Skip to main content
NCBI home page
Canadian Urological Association Journal logoLink to Canadian Urological Association Journal
. 2026 Jun;20(6Suppl1):S58–S65. doi: 10.5489/cuaj.9817

Poster Session 6: Training/Education

Sunday, June 28, 2026 • 16:45–18:00

PMCID: PMC13372449
Can Urol Assoc J. 2026 Jun;20(6Suppl1):S58.

MP 6.1. Structured reference letters in urology residency selection: Perspectives from Canadian trainees and program directors

Sanchit Kaushal 1,✉, David-Dan Nguyen 2,3, William Wu 4, Adree Khondker 2, David Bouhadana 5, Jennifer Bjazevic 6, Mélanie Aubé-Peterkin 5, Andrea Lantz Powers 7, Julien Letendre 8, Jasmir Nayak 9, Thomas de los Reyes 10, Jason Y Lee 2,4

Introduction

Letters of reference are a cornerstone of the Canadian urology residency selection process. Structured reference letters (SRLs) have emerged as a potential solution to improve standardization and fairness; however, their acceptability and structure for Canadian urology residency selection remains unknown.

Methods

We conducted a survey of a purposively sampled panel comprising of program directors (PDs) and residents from Canadian urology residency programs. The survey included 10 items (five quantitative and five qualitative) addressing three domains: opinions on SRLs, recommended content, and preferred letter structure. Responses collected on seven-point Likert scales were summarized using medians and interquartile ranges, while qualitative data were analyzed using content analysis and frequency counts of codes.

Results

Nine responses (six PDs, three residents) were received from seven unique residency programs. There was strong support for incorporating SRLs into residency applicants (median 6 [IQR 6–7]) and moderate agreement that SRLs are superior to narrative letters (median 5 [IQR 4–6]). Perceived benefits included increased standardization of evaluations (88.9%), greater efficiency for letter writers and file reviewers (44.4%), enhanced transparency (33.3%), and improved accuracy in distinguishing applicant attributes (22.2%). Concerns regarding exclusive reliance on SRLs included diminished granularity in differentiating candidates (44.4%), inability to capture unique applicant experiences (33.3%), and reduced ability to convey personalized insights (22.2%). Recommended global assessment domains included teamwork and collaboration (77.8%), communication (77.8%), initiative and work ethic (55.6%), professionalism and reliability (44.4%), and receptiveness to feedback (22.2%). Urology-specific domains emphasized included procedural skills appropriate to level of training (55.6%), foundational urologic knowledge (44.4%), demonstrated commitment to urology (44.4%), and overall residency potential or program fit (22.2%). All respondents supported the inclusion of a narrative comment section.

Conclusions

Among a cohort of Canadian urology PDs and resident trainees, there was strong support for the incorporation of SRLs into residency selection, particularly when combined with a narrative component. Findings from this study can inform a Delphi-based process for SRL development, including pilot testing to assess feasibility, usability, and potential impact on applicant evaluation.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S58.

MP 6.2. Open-access publishing: What has been the real impact on urology to date

Tomas Cordoba 1,✉, Othmane Zekraoui 2, Abbas Guennoun 1, Dean Elterman 3, Bilal Chughtai 4, Christopher JD Wallis 5,6, David-Dan Nguyen 6,7, D Robert Siemens 8, Michael Leveridge 8, Naeem Bhojani 1

Introduction

The publishing industry relies on unpaid research content and peer review, while access to this work is often restricted by paywalls. Open-access (OA) publishing reduces these barriers, and government mandates may contribute to broader dissemination and impact. Urologists may question the value of publishing OA in trusted urology journals. This study aimed to quantify the impact of OA publishing in urology across academic, societal, and clinical outcomes. Articles were classified as non-OA, gold, green, hybrid, or bronze.

Methods

Seventy-six urology journals, comprising fully open-access or subscription and mixed journals, were identified using the Clarivate Journal Citation Reports. Articles published from 2019–2023 were extracted from both databases to allow for citation accumulation. Citation impact was assessed using Web of Science. Societal impact was measured using Altmetric Attention Score and X mentions, and clinical impact using guideline citations. Citations between OA and non-OA articles were analyzed using linear regressions, and ANOVA with Tukey post-hoc testing was used to compare citations between OA subtypes. Altmetrics and guideline citations were analyzed using non-parametric tests and quasi-Poisson regression.

Results

A total of 39 607 articles were identified in Web of Science and 32 007 in Dimensions. OA articles showed a 104% citation advantage compared with closed articles (p<0.05). Citation impact differed across OA models, with hybrid outperforming gold by 70% (p<0.05). Altmetric Attention Score did not differ overall by access status but varied by subtype, with hybrid articles showing the highest attention. X mentions were twofold higher for OA articles (p<0.05). Clinical impact was similar across OA and closed models.

