Renal Cancer Classification in Tomographic Images Using Convolutional Neural Networks to Assist Early Diagnosis

• Vinícius Meireles Pereira Santos UFV http://orcid.org/0009-0006-1067-342X
• Ana Claudia Patrocinio UFU
• William Chaves de Souza Carvalho UFU
• Pedro Moises de Sousa UFV https://orcid.org/0000-0003-4563-0033

Abstract

Renal cancer is one of the 13 most frequent types of cancer in Brazil, primarily affecting individuals between 50 and 70 years of age, with around 12,000 cases and 4,000 deaths recorded in 2020. Early detection is crucial to prevent the progression of the disease, which can lead to kidney transplants, dialysis, or death. However, the lack of symptoms and screening tests make diagnosis difficult. This research proposes the use of Convolutional Neural Networks (CNNs) for the detection of renal cancer in CT images, classifying them as cancerous or normal. It is expected that the application of the obtained results can assist in early diagnosis, supporting healthcare professionals in providing faster and more accurate treatments and contributing to improving the patients' quality of life.

Keywords: Convolution Neural Networks., Kidney Cancer, Classification of Kidney Images