BRIEF RESEARCH REPORT published: 03 May 2022 doi: 10.3389/fsurg.2022.891596 3D Imaging Segmentation and 3D Rendering Process for a Precise Puncture Strategy During PCNL – a Pilot Study Otaš Durutović 1,2, Aleksandar Filipović 1,3*, Katarina Milić ević 4, Bhaskar Somani5, Esteban Emiliani6, Andreas Skolarikos7 and Milica M. Janković 4 1 Faculty of Medicine, University of Belgrade, Belgrade, Serbia, 2Clinic of Urology, University Clinical Centre of Serbia, Belgrade, Serbia, 3Center for Radiology and Magnetic Resonance Imaging, University Clinical Centre of Serbia, Belgrade, Serbia, 4 Laboratory for Biomedical Instrumentation and Technologies, Department of Signals and Systems, University of Belgrade, School of Electrical Engineering, Belgrade, Serbia, 5Faculty of Medicine, University Hospital Southampton, Southampton, United Kingdom, 6Department of Urology, Fundacion Puigvert, Autonomous University of Barcelona, Barcelona, Spain, 7 National and Kapodistrian University of Athens, 2nd Department of Urology, Sismanoglio Hospital, Athens, Greece Percutaneous nephrolithotomy (PCNL) is frequently used as the first-line treatment of large and Edited by: complex stones. The key point for successful complex stone removal with minimal risk of Clemens Mathias Rosenbaum, complications is to establish the most appropriate access route. Understanding the three- Asklepios Klinik Barmbek, Germany dimensional (3D) relationship of kidney stones and renal collecting systems is crucial for Reviewed by: planning and creating an optimal access route. By using a 3D volume segmentation tool a more Noor Buchholz, accurate approach to the renal collecting system and stone treatment could be planned. The U-merge Scientific Office, objective of this study was assessing the impact of 3D software in getting the desired access. United Kingdom Andreas Bruchbacher, Keywords: 3D reconstruction, 3D rendering, PCNL, puncture guidance, kidney stone Asklepios Klinik Barmbek, Germany *Correspondence: Aleksandar Filipović aleksandar.filipovic11@gmail.com INTRODUCTION Speciality section: This article was submitted to During renal stone treatment planning, non-contrast computerized tomography (NCCT) is a Genitourinary Surgery, a section of the highly recommended examination, used for assessing the stone size, number, and location. journal Frontiers in Surgery Non-excretory phases do not provide precise insight into detailed stone distribution inside the Received: 07 March 2022 renal collecting system. With the use of an additional excretory phase scan, the urologist can Accepted: 11 April 2022 study in-depth the relationship between the stone and the occupied pelvicalyceal system. Published: 03 May 2022 European Association of Urology (EAU) Urolithiasis Guidelines recommend percutaneous Citation: nephrolithotomy (PCNL) as the first-line treatment of large stones, more than 2 cm (1). Durutović O, Filipović A, Milić ević K, Urologists usually define a renal stone as being complex based on the complicated and difficult Somani B, Emiliani E, Skolarikos A and pelvicalyceal anatomy, branching of the stone into the pelvicalyceal system, or the existence of Janković MM (2022) 3D Imaging multiple stones in various calyces. PCNL treatment of complex renal stones frequently require Segmentation and 3D Rendering Process for a Precise Puncture several access tracts and clear out the stone completely (2–4). However, the overall complication Strategy During PCNL – a Pilot Study. rate, as well as the transfusion rate, is higher when using multiple tracts (5–7). The key point Front. Surg. 9:891596. for successful complex stone removal with minimal risk of complications is to establish the doi: 10.3389/fsurg.2022.891596 most appropriate access route (8). That means that the urologist should select and puncture Frontiers in Surgery | www.frontiersin.org 1 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning that calyx(s) which will lead to complete stone extraction, while Software Environment they will not increase perioperative complication rate. 3D Gastro CT Ex tool is an open-source tool developed on Understanding the three-dimensional (3D) relationship of Python 3. This tool is the extended version of the 3D Gastro kidney stones and renal collecting systems is crucial for CT tool that was previously introduced (12). The previous planning and creating an optimal access route. In recent years software version supports segmentation and 3D rendering of the use of advanced imaging and 3D reconstruction of CT abdominal CT scans for individual phases (native and vein data in the treatment of kidney stones is increasing (8). phase). The new software version, whose advantages in the Advanced imaging systems have been shown to provide the planning of PCNL treatment are illustrated through this 3D perception of target organs, surrounding anatomic paper, offers