Conclusions

OA publishing in urology is associated with higher citation impact and public attention, particularly for hybrid articles. Publishing model choice should reflect study goals, target audience, and clinical priorities.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S59.

MP 6.3. Enhancing the Canadian Undergraduate Urology Curriculum through large language model-generated multiple-choice questions and flashcards

Othmane Zekraoui 1,✉, Tiange Li 2,3, David-Dan Nguyen 2,4, Marie-Lyssa Lafontaine 5, Tomas Cordoba 5, Mahmoud Moustafa 5, Dawn L MacLellan 6, Julien Letendre 1,7, Malek Meskawi 5, Garson Chan 8, Sarah Neu 2, Dean Elterman 2, Bilal Chughtai 9, Brendan Wallace 10, Naeem Bhojani 1,5

Introduction

The Canadian Undergraduate Urology Curriculum (CanUUC), developed by the CUA’s medical education experts, is an educational resource for medical students. Since multiple-choice questions (MCQs) and flashcards are widely used for clerkship and exam preparation, it was determined that the CanUUC could be strengthened with these interactive learning materials; however, developing these materials can be time-consuming. Therefore, we developed a customized large language model-enabled system to generate MCQs and flashcards from CanUUC material and evaluated the quality and curricular alignment of its outputs.

Methods

We created a custom system enabled by OpenAI’s gpt-5-2025-08-07 model. We iteratively refined two system prompts to generate educational materials based on best practices, including Medical Council of Canada MCQ writing guidelines. The prompts included instructions to derive content only from CanUUC materials and provide appropriate source reference. The benign prostatic hyperplasia (BPH) and urolithiasis CanUUC slides were used to generate the items. Six evaluators blinded to items’ origin were involved (four urologists and two residents), and each item received two independent ratings using standardized five-point Likert scales capturing relevance, correctness, clarity, and distractor quality. Open-ended questions were completed if “1, 2, or 3” scores were selected. Suitability for education was assessed through a four-point Likert scale with no neutral option. Items with mean suitability score <3 were excluded. Results were summarized descriptively.

Results

Ninety items (45 BPH, 45 urolithiasis), including 20 MCQs and 25 flashcards per topic, were generated. Exact source citations were present for all MCQs (100%) and most flashcards (94%). Common MCQ flaws, such as negative stems and instances where the correct answer was obviously more detailed than the distractors, were absent. Among flashcards, 22% (11/50) allowed >1 plausible answer. Across 180 ratings, median scores were 5 (IQR 4–5) for all domains. The proportion of top ratings (4–5) was high: 86.1% for relevance, 85.6% and 86.7% for questions and explanations’ correctness, respectively; 85.5% and 90.5% for questions and explanations’ clarity, respectively; and 90.1% for distractors’ quality. Overall, 86.1% of ratings judged items suitable for medical education. Eleven items (five MCQs, six flashcards) with mean suitability score <3 were excluded. Items were revised following reviewers’ comments, and 87.8% of the initially generated items were included in the CanUUC.

Conclusions

Our customized system can produce curriculum-aligned MCQs and flashcards that meet item-writing standards and are mostly suitable for medical education. Expert review remains essential to identify flaws and improve quality before inclusion into the curriculum.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S59.

MP 6.4. Use of artificial intelligence in delivering objective structured clinical examinations for postgraduate urology training

Ali Hosseinzadeh 1,✉, Thomas de los Reyes 2, Katherine Anderson 2, Christopher Nguan 2

Introduction

Objective structured clinical examinations (OSCEs) are central to assessing clinical competence but are resource-intensive and prone to examiner variability. Advances in generative artificial intelligence (AI) offer potential to automate OSCE delivery and evaluation. This study assessed the feasibility and validity of a custom generative pre-trained transformer (GPT) to generate and deliver OSCE stations for postgraduate urology residents and to compare AI-based with faculty scoring.

Methods

We conducted a prospective validation study at the University of British Columbia Department of Urologic Sciences. ChatGPT-4 generated and administered two OSCE stations for postgraduate year 3–5 urology residents. Stations simulated common urologic scenarios and were reviewed by faculty to ensure clinical accuracy. Performances were scored by ChatGPT using structured rubrics and independently graded by three blinded faculty examiners. Agreement between AI and human grading was assessed using correlation coefficients, intraclass correlation coefficients (ICC), and Bland-Altman analysis. Scores were also compared with other OSCE stations to assess construct validity.

Results

Nine residents completed both stations. Mean human-graded and AI-graded scores were 51±19% vs. 65±16% for case 1 and 38±11% vs. 36±12% for case 2, respectively. Strong correlations were observed between AI and human graders (case 1: r=0.95, p<0.001; case 2: r=0.83, p=0.011), with moderate-to-high agreement (ICC 0.70 and 0.83). Bland-Altman analysis demonstrated minimal bias. Over 80% of participants agreed the stations reflected appropriate realism and educational relevance.

