Diferencias entre las configuraciones técnicas de Tomografía por Emisión de Positrones (PET) y Mamografía por Emisión de Positrones (PEM) para el diagnóstico del cáncer de mama

Abstract

The aim of this qualification work was to differentiate the technical configurations of positron emission tomography (PET) and positron emission mammography (PEM). The best technology for diagnosis was differentiated and a programming language was developed in Matlab; this algorithm analyses the textures of DICOM medical images, specific to breast, which were then selected and coregistered. An anatomical segmentation of discernible and non-discernible tumour lesions was performed, these segmented images were introduced into the developed code where the textures were analysed based on different methods, to implement a better estimation of data dimensionality. The texture analysis classification code was validated and the efficacy and safety assessment was determined. The diagnostic efficacy and safety assessment between (PET) and (PEM) in the detection of breast cancer and by means of the GUIDE graphical user interface, improved the discernment of the implemented code through the execution of the programme. It is concluded that (PEM) is the most useful technology for breast cancer diagnosis, due to the fact that its accuracy, precision and sensitivity are in the range of (92% to 94%), while for (PET) technology, it is in the range of (83% to 86%); the error rate is (15%) for (PEM) and for (PET) is (35%), it is also identified that the ROC Curve of the Area Under the Curve (AUC) index is (35%), is with a difference of (12%), which is considered to be of high diagnostic value that testifies to a notably high index of diagnostic efficacy and safety,, thus predicting more accurately the response to treatment.