the option for 3D hybrid visualization of native structures, depth perception, and spatial orientation during and delayed phase, which will be described in detail. The endoscopic and minimally invasive surgery (9, 10). following libraries and toolkits were used in the tool There is a continuous search for new and improved methods that development process: matplotlib (13), ndimage (14), could help and refine the course of the PCNL procedure by SimpleITK (15–17), and VTK (Visualization Toolkit) (18). achieving a more accurate approach to the renal collecting system PyQt5 binding for Qt v5 was used for designing an intuitive and stone treatment while decreasing the risk of complications (11). and user-friendly graphical interface (19). The tool offers For the purpose of this study an elective semi-automated options for reading and viewing different image formats that instrument was created for a more exact and quantitative are supported by SimpleITK reader (e.g., Dicom, MetaImage, evaluation of the renal collecting system and the stone by etc.). Export options for saving rendering results in.stl and .jpg using a 3D volume segmentation tool. This could help for formats are available. The source code of 3D Gastro CT Ex is preoperative comprehensive PCNL access route planning but available for download from the Github repository: https:// is also viable for real-time use in the operating room. The aim github.com/milicevickatarina/3D-Gastro-CT-Extended. of this study was to assess the impact of 3D software in getting the desired access. Image Preprocessing Image preprocessing of axial CT slices (512 × 512) included band shifting on a histogram, image noise re-movement, and METHODS co-registration process. First, the band [−548, 800] was shifted to [0,255] using a linear intensity transfer function. Second, In this pilot case study, we enrolled 5 patients, which were selected image noise removal was performed by median filtering. with kidney stones, planned for PCNL procedure. CT scans were Finally, Mattes mutual information was used as a criterion for performed using a 64 multi-detector row CT scanner (General the co-registration of native and delayed phase using the Electric) with 0.625 mm axial plane reconstruction. The scan was SimpleITK class Image Registration Method (sitkLinear was performed before and after contrast administration and both used as interpolator and gradient descent was used for the phases (plain and excretory) were used for the segmentation optimization process) (20). process and fusion of the 3D images into a single semi- transparent model demonstrating the stone distribution in the collecting system. While NCCT was used for stone segmentation, Segmentation and 3D Rendering the excretory phase was used for collecting system segmentation. The overall segmentation process on native and delayed phases The scan delay time for excretory phase after i.v. contrast is presented in Figure 1. First, the segmentation procedure administration was 10–15 min, depending on the kidney implies the calculation of histograms for the whole CT volume excretion and degree of obstruction if present. The data set was of both phases. Skeleton segmentation includes the following routinely analyzed in Advantage Workstation Software V 4.6. steps: The contrast medium used was Ultravist 370 mg J/mL, 1. Volume binarization from the native phase based on the approximately 70–100 mL depending on the patient’s weight. histogram thresholding method (default threshold values: After NCCT acquisition DICOM images were used for lower = 117 and upper = 180, light green line and light red segmentation and 3D reconstruction of the renal stones using line on native phase histogram in Figure 1 respectively) the 3D Gastro CT Ex tool. The same software was used for 2. Particle removal from the volume extracted in step 1 based segmentation and 3D reconstruction of pelvicalyceal system on opening and closing of the volume (cube structural images obtained at the excretory phase scan. The reconstructed element with the dimension 3) 3D model represented the interrelation between the 3. Dilatation of the volume extracted in step 2 pelvicalyceal system and stone with the greater anatomical 4. Volume binarization from the delayed phase based on the details. Visualization is achieved by adjusting the scanned histogram thresholding method (default threshold values: object transparency. Segmentations were performed to obtain lower = 130 and upper = 255, light green line and light red stone volume (SV) and pelvicalyceal system volume (PSV). line on delayed phase histogram in Figure 1 respectively) After volume rendering of the collecting system and renal 5. Particle removal from the volume extracted in step 3 based stone, the stone volume (SV) was analyzed and compared with on opening and closing of the volume (cube structural the pelvicalyceal system (PS) volume. The volumetric ratio of element with the dimension 3) the stone and collecting system (SV/PS) was also analyzed. 