Conclusions

AI-assisted OSCE generation and evaluation using ChatGPT is feasible and demonstrates close alignment with faculty grading in postgraduate urology training. This approach may serve as a scalable adjunct to competency-based assessment, reducing examiner burden while maintaining validity, provided appropriate human oversight is maintained.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S59–S60.

MP 6.5. Stakeholder perspectives on undergraduate urology education in Canada: The roles of urology and surgery interest groups and the Canadian Urology Student Interest Group

Hyunwoong Harry Chae 1,✉, Nicole Acupinpin 2, David-Dan Nguyen 3,4, Stacy de Lima 2, Katherine Anderson 6, Jason Y Lee 3, Dawn L MacLellan 2, David Bouhadana 5

Introduction

Medical school interest groups (IGs) can complement undergraduate education by providing exposure to specialties beyond formal curricula. The Canadian Urology Student Interest Group (CUSIG) is a novel, national model of IG aiming to advance Canadian urology education that has ongoing collaborations with the Canadian Undergraduate Urology Committee (CanUUC) of the Canadian Urological Association (CUA). To inform CUSIG’s ongoing development and collaborations, a survey of Canadian urologists’ perspectives on the role of urology/surgery IGs in undergraduate urology education was conducted.

Methods

A survey assessing demographics, perceptions of IGs and CUSIG, and involvement with IGs was developed, reviewed by Canadian urology educators, and distributed to CanUUC members, urology program directors, and other Canadian urology faculty. Data were summarized descriptively.

Results

Forty-three Canadian urologists responded. Respondents believed shadowing (37/43, 86%), mentorship (29/43, 67.4%), and research programs (25/43, 58.1%) to be the most effective IG initiatives. While 28/43 (65.1%) respondents were not previously aware of CUSIG, they agreed that CUSIG could help advance Canadian urology education (3.69±0.80). Stakeholders identified that the key priorities for CUSIG should be increasing exposure (37/43, 86.0%), student-resident-staff collaborations (27/43, 62.8%), and access to research (23/43, 53.5%). Barriers to staff engagement with IGs and CUSIG included lack of awareness (22/43, 51.2%) and time (17/43, 39.5%).

Conclusions

Canadian urologists endorsed CUSIG’s national IG model and identified clear priorities, including increasing clinical and research exposure, fostering mentoring relationships, increasing staff awareness, and delivering feasible, low-burden initiatives. These findings can guide CUSIG’s ongoing development and collaborations with CanUUC and the CUA to help address gaps in Canadian undergraduate urology curricula.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S60.

MP 6.6. Artificial intelligence in urology training: Enhancing annotation, feedback, and evaluation in robotic, laparoscopic, and endoscopic surgery

Jackie Han 1,✉, Le Yi He 2, Ankit Parmar 1, Kofi Brako 1, Edward Matsumoto 1,3

Introduction

The integration of artificial intelligence (AI) into surgical training is rapidly evolving, driven by advancements in machine learning; however, its potential in urology remains under-explored. This study is the first to provide a comprehensive overview of AI-based tools in urologic procedures and aims to map the current landscape of AI’s educational applications in urology.

Methods

A systematic search of MEDLINE, PubMed, Embase, Cochrane, Scopus, and Engineering Village identified studies exploring AI applications in video-based surgical education and assessment. Search terms included AI, machine learning, urologic procedures, and training/assessment components. Results were screened in Covidence®. AI applications focused on urologic procedures were included. For every study, two reviewers independently conducted screening. Data were synthesized thematically to evaluate AI’s application in urology training.

Results

Our search yielded 2774 studies, out of which 59 relevant ones were identified (Figure 1). AI was most frequently applied with robotic-assisted radical prostatectomy (RARP), followed by nephrectomy (RAPN) (Figure 2). AI applications were broadly categorized into three domains: 1) annotation, where key anatomy and instruments from procedural videos are labelled; 2) feedback, such as recognizing surgical phases or monitoring surgical milestone events; and 3) evaluation, where the surgical gestures are recognized or evaluated to stratify skill level and predict patient outcomes.

MP 6.6. Figure 1.

MP 6.6. Figure 1

Open in a new tab

Flowchart of screened references and reasons of exclusion.

MP 6.6. Figure 2.

MP 6.6. Figure 2

Open in a new tab

AI application domains across RARP, RAPN, and endourology procedures.

Conclusions

The emergence of AI use in urologic procedures underscores its transformative potential in procedural education and training. AI has wide applications in annotation, feedback, and assessment across different procedures. While prostatectomy dominates in the literature, the adaptability of AI frameworks exists across other urologic procedures. New commercially available tools demonstrate promising results, making them potentially beneficial additions to urology training programs. Future efforts should focus on multicentric collaboration and longitudinal skill assessments.

Acknowledgements

The content of the abstract was accepted by CUAJ and scheduled for publication in the June 2026 issue of the journal.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S60.

