Quantitative Evaluation of Encrustations in Double-J Ureteral Stents With Micro-Computed Tomography and Semantic Segmentation

Accurate evaluation of stent encrustation patterns, such as volume distribution, from different patient groups are valuable for clinical management and the development better stents. This study quantitatively compares patterns stone kidney transplant patients. Twenty-seven double-J ureteral stents were collected patients with disease or who underwent transplantation. Encrustations on samples quantified by means micro−Computed Tomography semantic segmentation using a Convolutional Neural Network model. Luminal per unit was derived to represent level, which did not differ between in first six weeks. However, showed higher levels over prolonged indwelling times (p = 0.02). Along shaft body, group near ureteropelvic junction compared ureterovesical 0.013), whereas no difference. Possible explanations discussed regarding vesicoureteral reflux. In both groups, pigtails more susceptible encrustations, difference renal bladder pigtail identified. The method presented this is also applicable other image analysis tasks urology.