Background/Objectives: Three-dimensional (3D) image-guided robotic-assisted partial nephrectomy (3D-IGRAPN) integrates patient-specific anatomical models to optimize surgical planning and intraoperative guidance in the management of renal tumors. This study aimed to assess medium-term functional and oncologic outcomes of 3D-IGRAPN in a large, prospective cohort. Methods: All consecutive patients undergoing 3D-IGRAPN between January 2016 and March 2023 at a tertiary referral center were prospectively included in the UroCCR database (NCT03293563). Patient-specific 3D models were generated from preoperative CT scans and used intraoperatively. The primary endpoint was trifecta achievement, defined as an absence of major complications (Clavien–Dindo ≥ 3), negative surgical margins for malignant tumors, and ≥90% preservation of baseline renal function at 3 months. Secondary endpoints included functional outcomes, complication rates, local recurrence, and metastasis rates, as well as cancer-specific and overall survivals. Results: Among 568 patients (586 surgeries), the trifecta was achieved in 55.2% of evaluable malignant cases. Severe complications occurred in 33 cases (5.6%), and positive surgical margins were reported in 27 cases (5.1%) out of 528 surgeries involving malignant lesions. Renal function was preserved in 59.9% of patients at 3 months. At a mean follow-up of 31.5 months, recurrence and metastasis rates were 7.4% and 8.6%, respectively. Cancer-specific and overall survival at follow-up were 96.5% and 89%. Conclusions: 3D-IGRAPN demonstrates favorable functional and oncologic outcomes, even in complex tumors. These results support the integration of 3D modeling as a standard tool in image-guided nephron-sparing surgery.

Radical nephroureterectomy is the standard treatment for renal upper urinary tract urothelial cancer, but may expose patients to renal failure and hemodialysis. We aimed to evaluate the outcomes of nephron-sparring surgery in three selected cases of localized renal UUTUC confined to a single calyx treated by segmental 3D image-guided robot assisted partial nephrectomy. All tumors were non-invasive and confined to either the superior or inferior calyx, refractory to endoscopic treatment and without evidence of synchronous lesions, lymph nodes involvement, or metastasis. All procedures were performed using the Da Vinci Xi robot, with preoperative planning and intraoperative guidance based on a 3D-model reconstructed from CT-urography. Key surgical principles included selective devascularization, avoidance of main artery clamping, and no opening of the concerned excretory tract. Mean patient age was 58 years, estimated blood loss was 166 mL and mean operative time was 272 minutes. All patients were discharged on postoperative day 1, with no perioperative complications. On patient experienced recurrences treated successfully by ureteroscopy. At 23,3 months of median follow-up, all patients were disease-free. Although limited to three patients, this initial experience suggests the potential value of this technique. However, larger studies with long-term follow-up are warranted to confirm its safety and efficacy.

BACKGROUND Augmented reality (AR) has emerged as a promising tool to enhance surgical precision during robot-assisted partial nephrectomy (RAPN), particularly by enabling the intraoperative overlay of 3D anatomical models. However, real-time AR implementation requires robust image segmentation of anatomical structures, such as the kidney, which remains technically challenging in dynamic laparoscopic environments. OBJECTIVE This study aimed to develop and validate a large, annotated image dataset to train deep learning models for automated segmentation of the renal parenchyma during RAPN, as a prerequisite for real-time AR guidance. METHODS We conducted a single-center, observational image annotation study using prospectively collected surgical videos from 131 RAPN procedures performed between 2022 and 2024. Patients had localized renal tumors, with 11 presenting with multifocal disease (160 tumors in total). A total of 48,000 frames were extracted based on image sharpness, diversity, and the exclusion of artifacts. A subset of 454 images was annotated by 9 contributors (surgeons, engineers, and nonexperts) after structured training. Interannotator agreement was assessed using the Dice similarity coefficient (DSC) and sensitivity against an expert reference. A convolutional neural network (AlbuNet-34) was trained using 12,546 annotated images and evaluated on a validation set of 3137 images. Model performance was analyzed overall and across surgical phases. RESULTS Annotators achieved high agreement, with median DSC values ranging from 0.91 to 0.95 and sensitivity consistently more than 0.89. The deep learning model achieved a mean DSC of 0.75 (SD 0.23) and a sensitivity of 0.71 (SD 0.24) on the validation set. Segmentation accuracy varied significantly across surgical phases, with lower performance observed during tumor resection and tumor bed reconstruction due to increased visual complexity. CONCLUSIONS This study demonstrates the feasibility of automated renal parenchyma segmentation using deep learning in real-world intraoperative settings. Although current performance remains below that of expert-level annotations, the creation of a large, annotated dataset and the implementation of a structured multiannotator workflow represent key milestones toward reliable AR-assisted surgery. Ongoing refinements in annotation quality, dataset diversity, and neural network optimization are expected to enhance future real-time AR applications in urology.

Background: Augmented reality (AR) has emerged as a promising tool to enhance surgical precision during robot-assisted partial nephrectomy (RAPN), particularly by enabling the intraoperative overlay of 3D anatomical models. However, real-time AR implementation requires robust image segmentation of anatomical structures, such as the kidney, which remains technically challenging in dynamic laparoscopic environments. Objective: This study aimed to develop and validate a large, annotated image dataset to train deep learning models for automated segmentation of the renal parenchyma during RAPN, as a prerequisite for real-time AR guidance. Methods: We conducted a single-center, observational image annotation study using prospectively collected surgical videos from 131 RAPN procedures performed between 2022 and 2024. Patients had localized renal tumors, with 11 presenting with multifocal disease (160 tumors in total). A total of 48,000 frames were extracted based on image sharpness, diversity, and the exclusion of artifacts. A subset of 454 images was annotated by 9 contributors (surgeons, engineers, and nonexperts) after structured training. Interannotator agreement was assessed using the Dice similarity coefficient (DSC) and sensitivity against an expert reference. A convolutional neural network (AlbuNet-34) was trained using 12,546 annotated images and evaluated on a validation set of 3137 images. Model performance was analyzed overall and across surgical phases. Results: Annotators achieved high agreement, with median DSC values ranging from 0.91 to 0.95 and sensitivity consistently more than 0.89. The deep learning model achieved a mean DSC of 0.75 (SD 0.23) and a sensitivity of 0.71 (SD 0.24) on the validation set. Segmentation accuracy varied significantly across surgical phases, with lower performance observed during tumor resection and tumor bed reconstruction due to increased visual complexity. Conclusions: This study demonstrates the feasibility of automated renal parenchyma segmentation using deep learning in real-world intraoperative settings. Although current performance remains below that of expert-level annotations, the creation of a large, annotated dataset and the implementation of a structured multiannotator workflow represent key milestones toward reliable AR-assisted surgery. Ongoing refinements in annotation quality, dataset diversity, and neural network optimization are expected to enhance future real-time AR applications in urology.