MP 6.7. Financial literacy among graduating Canadian urology residents compared to faculty

Archit Jain 1,✉, Naji J Touma 1

Introduction

Personal financial literacy influences well-being, particularly during the transition from residency to independent practice, when residents transform into small business owners. Despite growing emphasis on wellness, many residency programs provide little education on financial management. This study evaluated the financial literacy of graduating Canadian urology residents in comparison to staff urologists. It also examined perceptions of financial competence and the frequency of financial discussions between both groups.

Methods

We conducted a cross-sectional survey study at a national preparatory exam for graduating Canadian urology residents. The survey was administered to all residents and faculty attending the event, and the response rate was 100% in both groups. We created a 40-item survey with demographic questions (11), Likert-scale questions about comfort and perception of financial literacy (5), reimbursement-related questions to a urology practice (12), and validated financial literacy questions (9 true/false, 11 multiple-choice). Descriptive statistics were generated, and responses were analyzed using Welch’s t-test for group comparisons.

Results

A total of 38 residents and 13 staff completed the survey. Staff achieved higher objective financial literacy scores (77% vs. 63%, p=0.004). Staff also reported a greater perception in financial literacy skills (p<0.05). Both groups viewed financial literacy as highly important (p=0.765). Most respondents felt finances were discussed “not enough” during residency. Regarding reimbursement, both groups were equally likely to be wrong about true compensation, with a delta varying from 0.4–44.1%.

Conclusions

Staff urologists outperformed residents in both validated financial literacy questions and perceived self-competence in financial literacy. Despite these differences, both groups indicated a lack of adequate discussion of financial topics during residency. Discrepancies in perceived vs. actual reimbursement underscore the importance of structured financial education within residency programs to ease the transition to practice.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S61.

MP 6.8. Research productivity in Canadian academic urology: Does subspecialty matter?

Andrew R Maher 1,✉, Kennedy L Dirk 2, Naji J Touma 2

Introduction

Research productivity is an important factor in assessing scholarly activity and predicting academic advancement. The H-index is an established metric that calculates the ‘H’ number of papers a researcher has published with at least ‘H’ number of citations. Another metric is the i-10 score, describing the number of papers with at least 10 citations. Previous studies demonstrated apparent differences in academic productivity among American urologists largely explained by career duration and subspecialty distribution; however, little is known regarding scholarly output across urologic subspecialties among Canadian urologists.

Methods

Academic urologists were identified across the 14 Canadian academic urology residency programs. Research productivity was measured using the H-index (Scopus) and i-10 score (Google Scholar). Faculty were grouped into seven fellowship subspecialties: oncology, endourology/MIS, pediatrics, female/reconstruction, andrology/fertility, transplant, and general urology. Kruskal-Wallis test was used to compare H-index and i-10 scores across subspecialties. Regression adjusted for career duration and academic rank was conducted.

Results

We identified a total of 263 academic urologists; 222 (84.4%) males and 41 (15.6%) females. The most common subspecialties were oncology (30.0%), general urology (16.7%), and endourology/MIS (15.2%). Significant differences (p<0.01) in H-index scores between subspecialties were observed; oncology had the highest median H-index (31.5), followed by pediatrics (20.5), and andrology/fertility (15.0). Similarly, significant differences (p=0.01) in median i-10 scores were observed; the top subspecialties were oncology (125.0), pediatrics (64.5), and general urology (36.0).

Conclusions

Subspecialty-driven differences in scholarly productivity exist. Understanding these patterns is essential for evaluating academic structures, mentorship models, and promotion metrics in urology.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S61.

MP 6.9. Team training with robotic surgery simulators: An analysis of communication and surgical skills

Jeffrey A Sioufi 1,2,3,✉, Anne Yin 1,4, Caroline White 5, Sero Andonian 1,2

Introduction

Robotic surgery is challenging to learn due to the lack of tactile feedback, limited access, and complex techniques. Many simulators were developed as training tools, but none address the communication challenges inherent to robotic surgery. The RobotiX Mentor and LAP Mentor Express are the first simulators of their kind to enable team training, allowing a second trainee to participate as a bedside assistant. Our study aimed to assess whether there is a correlation between communication and performance during these exercises.

Methods

This prospective, observational study included 12 urology residents, and six attending urologists at an academic institution (n=18). Following completion of demographic questionnaires, participants were randomly divided into surgeon/assistant pairs. Participants completed individual warmup exercises on each simulator, followed by three increasingly difficult team training exercises. They then switched surgeon/assistant roles and repeated the exercises. Performance scores were obtained for each exercise. Audiovisual recordings of each simulation were scored by two reviewers using the validated Clinical Teamwork Scale to assess communication skills on a 0–10 scale. Participants were divided into novice, intermediate and expert groups based on training level. Each simulation served as an individual data point (n=54). Pearson’s correlation was used to compare communication and performance scores.

