Desenvolvimento de um algoritmo avançado para mensuração de densidade óssea baseado na densitometria radiográfica

Abstract

Radiographic densitometry is a technique that allows us to evaluate bone density, using as reference an inert material, such as aluminum. Those techniques allow for the support of the diagnosis of diseases and bone abnormalities such as osteoporosis. To increase the technical precision, some authors proposed the use of mathematical equations, as the minimum square method, LaPlace's theorem and the Sarrus rule, which will result in a better fit of the characteristic absorption curve to X-rays through the study object. The use of technologies such as Java and MySql-open source-Allows the development of advanced algorithms, able to analyze and process digital or scanned radiographic images and calculate density values close to the study object values. Currently many paid solutions are supported to perform radiographic densitometry. The objective of this study is to develop an algorithm that makes it possible to measure bone density from the X-ray densitometry. To test the software, a number of test will be conducted using both inert material and bone materials. It is expected that with the implementation of advanced algorithms can provide better precision and accuracy, once applications using the polynomial 3rd degree demonstrated excellent accuracy, but a bad correctness in some steps cases.