查看文章

Kidney cancer has an high number of diagnosed cases each year, being invasive techniques the most common treatment [5]. Machine learning techniques have been proven to provide valuable tools in medical imaging, increasing the accuracy in diagnosis for several diseases [2]. In recent years, deep learning techniques have increased the performance of classical machine learning approaches in several imaging problems, including medical image analysis [4]. In this work, we propose a deep learning approach for kidney cancer detection and segmentation, as participants of the Kits19 challenge.