Results

Novices’ communication scores were two points lower than intermediates and experts on average, both as surgeons (p<0.001) and assistants (p=0.009). When novices assisted, the surgeons’ communication scores were two points higher, on average (p=0.002). Surgeon communication scores correlated with faster task completion time (p=0.013). Although surgeon experience correlates with higher communication scores, we found no correlation between surgeon/assistant experience level and completion time (p=0.451). Interestingly, there was no correlation between assistant communication and completion time (p=0.448).

Conclusions

Our findings indicate that surgeon communication scores correlate with better performance. Additionally, novices communicated less effectively than intermediates and experts during team training exercises. This shows that communication is an important skill in robotic surgery, which improves with experience. The use of such simulations could allow trainees to improve their communication skills, leading to better performance.

Funding: Pathy Family, Urology Department of McGill University, Canadian Institutes of Health Research, and Fonds de recherche du Québec.

Acknowledgements

This abstract was presented as an interactive poster session at the AUA 2026 Annual Meeting.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S61.

MP 6.10. The CUAJ trainee peer-reviewer training program: Pilot evaluation of a structured peer review curriculum for Canadian urology residents and fellows

David-Dan Nguyen 1,2,3,✉, Adriana Modica 1, Andrew MacNeily 1,4, Sero Andonian 1,5, Kristen McAlpine 1,2,3, Naeem Bhojani 1,6, Michael Leveridge 1,7

Introduction

Peer review underpins academic publishing and represents a core scholarly competency; however, most trainees receive little structured or mentored instruction in conducting high-quality reviews. As the demand for rigorous peer review grows, journals must also cultivate the next generation of reviewers. For a society journal such as the Canadian Urological Association Journal (CUAJ), this also supports strengthening scholarly capacity within its Canadian urology community. We therefore developed and piloted the CUAJ Trainee Peer-Review Training Program.

Methods

We designed and implemented a one-year, structured, and mentored trainee peer-reviewer training program for Canadian urology residents and fellows. The program included: 1) a didactic seminar introducing principles of peer review, ethical considerations, and best practices; 2) a structured mock peer-review exercise with facilitated group feedback; and 3) completion of three mentored formal manuscript reviews under the supervision of editorial board members, with individualized feedback. Participants completed baseline and exit surveys assessing self-reported knowledge, confidence, and peer-review competencies, as well as providing feedback on the initiative.

Results

Nine trainees completed the program. Participants reported increased familiarity with the peer-review process (median 7 [IQR 4–8] before vs. 9 [IQR 8–10] after, p=0.02) and confidence in conducting peer review (5 [IQR 4–8] before vs. 9 [IQR 8–9] after, p=0.01). Improvements were also observed across multiple domains of critical appraisal and peer-review skills, including evaluation of study design, statistical methods, interpretation of results, and provision of constructive feedback. Participants reported a high likelihood of continuing to engage in peer review after the program (median 9, IQR 8–10). All participants indicated they would recommend the initiative to other trainees, and overall satisfaction with the program was high (median 9, IQR 8–10).

Conclusions

The CUAJ Trainee Peer-Review Training Program was feasible and well-received by trainees. In its pilot year, the program provided structured exposure to the peer-review process and was associated with improvements in trainees’ self-reported familiarity, confidence, and comfort with evaluating key components of scientific manuscripts. This trainee-focused peer-review initiative may represent a practical and sustainable strategy to strengthen the Canadian urology reviewer pipeline while equitably introducing trainees to the principles of critical appraisal and scholarly publishing.

Funding: Dr. David-Dan Nguyen is supported by a Canadian Institutes of Health Research (CIHR) Vanier Canada Graduate Scholarship (CGV-192647), a CMCC/Atrium Hold’em for Life Oncology Fellowship, a Schwartz Reisman Institute for Technology and Society Graduate Fellowship, and an AI4PH Doctoral Scholarship. This program was supported by the Canadian Urological Association.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S62.

MP 6.11. Urologists’ attitudes and perceptions toward artificial intelligence medical scribes

Mohammadali Saffarzadeh 1,✉, Nathan C Wong 1

Introduction

Administrative tasks are leading contributors to burnout among urologists. Artificial intelligence (AI) medical scribes use natural language processing to generate real-time clinical notes while reducing providers’ documentation burden; however, urologists’ experience with AI scribes and their readiness to integrate them into their practice is mainly unexplored. This study qualitatively and quantitatively explored urologists’ perceptions and experiences with AI scribes.

Methods

We conducted a two-pronged evaluation of urologists’ perspective of AI medical scribes: 1) cross-sectional survey study of practicing urologists in British Columbia at the annual BC Urologic Society Meeting 2025; and 2) focus group of select urologists across academic and community settings. Semi-structured discussions were centered around experience with AI scribes, perceived benefits and concerns, and openness to adoption. Descriptive and thematic analysis was performed.