6. The intersection of volumes obtained in step 3 and step 5. Frontiers in Surgery | www.frontiersin.org 2 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning FIGURE 1 | The flow chart of the segmentation and 3D rendering process. A renal collecting system is extracted by the subtraction of after retrograde contrast injection through the ureteral volumes archived in steps 6 and 5. catheter. Tract was dilated up to 30 Fr Amplatz sheath Calculus segmentation includes the following steps: (standard PCNL) with the use of 26 Fr nephroscope. 1. Volume binarization from the native phase based on the histogram thresholding method (default threshold values: lower = 250 and upper = 255, dark green line and dark red RESULTS line in Figure 1 respectively) All patients were completely investigated by conventional 2. Particle removal from the volume extracted in the previous protocols proposed by guidelines. After confirming the step is based on the opening and closing of the volume indication for surgery, an additional puncture strategy was (cube structural element with the dimension 2). made after 3D reconstruction using the 3D Gastro CT Ex 3D rendering was performed by Marching Cubes algorithm tool. Images from conventional CT scanners were compared (21, 22). VTK renderer was used for the implementation of with new images and an optimal axis for percutaneous rotation and zoom options (Figure 1). The processing time puncture was made upon case characteristics. Due to the 3D of DICOM files required to obtain a semi-automatic 3D reconstructed semi-transparent images, it was possible to image is from 45 to 90 min on a computer with the analyze volumetric datasets in both anterior and posterior following features: processor Intel (R) Core (TM) i7-8565U view (Figures 2 and 3). Patients from the study group had CPU@1.80 GHz, 8GB RAM, graphics card NVIDIA GeForce complex renal stones (Guy stone score 3 and 4) (23). All cases MX250 were treated by using a single puncture (Figure 4). The use of At the operating room, the patient was positioned on the flexible nephroscope was indicated by case characteristics and table in a prone position with all the coordinates and surgery flow. Stone-free status was achieved in all cases, measurements taken out from both conventional CT confirmed by follow up imaging. No perioperative evaluation and 3D model translated on the skin of the lumbar complication occurred. In our series there was no unsuccessful area to enable the planned access route. Puncture was outcome of PCNL procedures performed regardless the use of performed under ultrasound and pulsed fluoroscopy control the 3D software for puncture planning. Frontiers in Surgery | www.frontiersin.org 3 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning FIGURE 3 | Posterior view. FIGURE 2 | Anterior view. PCNL is routinely performed under 2D visualization with ultrasound and/or fluoroscopy guidance. In our technique the use of flexible nephroscope is not mandatory but flexible DISCUSSION nephroscope should always be available during PCNL in aim to clear all stones from the pelvicalyceal system but also upper ureter. The pelvicalyceal system with a complex stone represents a By using a 3D model of CT images in planning the route and complex 3D structure. Thus, for accurate access route controlling the needle path, we would be able to use all the planning, the semi-transparent 3D model representing the advantages of multidetector CT (MDCT) without raising the correlation between the collecting system and stone seems to risk of radiation dose as it would be the case with CT be very useful. fluoroscopy or Cone-beam CT guidance. 89% of success rate Rendering of the CT images with the ability to adjust the for percutaneous kidney access has been reported when using transparency of different tissue layers allowed the operator to Cone-beam CT for guidance, but it comes with a higher see the stone through the renal collecting system, locate the radiation dose than conventional fluoroscopy guidance (26). stone and navigate the puncture needle. This semi-transparent An accurate preoperatively defined puncture trajectory may also 3D model could provide additional information during contribute to decreasing total fluoroscopy time to the lowest possible endourological procedures and therefore raise the possibility level. Once the operator has a clear preoperative puncture plan of achieving one-stage clearance of complex renal stones. The during fluoroscopy and ultrasound-guided PCNL procedures, greatest advantage of using a 3D software is in creating a clear fluoroscopy can be used only for access tract formation and wire preoperative plan and selecting a target access route. Once position