João Antonio Zuffo

Segmentation of breast tumors in mammograms by fuzzy region growing

Segmentation of tumor regions in mammograms is not easy due to the low contrast and the fuzzy nature of the boundaries of malignant tumors. General image segmentation procedures do not consider the uncertainty present around the boundaries of a tumor region. In this paper we present a segmentation method based on fuzzy region growing. The procedure starts with a seed pixel, and uses a fuzzy membership function based upon statistical measures of the region being grown. Results of testing with several mammograms indicate that the method can provide boundaries of tumors close to those drawn by an expert radiologist. The regions obtained preserve the transition information present around tumor boundaries. Statistical measures computed from the resulting regions have shown the potential to classify masses and tumors as benign or malignant.

Segmentation of breast tumors in mammograms using fuzzy sets

Defining criteria to determine precisely the boundaries of masses in mammograms is a difficult task. The problem is com- pounded by the fact that most malignant tumors possess fuzzy boundaries with a slow and extended transition from a dense core region to the surrounding less-dense tissues. We propose two seg- mentation methods that incorporate fuzzy concepts. The first method determines the boundary of a mass or tumor by region growing after a preprocessing step based on fuzzy sets to enhance the region of interest (ROI). Contours provided by the method have demonstrated a good match with the contours drawn by a radiolo- gist, as indicated by good agreement between the two sets of con- tours for 47 mammograms. The second segmentation method is a fuzzy region-growing method that takes into account the uncertainty present around the boundaries of tumors. The difficult step of decid- ing on a crisp boundary is obviated in the proposed method. Mea- sures of inhomogeneity computed from the pixels present in a suit- ably defined fuzzy ribbon have indicated potential use in classifying the masses and tumors as benign or malignant, with a sensitivity of 0.8 and a specificity of 0.9.

Detection of Breast Tumor Boundaries Using ISO-Intensity Contours and Dynamic Thresholding

Mammograms are, at times, difficult to interpret: developing signs of cancer may be masked by superimposed tissues. Additional diagnostic procedures may be recommended when the original mammogram is equivocal.

Fuzzy fusion operators to combine results of complementary medical image segmentation techniques

The detection of masses and tumors in a mammogram is a difficult problem that could benefit from the use of multiple approaches. We propose an abstract concept of information fusion based on a finite automaton and fuzzy sets to integrate and evaluate results of multiple image segmentation procedures. We give examples on how the method can be applied to the problem of mammographic image segmentation, combining results of region growing and closed-contour detection techniques. We also propose a measure of fuzzyness to assess the agreement between a segmented region and a reference contour. Application of the fusion technique to breast tumor detection in mammograms indicates that the fusion results agree with the reference contours provided by a radiologist to a higher extent than the results of the individual methods.

Fuzzy fusion of results of medical image segmentation

We propose an abstract concept of data fusion based on finite automata and fuzzy sets to integrate and evaluate different sources of information, in particular results of multiple image segmentation procedures. We give an example of how the method may be applied to the problem of mammographic image segmentation to combine results of region growing and closed- contour detection techniques. We further propose a measure of fuzziness to assess the agreement between a segmented region and a reference contour. Results of application to breast tumor detection in mammograms indicate that the fusion results agree with reference contours provided by a radiologist to a higher extent than the results of the individual methods.

Platform to enforce multiple access control policy in grid hosting environment

Computational grid aims to get a better improvement of the existents resources by the use of distributed and flexible systems. However, the utilization of this system brings new challenges in relation to security, requiring an access control service that can be adequate for different conditions of heterogeneous environments and allows its integration with pre existents mechanism. This work presents a flexible platform that integrates multiple policy models to enforce access control in grid hosting environments by controlling the actions of submitted applications. The results of the operational test show how the current platform can realize access control based on IDS systems and enforces a Least Privilege policy model.

A programming environment for high-performance volume visualization applications

Abstract Volume visualization is an important tool in many scientific applications, requiring intensive processing and dealing with large amounts of data. For interactive and specialized applications one needs to use high-performance computing techniques. This paper describes the Parallel Volume Visualization (PVV) toolkit. PVV is an Application Programming Interface (API) for developing volume visualization applications. It was designed to be easily portable across high performance parallel computers such as shared memory and distributed memory parallel computers. Good speed-up rates have been achieved with a preliminary implementation of PVV on a SGI 4D480 multiprocessor.