Ureteroscopy is an efficient endoscopic minimally invasive techniquefor the diagnosis and treatment of upper tract urothelial carcinoma (UTUC) U-Net is based on an ensemble of 4 parallel CNNs to process single and multi-frame information . The method was evaluated using a custom dataset of 11videos (2,673 frames) which were collected and manually annotated from 6patients . We obtain a Dice similarity coefficient of 0.80, outperforming previous state-of-the-art methods . Conclusion: The obtained results show that spatial-temporal information can be effectively exploited by the ensemble model to improve hollow lumensegmentation in ureteroscopic images. The method is effective also in the presence of poor visibility, occasional bleeding, or specular reflections. The Method is effectiveAlso in the absence of . poor visibility or . occasional bleeding or speculant reflections, or . specular reflection, or a lack of visibility, or an occasional bleeding in the images. It was effective also effective also

Author(s) : Jorge F. Lazo, Aldo Marzullo, Sara Moccia, Michele Catellani, Benoit Rosa, Michel de Mathelin, Elena De Momi

Links : PDF - Abstract

Code :

Keywords : method - effective - visibility - occasional - bleeding -

Leave a Reply

Your email address will not be published. Required fields are marked *