Repozytorium publikacji - Politechnika Gdańska

Renal tumor malignancy classification is one of the crucial tasks in urology, being a primary factor included in the decision of whether to perform kidney removal surgery (nephrectomy) or not. Currently, tumor malignancy prediction is determined by the radiological diagnosis based on computed tomography (CT) images. However, it is estimated that up to 16% of nephrectomies could have been avoided because the tumor that had been diagnosed as malignant, was found to be benign in the postoperative histopathological examination. The excess of false-positive diagnoses results in unnecessarily performed nephrectomies that carry the risk of periprocedural complications. In this paper, we present a machine-aided diagnosis system that predicts the tumor malignancy based on a CT image. The prediction is performed after radiological diagnosis and is used to capture false-positive diagnoses. Our solution is able to achieve a 0.84 F1-score in this task. We also propose a novel approach to knowledge transfer in the medical domain in terms of colorization based pre-processing that is able to increase the F1-score by up to 1.8pp.

Autorzy

• Aleksander Obuchowski link otwiera się w nowej karcie ,
• Barbara Klaudel link otwiera się w nowej karcie ,
• Roman Karski link otwiera się w nowej karcie ,
• Bartosz Rydziński link otwiera się w nowej karcie ,
• Mateusz Glembin,
• dr inż. Paweł Syty link otwiera się w nowej karcie ,
• dr inż. Patryk Jasik link otwiera się w nowej karcie