Results

The survey was completed by 47 urologists and 21 participated in one of three focus group sessions. Current documentation practices of urologists were dictation (36.2%), templated notes (21.3%), typed notes (19.1%), AI scribes (17%), and handwritten notes (6.4%) (Table 1). Improved workflow efficiency (74.5%) and reduced administrative workload (72.3%) were the top perceived benefits of AI scribes. Most urologists (93%) were open to try or adopt AI scribes. Prominent concerns were accuracy, medicolegal accountability, patient privacy, upfront costs, and initial workflow disruption.

MP 6.11. Table 1.

Current practices and perceptions of Artificial Intelligence (AI) medical scribes (N=47)

Questions Number (%)
How do you currently document medical notes?
 Dictation using speech recognition software 17 (36.2%)
 Templates 10 (21.3%)
 Typed notes during patient visit 9 (19.1%)
 AI Scribe 8 (17.0%)
 Written notes, then transcribe later 3 (6.4%)
 Scribe (human) 0 (0%)
 Other 0 (0%)
How many hours do you spend documenting per week?
 <1 hour 3 (6.4%)
 1–2 hours 6 (12.8%)
 2–4 hours 10 (21.3%)
 4–6 hours 16 (34.0%)
 6–8 hours 5 (10.6%)
 >8 hours 7 (14.9%)
What do you perceive as the primary benefit of using AI scribes?
 Improved workflow efficiency 35 (74.5%)
 Reduced administrative workload 34 (72.3%)
 More time for patient care 18 (38.3%)
 Increased documentation accuracy 17 (36.2%)
 No benefit 0 (0%)
 Other 0 (0%)
What are the challenges to adopting AI scribes? (check all that apply)
 Integration challenges with existing EMR 34 (72.3%)
 Time-intensive training and learning curve (takes too long) 18 (38.3%)
 Privacy and data security concerns 16 (34.0%)
 Medicolegal liability issues 13 (27.7%)
 Costs 10 (21.3%)
 Documentation accuracy concerns 7 (14.9%)
 Risk of over-reliance and automation bias 7 (14.9%)
 Skepticism about the utility of AI 6 (12.8%)
 Limited customization for individual documentation styles 3 (6.4%)
 Other (please specify) 1 (2.1%)
How likely are you to try AI scribe technology in your practice?
 Very likely 39 (83.0%)
 Somewhat likely 5 (10.6%)
 Neutral 3 (6.4%)
 Not likely 0 (0%)
Open in a new tab

Conclusions

AI medical scribes are viewed as promising adjuncts to reduce documentation time and improve efficiency. Many urologists remain cautious about accuracy, privacy, and legal implications. Seamless integration, strong data security, and clinical oversight will be crucial for broad adoption. These insights may guide the implementation of AI scribes in urologic practice.

Funding: This research received funding support from the Canadian Urological Association Community Urologist CPD Grant.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S62–S63.

MP 6.12. Abstracts of systematic reviews presented at major urologic conferences are not transparently reported

Pardis Ziaeefar 1,✉, Irin Lin 2, Ali Abud 3, Bowen Yao 1, Philipp Dahm 1

Introduction

Systematic reviews (SRs) play a major role in informing evidence-based clinical practice and are frequently presented at urologic meetings. We performed this study to evaluate to what degree submitted abstracts are transparently reported.

Methods

Based on a protocol registered in OSF, we searched American Urological Association (AUA) and European Association of Urology (EAU) abstract proceedings from 2021–25 for SRs. We included SRs of studies of clinical outcomes and excluded methodologic studies. All steps of the study screening and data abstraction process were completed independently and in duplicate by two members of the review team. Transparent reporting was assessed using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) from 2020.

Results

We included 327 abstract proceedings, of which 154 (47.0.%) and 174 (53.0%) were from the AUA and EAU, respectively. In accordance with PRISMA 2020, most (283; 86.3%) self-identified as SRs in the title and nearly all (312, 95.1%) had an explicit objective. Both inclusion and exclusion criteria were rarely (21, 6.4%) reported and less than half (159, 48.5.0%) identified both their information sources and the date of the latest search. Approximately one in three (107, 32.6%) reported assessing risk of bias (RoB), and only a few (19, 5.8%) used the GRADE approach to rate the certainty of evidence. Less than one-quarter (76, 23.2%) provided information about registration, and very few indicated funding (15, 4.6%). SR abstracts presented at the EAU (vs. AUA) meeting were more likely to report RoB assessment (43.9% vs. 20.9%, p<0.001), and a protocol (26.2% vs, 17.9%, p=0.004).

Conclusions

Abstracts of SRs submitted to major urologic meetings lack transparency in a majority of PRISMA 2020 domains. GRADE use is rarely reported. There is a need for improved reporting quality through better training and education, as well as an enforcement of established standards.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S63.