confirmation. Shortening of fluoroscopy time during having a clear preoperative puncture plan further steps of PCNL is an important technical requirement to ensure safety and puncture guidance and control can be performed very efficiency in PCNL while also reducing the radiation exposure of precisely, usually with less fluoroscopy use. In this way the the patient and surgical team (11). total fluoroscopy time can be reduced. As the total Our puncture technique involves ultrasound guidance with fluoroscopy time is a surrogate for radiation exposure, we the assistance of pulsed fluoroscopy. Compared with our initial assume that preoperative planning is a good way to also cases which were performed by ultrasound and continuous achieve the goal of reducing the use of fluoroscopy to a fluoroscopy we have already achieved significant reduction in minimum in accordance with ALARA principles. fluoroscopy time from 155.4 s (median value) to 76.8 s (27). It has been proven that the surgeon is more comfortable with If we compare the total fluoroscopy time values of PCNL the initial puncture due to the 3D preoperative planning (24, 25). procedures performed without the use of 3D software already Frontiers in Surgery | www.frontiersin.org 4 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning FIGURE 4 | Images from surgery. reported in our larger series (27) with the total fluoroscopy time majority of patients have normal renal function the use of values of the PCNL procedures performed in this study (after contrast offers a huge benefit in term of detail determination creating the semi-transparent 3D model of pyelocaliceal system and distinction between stone and pelvicalyceal system. and stone) a significant reduction in the total fluoroscopy time Measurements of puncture angles and distances from the is noted when 3D software was used (76.8 s vs 27.7 s). referent points were collected from a 3D model and translated Bearing in mind that in this study the results obtained on a in the operating room with accuracy. Vascular structures’ small sample of five patients are presented, we cannot speak segmentation was not taken into consideration because the from the relevance of the statistically obtained data. This initial puncture of the renal collecting system was performed reduction can also be a result of an increased experience and by an experienced urologist with real-time ultrasound skills of surgeon (OD), as his volume increased during the years. guidance. The vessel segmentation process could take To this study, CT images were segmented by focusing on renal additional time before the procedure starting point without stone, collecting system, and bone structures serving as a any substantial decrease in the rate of vascular injuries. landmark for translation of measurements on the patient’s skin. From the point of PCNL safety, avoidance of unnecessary Contrast is not absolutely mandatory for MDCT examination, puncture(s) is crucial. A greater rate of complications has especially in cases with damaged renal function. As the been reported during calyceal stones treatment in comparison Frontiers in Surgery | www.frontiersin.org 5 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning with the treatment of pelvic stones (28). When planning patient. This was especially the case with smaller stones in the treatment with rigid instruments, the access route should be dilated collecting system. After 3D model calculation and taking as straight as possible from the skin through the calyx in the measurements in a few cases the stone had changed its position axial line of the calyx through the infundibulum of the certain most likely due to patient positioning. The CT scan was calyx and into the renal pelvis. Any additional intraoperative routinely performed while the patient was in the supine position angulation of the access sheath would increase the risk of and the PCNL procedure was performed in a prone position. renal injury and bleeding. Every percutaneous kidney puncture is associated with a risk of bleeding. With the need for tract formation and dilatation, this risk is increased. The CONCLUSION transfusion rate is higher in cases where multiple tracts were used during PCNL. Therefore, a decrease in access tracts is Having a clear and precise preoperative puncture plan is the key one of the goals of endourologists. The use of flexible point to ensure efficiency in PCNL. Optimal access according to nephroscope can decrease the need for additional puncture. In stone distribution and renal collecting system anatomy may our opinion, even with flexible nephroscope the importance of contribute to both PCNL procedure efficacy and safety. This optimal access axis remains of huge importance. Optimal semi-automated segmentation tool for 3D visualization of the access offers two biggest benefits: avoidance of torqueing