Sistema de análise de imagens pulmonares implementado em ambiente web

Abstract

Allied with medicine, computer vision assists in the respiratory diseases diagnosis speeding up prognosis definition through digital image processing means. Computed tomography comes from this association. Invented by Allan Cormack and Godfrey Hounsfield, CT uses X-rays to obtain detailed images of cross-sections of the body that are processed by computers. Due to the insertion of computer systems in clinical applications, DICOM, the international standard for medical images and related information, emerged as a proposed solution to the demand to universalize the file format generated by diagnostic imaging devices such as CT scanners. Computational tools are developed and act as an intermediate interface between tomographs and computer systems. In this context, the first LISA version - Lung Image System Analysis - was developed as a desktop software, in 2010, aiming torax computed tomographies segmentation easing this way pulmonar diseases identification by radiologists or pulmonologists analysis. Like any traditional software, there is a need to install dependencies for it to run. In this work, it is introduced a web solution for LISA, using current technologies, as a new multiplatform version capable of running via an internet connection and without the need for installation.