MP 6.13. The resident physician’s learning curve for ultrasound-guided renal access

Manuel Armas-Phan 1, David B Bayne 1, Yi Li 1, Scott Wiener 1,2, Kevin Liu 3,✉, Fadl Hamouche 1,3, Samuel Zetumer 4, David T Tzou 1,5, Justin Ahn 1, Marshall Stoller 1, Vincente Elorrieta 6, Feres Camargo 6, Thomas Chi 6

Introduction

Obtaining percutaneous access to the collecting system remains an obstacle to the widespread adoption of percutaneous nephrolithotomy (PCNL) among urologists. This study sought to characterize the learning curve of urology trainees for ultrasound-guided percutaneous renal access with the hypothesis that less cases are required to reach competency compared to fluoroscopic-guided access.

Methods

This prospective, observational study analyzed fourth-year urology ultrasound-naive residents attempting consecutive ultrasound-guided percutaneous renal access for percutaneous nephrolithotomy. Primary outcomes included access success and time to access completion. A representative resident learning curve was generated by combining the access attempt series for each resident and then applying local polynomial smoothing with a Gaussian Kernel function.

Results

Ten fourth-year urology residents collectively performed 48 percutaneous renal access attempts under ultrasound guidance with an overall success rate of 73% (35/48). There were no differences in patient characteristics or outcomes after PCNL between successful and non-successful attempts. Local polynomial smoothing estimated that a fourth-year urology resident had a 62% probability of successfully obtaining ultrasound-guided percutaneous renal access after the first case. On the fifth, sixth, and seventh successive cases, the probability increased to 75%, 77%, and 89%, respectively. The mean first access completion time for a fourth-year urology resident was 3.6 minutes. On the seventh, eighth, and ninth successive cases, the completion time decreases to 3.4, 3.1, and 2.8 minutes, respectively.

Conclusions

Fourth-year urology residents with basic ultrasound training can learn ultrasound-guided percutaneous renal access in the majority of percutaneous nephrolithotomy cases after a relatively low number of cases. The learning curve to improve proficiency in ultrasound-guided renal access techniques in trainees appears to be as short as 5–7 successive PCNL cases.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S63.

MP 6.14. From Prairie precision to national impact: The Saskatoon origins of minimally invasive urology in Canada

Donald Fentie 1, Trustin Domes 1, Garson Chan 1, Kunal Jana 1,✉

Introduction

Canada’s journey with minimally invasive surgery (MIS) in urology began not in the largest academic centers, but in a Prairie city known for community-based excellence. This historiographic review celebrates the contributions of Drs. Donald Fentie and Peter Barrett of Saskatoon, who pioneered laparoscopic urology in Canada in the early 1990s. Their local innovations reshaped national surgical practice, laying the groundwork for MIS adoption across community and academic centers alike.

Methods

We conducted a narrative historiographic review of peer-reviewed publications, institutional records, and oral histories. Key early MIS urologic procedures and training efforts in Saskatoon were analyzed in the broader context of technological change and surgical culture in Canada. Emphasis was placed on mentorship, skill transfer, and the real-world pathways by which MIS spread to communities across the country.

Results

Dr. Fentie and Dr. Barrett performed Canada’s first laparoscopic nephrectomy in January 1993, with advancing early laparoscopic pelvic procedures. Operating out of Saskatoon, a mid-sized city without a formalized urology training program, these surgeons published landmark early papers, developed safe and reproducible techniques, and most importantly, trained others. They welcomed visiting urologists, hosted resident electives, and offered hands-on mentorship long before structured MIS training was widespread. Their approach blended surgical rigor with community-centered accessibility, enabling smaller centers to adopt advanced techniques and pushing the boundaries of what community urology could achieve.

Conclusions

The Saskatoon MIS story is a powerful example of how local excellence can spark national transformation. Drs. Fentie, and Barrett not only changed how urologic surgery was performed, they changed where and by whom it could be performed. Their legacy reminds us that surgical innovation need not be confined to academic giants. As we celebrate community urology at CUA 2026 in Saskatoon, we honor a legacy born here; proof that vision, mentorship, and practical excellence in a community setting can shape the future of Canadian urology.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S63–S64.

MP 6.15. Factors associated with future academic appointment of Canadian urology graduates: A cross-sectional study of the 2005–2015 cohort

Ihtisham Ahmad 1, Zizo Al-Daqqaq 2, Gurdit Sood 1, David-Dan Nguyen 3, Maximilien Laviolette-Brassard 4, Alan Cheng 5,✉, Shiv Kandappah 6, Adree Khondker 3, Monica A Farcas 3, Naeem Bhojani 4, Quoc-Dien Trinh 7, Christopher JD Wallis 3, Yonah Krakowsky 3, Jason Y Lee 3

Introduction

Career goals vary between Canadian urology residency graduates, with both community and academic paths being pursued. The purpose of this study was to investigate factors that are associated with their future academic appointment.

