RADIÔMICA E APRENDIZADO DE MÁQUINA PARA IDENTIFICAÇÃO DA CODELEÇÃO CROMOSSÔMICA 1p/19q EM GLIOMAS DE BAIXO GRAU

Abstract

Gliomas represent approximately 40% of all brain turmors, and approximately 50% of all gliomas are low-grade, almost entirely represented by oligodendrogliomas and astrocytomas, both grade II or III. Since 1p/19q chromosome codeletion is an important tumor marker of positive prognostic (benign) factor in these gliomas, we propose the use of radiomics in image analysis as a comprehensive quantifier of non-invasive tumor phenotypes to identify 1p/19q chromosome codeletion in low-grade gliomas. This is an approach potentially used in oncology, helping in the detection, diagnosis and prognosis of cancer, prediction of response to treatment and monitoring of the state of the disease. Using the PyRadiomics platform, which extracts radiomic features that quantify the intensity, shape and texture of the tumor image in magnetic resonance images, a dataset with 159 patients diagnosed with low-grade gliomas was used and 120 features were extracted . Then, 4 experiments were implemented where data pre-processing techniques were applied, such as selection, scaling, resizing and data augumentation, and supervised machine learning. After evaluating a set of test data in each experiment, promising results were obtained for accuracy, sensitivity, specificity, precision, area under the curve (AUC) and F1 score. These results were competitive when compared to the state of the art. This study describes an exploratory research whose results underscore that radiomics in combination with machine learning constitutes a promising approach in identifying 1p/19q chromosome codeletion in low-grade gliomas. University of Brasilia - UnB, Postgraduate Program in Electronic Systems and Automation.