and collecting system and stone interrelation is proven to be very in that way less possibility of bleeding and better visibility useful for preoperative PCNL access planning. Currently, this during the procedure, especially in complex cases with stone might be a time-consuming process, but in the future, it distribution in different calices. Another benefit of detailed should be a completely automated process with widespread preoperative planning and selection of an optimal access route use and adoption for all complex stone procedures. is in possibility to easily approach to all other calices that contain stones (such as in parallel calices, that sometimes can not easily be approached, even with flexible nephroscope from DATA AVAILABILITY STATEMENT the punctured calyx) or might be a possible place of stone migration (usually upper calyx). The datasets presented in this study can be found in online Stone burden and volume distribution in the renal collecting repositories. The names of the repository/repositories and system is proven to be a significant determinant of the stone-free accession number(s) can be found in the article. rate (SFR) (24), which was defined as no residual fragments larger than 3 mm in diameter (29, 30). In our initial experience and case series, the use of this tool helped significantly in AUTHOR CONTRIBUTIONS preoperative planning and surgical strategy, allowing access to secondary calices from the main axis tract more easily. OD is responsible for study design, data collection, data There were no additional costs for this kind of evaluation, the interpretation, manuscript preparation, funds collection; AF software is free for use and its utilize was simple and intuitive. for data collection, data interpretation, manuscript Collaboration with radiologists and software engineers was preparation, literature search; KM for data interpretation, necessary, especially as the program offers the opportunity of manuscript preparation, literature search; BS for manuscript rotation of the image and creating a picture in all positions preparation, data interpretation, literature search; EE for used for PCNL, whether prone or supine or Valdivia modified study design; AS for study design, manuscript preparation, used for retrograde intrarenal surgery (RIRS). literature search; MJ for study design, data interpretation, The limitation of the study was a shift and change in the manuscript preparation, literature search. All authors position of the stone that can happen with the positioning of the contributed to the article and approved the submitted version. REFERENCES 5. Kukreja R, Desai M, Patel S, Bapat S, Desai M. First prize: factors affecting blood loss during percutaneous nephrolithotomy: prospective 1. Türk C, Petřík A, Sarica K, Seitz C, Skolarikos A, Straub M, et al. EAU Study. J Endourol. (2004) 18(8):715–22. doi: 10.1089/end.2004.18.715 guidelines on interventional treatment for urolithiasis. Eur Urol. (2016) 6. Turna B, Umul M, Demiryoguran S, Altay B, Nazli O. How do increasing 69(3):475–82. doi: 10.1016/j.eururo.2015.07.041 stone surface area and stone configuration affect overall outcome of 2. Brehmer M, Beckman MO, Magnusson A. Three-dimensional computed percutaneous nephrolithotomy? J Endourol. (2007) 21(1):34–43. doi: 10. tomography planning improves percutaneous stone surgery. Scand J Urol. 1089/end.2005.0315 (2014) 48(3):316–23. doi: 10.3109/21681805.2013.876552 7. Turna B, Nazli O, Demiryoguran S, Mammadov R, Cal C. Percutaneous 3. Fernström I, Johansson B. Percutaneous pyelolithotomy: a new extraction nephrolithotomy: variables that influence hemorrhage. Urology. (2007) 69(4): technique. Scand J Urol Nephrol. (1976) 10(3):257–9. doi: 10.1080/ 603–7. doi: 10.1016/j.urology.2006.12.021 21681805.1976.11882084 8. Christiansen AR, Shorti RM, Smith CD, Prows WC, Bishoff JT. 4. Raza A, Moussa S, Smith G, Tolley DA. Upper-pole puncture in percutaneous Intraoperative utilization of advanced imaging modalities in a complex nephrolithotomy: a retrospective review of treatment safety and efficacy. BJU kidney stone case: a pilot case study. World J Urol. (2018) 36(5):733–43. Int. (2008) 101(5):599–602. doi: 10.1111/j.1464-410X.2007.07388.x doi: 10.1007/s00345-018-2260-4 Frontiers in Surgery | www.frontiersin.org 6 2022 | Volume 9 | Article 891596 Durutović et al. 3D-Rendering for PCNL Puncture Planning 9. Zhang X, Chen G, Liao H. High-quality see-through surgical guidance system 24. Atalay HA, Canat L, Bayraktarlı R, Alkan I, Can O, Altunrende F. Evaluation of using enhanced 3-D autostereoscopic augmented reality. IEEE Trans Biomed stone volume distribution in renal collecting system as a predictor of stone-free Eng. (2016) 64(8):1815–25. doi: 10.1109/TBME.2016.2624632 rate after percutaneous nephrolithotomy: a retrospective single-center study. 