution of this work is the mathematical modeling of the constraints, that could be unary, multiple, numeric and implicit constraints. In the results it is presented a case study of a human resource allocation in a cooperative health service. Based on the results, it is proposed that for a real allocation problems solution, the best approach is to combine the remaining minimum values heuristics with the grade heuristics, to select the best unit allocation to be filled, and then use the proposed heuristic to select the best physician to the chosen unit allocation. This association shows a satisfactory result for the human resource allocation problem of the case study, with an algorithm convergence time of 46.7min with no backtracks. The same problem when manually resolved took about more than 50h.

Applying a novelty filter as a matching criterion to iris recognition for binary and real-valued feature vectors

The main contributions of this paper are proposing a robust matching measure that employs multiple images of a subject to enroll an iris and that can be used with both types of feature vectors, real-valued and binary feature vectors. The first one is obtained using wavelet transforms and pixel intensity images and the second one using binary wavelet coefficients. The validation of the new matching measure proposed was done considering two utilization modes of the biometric system: verification mode and identification mode. The performance of the new matching measure is comparable to other published results. The vector with lower size was the one that uses binary wavelet coefficients, with only 10 bytes of information. Other authors reported binary feature vector sizes much greater than this one. Iris codification with vectors of lower sizes accounts for the construction of iris recognition embedded systems.

A sputum smear microscopy image database for automatic bacilli detection in conventional microscopy

In this work, we present an image database for automatic bacilli detection in sputum smear microscopy. The database comprises two parts. The first one, called the autofocus database, contains 1200 images with resolution of 2816 × 2112 pixels. This database was obtained from 12 slides, with 10 fields per slide. Each stack is composed of 10 images, with the fifth image in focus. The second one, called the segmentation and classification database, contains 120 images with resolution of 2816×2112 pixels. This database was obtained from 12 slices, with 10 fields per slice. In both databases, the images were acquired from fields of slides stained with the standard Kinyoun method. In both databases, accordingly to the background content, the images were classified as belonging to high background content or low background content. In all 120 images of segmentation and classification database, the identified objects were enclosed within a geometric shape by a trained technician. A true bacillus was enclosed in a circle. An agglomerated bacillus was enclosed by a rectangle and a doubtful bacillus (the image focus or geometry does not allow a clear identification of the object) was enclosed by a polygon. These marked objects could be used as a gold standard to calculate the accuracy, sensitivity and specificity of bacilli recognition.

Detecting Natural Gas Leaks Using Digital Images and Novelty Filters

This paper presents a new technique for detecting natural gas leaks in the oil and gas industry. More precisely, the detection is done in wellheads of industry installations. In the literature, other methods are already used, but with some drawbacks. One technique detects gas leaks measuring the CH4 concentration through the principle of catalytic combustion but suffers from reduced life span and a narrow detection range of sensors. Another technique that measures infrared spectrum absorption suffers from high false negative values in the presence of steam. The technique proposed in this study uses radiation in the visible range that can be captured through CCD cameras already present in Closed-Circuit Television systems used to monitor wells. The proposed method uses the novelty filter concept to detect the leak and to identify the region where it occurs. The proposed technique is a pioneering study of natural gas detection with CCD in visible range. The results presented are promising, showing sensitivity and specificity equal to 100%.

Breast tumor classification in ultrasound images using support vector machines and neural networks

Introduction The use of tools for computer-aided diagnosis (CAD) has been proposed for detection and classification of breast cancer. Concerning breast cancer image diagnosing with ultrasound, some results found in literature show that morphological features perform better than texture features for lesions differentiation, and indicate that a reduced set of features performs better than a larger one. Methods This study evaluated the performance of support vector machines (SVM) with different kernels combinations, and neural networks with different stop criteria, for classifying breast cancer nodules. Twenty-two morphological features from the contour of 100 BUS images were used as input for classifiers and then a scalar feature selection technique with correlation was used to reduce the features dataset. Results The best results obtained for accuracy and area under ROC curve were 96.98% and 0.980, respectively, both with neural networks using the whole set of features. Conclusion The performance obtained with neural networks with the selected stop criterion was better than the ones obtained with SVM. Whilst using neural networks the results were better with all 22 features, SVM classifiers performed better with a reduced set of 6 features.

Handwritten Digit Recognition Using SVM Binary Classifiers and Unbalanced Decision Trees

In this work, we use SVM binary classifiers coupled with a binary classifier architecture, an unbalanced decision tree, for handwritten digit recognition. According to input variables, two classifiers were trained and tested. One using digit characteristics and the other using the whole image as input variables. Developed recently, the unbalanced decision tree architecture provides a simple structure for a multiclass classifier using binary classifiers. In this work, using the whole image as input, 100% handwritten digit recognition accuracy was obtained in the MNIST database. These are the best results published in the literature for the MNIST database.