| Sumario: | Simple Summary: Three-dimensional (3D) reconstruction of patient-specific renal anatomy is increasingly used to support planning in kidney surgery, but the models are usually produced by external providers. Synapse3D™ (V6.8.0017EU Build 0052, Fujifilm) is a clinician-oriented software package that allows a surgical team to generate 3D kidney models in-house from routine CT scans. In this study, three urology residents with no prior experience completed a two-day training programme and each independently reconstructed the same 20 consecutive cases scheduled for robotic kidney surgery. Every reconstruction was completed successfully. Reconstruction time fell by approximately 0.6 min for each additional case, and after approximately ten cases the mean time per case had fallen below 18 min; the rate of improvement was similar for all three residents. After surgery, the two operating surgeons reviewed each model and rated how closely it matched the anatomy they had encountered in the operative field. This study shows that in-house 3D reconstruction is technically feasible and can be learned quickly. Background/Objectives: To evaluate the feasibility and the learning curve of an insourced, clinician-performed three-dimensional (3D) image reconstruction workflow using Synapse3D™ software in a robotic urologic surgery setting. Methods: In this prospective single-centre case series, 20 consecutive patients scheduled for robotic kidney surgery were enrolled over a 30-day period. Three urology residents without prior experience in 3D reconstruction completed a structured two-day training programme including eight supervised practice reconstructions. Subsequently, each one reconstructed all 20 study cases independently. Reconstruction time was recorded for every reconstruction. After surgery, the two operating surgeons independently rated each model on a 5-point Likert scale for anatomical fidelity, defined as the correspondence between the model and the anatomy encountered intraoperatively. Results: All 60 reconstructions were completed successfully, with no failed reconstructions, no DICOM (Digital Imaging and Communications in Medicine) import errors and no software crashes; assistance from the manufacturer was never required. Mean reconstruction time was 19 min (range 11–29 min). Reconstruction time decreased by 0.63 min per case (95% CI 0.51–0.74; p < 0.001). Median time fell from 23 min in the first ten cases to 15 min in the last ten (p < 0.001), and the reduction was significant for each operator analysed separately. Anatomical fidelity was rated 4 (IQR 1) by both surgeons. The two surgeons rated overall system usability 3/5 and 5/5. Conclusions: Insourced 3D reconstruction with Synapse3D™ is technically feasible and is learned rapidly and reproducibly by urology residents after limited training.
|
|---|