10. Wang R, Geng Z, Zhang Z, Pei R, Meng X. Autostereoscopic augmented Urolithiasis. (2018) 46(3):303–9. doi: 10.1007/s00240-017-0995-9 reality visualization for depth perception in endoscopic surgery. Displays. 25. Keyu G, Shuaishuai L, Raj A, Shuofeng L, Shuai L, Yuan Z, et al. A 3D (2017) 48:50–60. doi: 10.1016/j.displa.2017.03.003 printing personalized percutaneous puncture guide access plate for 11. Zeng G, Zhong W, Pearle M, Choong S, Chew B, Skolarikos A, et al. percutaneous nephrolithotomy: a pilot study. BMC Urol. (2021) 21(1):184. European association of urology section of urolithiasis and international doi: 10.1186/s12894-021-00945-x alliance of urolithiasis joint consensus on percutaneous nephrolithotomy. 26. Ritter M, Rassweiler M-C, Michel MS. The Uro Dyna-CT enables three- Eur Urol Focus. (2021) S2405–4569(21):00065–00071. doi: 10.1016/j.euf. dimensional planned laser-guided complex punctures. Eur Urol. (2015) 2021.03.008 68(5):880–4. doi: 10.1016/j.eururo.2015.07.005 12. Milićević K, Durutović O, Janković M. Open-source tool for 3D segmentation 27. Durutovic O, Dzamic Z, Milojevic B, Nikic P, Mimic A, Bumbasirevic U, and rendering of abdominal CT scans, 8th International Conference on et al. Pulsed versus continuous mode fluoroscopy during PCNL: safety and Electrical, Electronic and Computing Engineering IcETRAN 2021, pp. 151–154, effectiveness comparison in a case series study. Urolithiasis. (2016) 44(6): ETRAN Society, Belgrade, Academic Mind, Belgrade, Ethno Village Stanišići, 565–70. doi: 10.1007/s00240-016-0885-6 Republic of Srpska. (2021). 28. Xue W, Pacik D, Boellaard W, Breda A, Botoca M, Rassweiler J, et al. 13. Hunter JD. Matplotlib: a 2D graphics environment. Comput Sci Eng. (2007) Management of single large nonstaghorn renal stones in the CROES PCNL 9(3):90–5. doi: 10.1109/MCSE.2007.55 global study. J Urol. (2012) 187(4):1293–7. doi: 10.1016/j.juro.2011.11.113 14. Available from: https://docs.scipy.org/doc/scipy/reference/ndimage.html# 29. Tanriverdi O, Boylu U, Kendirci M, Kadihasanoglu M, Horasanli K, Miroglu (Accessed February 2022). C. The learning curve in the training of percutaneous nephrolithotomy. Eur 15. Beare R, Lowekamp BC, Yaniv Z. Image segmentation, registration and Urol. (2007) 52(1):206–12. doi: 10.1016/j.eururo.2007.01.001 characterization in R with SimpleITK”. J Stat Softw. (2018) 86(8):6595. 30. Jang WS, Choi KH, Yang SC, Han WK. The learning curve for flank doi: 10.18637/jss.v086.i08 percutaneous nephrolithotomy for kidney calculi: a single surgeon’s 16. Yaniv Z, Lowekamp BC, Johnson HJ, Beare R. SimpleITK image-analysis experience. Korean J Urol. (2011) 52(4):284–8. doi: 10.4111/kju.2011.52.4.284 notebooks: a collaborative environment for education and reproducible research. J Digit Imaging. (2018) 31(3):290–303. doi: 10.1007/s10278-017- Conflict of Interest: The authors declare that the research was conducted in the 0037-8 absence of any commercial or financial relationships that could be construed as 17. Lowekamp BC, Chen DT, Ibáñez L, Blezek D. The design of SimpleITK”. a potential conflict of interest. Front Neuroinform. (2013) 7(45):1–14. doi: 10.3389/fninf.2013.00045 18. Schroeder W, Martin K, Lorensen B. The Visualization Toolkit. 4th edn Publisher’s Note: All claims expressed in this article are solely those of the authors New York, USA: Kitware (2006). and do not necessarily represent those of their affiliated organizations, or those of 19. Available from: https://pypi.org/project/PyQt5/ (Accessed February 2022). the publisher, the editors and the reviewers. Any product that may be evaluated in 20. Available from: https://simpleitk.readthedocs.io/en/master/registrationOverview. this article, or claim that may be made by its manufacturer, is not guaranteed or html (Accessed February 2022). endorsed by the publisher. 21. Lorensen WE, Cline HE. Marching cubes: a high resolution 3D surface construction algorithm. ACM Siggraph Comput Graph. (1987) 21(4):163–9. Copyright © 2022 Durutović, Filipović, Milićević, Somani, Emiliani, Skolarikos and doi: 10.1145/37402.37422 Janković. This is an open-access article distributed under the terms of the Creative 22. Available from: https://github.com/lorensen/VTKExamples (Accessed February Commons Attribution License (CC BY). The use, distribution or reproduction in 2022) other forums is permitted, provided the original author(s) and the copyright 23. Thomas K, Smith NC, Hegarty N, Glass JM. The Guy’s stone score—grading owner(s) are credited and that the original publication in this journal is cited, in the complexity of percutaneous nephrolithotomy procedures. Urology. (2011) accordance with accepted academic practice. No use, distribution or reproduction 78(2):277–81. doi: 10.1016/j.urology.2010.12.026 is permitted which does not comply with these terms. Frontiers in Surgery | www.frontiersin.org 7 2022 | Volume 9 | Article 891596