Methods

We performed a retrospective, cross-sectional study of graduates from Canadian urology programs to allow for ≥10 years of followup (2005–2015). Demographic, training, and bibliometric data were extracted from public sources. Academic appointment was defined as a full-time professorship (full/associate/assistant), active as of December 2025. Multivariable logistic regression was used to identify factors associated with a full-time academic appointment, with secondary linear regression models used to investigate scholarly output such as H-index and total publications.

Results

We identified 210 graduates from 11 of 13 programs; data from two programs was not publicly available. Among them, 70 (33%) held full-time academic appointments. In the primary multivariable model, completion of a graduate degree (OR 4.48, 95% CI 1.90–11.0), fellowship training (OR 3.50, 95% CI 1.34–10.4), and a higher number of publications during residency (OR 1.22 per publication, 95% CI 1.08–1.39) were independently associated with academic appointment (Tables 1, 2). In secondary analyses, these three factors were also significant predictors of greater long-term scholarly output (higher H-index and total publications). Pre-residency publications and gender were not significant predictors (p>0.05). Graduates in full-time academic practice were significantly more likely to have completed U.S. fellowships (65% vs. 43%, p=0.01) and to practice in the U.S. (26% vs. 12%, p=0.02) compared to community-focused peers.

MP 6. 15. Table 1.

Comparing career and bibliometric characteristics of Canadian urology graduates from 2005–2015

Variable Overall, n=210 Community-focused practice, n=140 Full-time academic practice, n=70 p
Male gender n (%) 177 (84%) 119 (85%) 58 (83%) 0.69
Median years since starting residency (IQR) 19 (16, 22) 16 (18, 22) 20 (17, 22) 0.16
Clinician-investigator program completed, n (%) 4 (2%) 1 (0.7%) 3 (4%) 0.11
Graduate degree completed, n (%) 54 (26%) 18 (13%) 36 (51%) <0.001*
Master 48 (23%) 18 (13%) 30 (43%) <0.001*
Doctorate 7 (3%) 1 (0.7%) 6 (9%) 0.01*
Fellowship completed, n (%) 152 (72%) 88 (63%) 64 (91%) <0.001*
 Men’s health/fertility 8 (5%) 3 (3%) 5 (8%) 0.28
 MIS/endourology 58 (38%) 44 (50%) 14 (22%) 0.001*
 Oncology 42 (28%) 18 (20%) 24 (38%) 0.03*
 Pediatrics 15 (10%) 11 (13%) 4 (6%) 0.27
 Reconstruction/women’s health 23 (15%) 11 (13%) 12 (19%) 0.36
 Transplantation 6 (4%) 1 (1%) 5 (8%) 0.08
Second fellowship completed, n (%) 15 (7%) 9 (6%) 6 (9%) 0.72
Median pre-residency publications (IQR) 0 (0, 1) 0 (0, 0) 0 (0, 1) 0.04*
Median residency publications (IQR) 2 (1, 4) 1 (0, 3) 3 (1, 6) <0.001*
 Median total publications (IQR) 9 (3, 29) 5 (2, 11) 46 (20, 98) <0.001*
 Median first authorships (IQR) 3 (1, 8) 2 (0, 4) 10 (4, 21) <0.001*
 Median co-authorships (IQR) 5 (1, 19) 3 (1, 6) 26 (11, 64) <0.001*
 Median last authorships (IQR) 0 (0, 4) 0 (0, 0) 7 (2, 21) <0.001*
Median H-index 6 (2, 12) 3 (2, 7) 17 (8, 26) <0.001*
Open in a new tab
*

Statistically significant. Community-focused practice includes part-time or no academic appointments.

MP 6.15. Table 2.

Logistic regression model predicting future fulltime academic appointment of Canadian urology graduates from 2005–2015

Variable OR (95% CI) p
Male gender 0.54 (0.20–1.43) 0.21
Residency program (random program as reference)
 2 1.39 (0.27–6.86) 0.69
 3 1.25 (0.16–8.77) 0.83
 4 0.12 (0.01–0.84) 0.04*
 5 1.87 (0.34–10.16) 0.46
 6 0.74 (0.16–3.36) 0.69
 7 0.29 (0.04–1.68) 0.18
 8 0.71 (0.18–2.86) 0.63
 9 0.17 (0.03–0.80) 0.03*
 10 2.74 (0.56–14.0) 0.22
Clinician-investigator program completed 0.14 (0.01–4.38) 0.21
Graduate degree completed 4.48 (1.90–11.0) <0.001*
Fellowship completed 3.50 (1.34–10.4) 0.02*
Pre-residency publications 1.23 (0.92–1.70) 0.19
Residency publications 1.22 (1.08–1.39) 0.002*
Open in a new tab
*

Statistically significant.

Conclusions

Graduate degree, fellowship training, and residency publication output are factors associated with future academic appointment among Canadian urology training program graduates. These findings offer an evidence-based profile to guide trainees and program directors in career planning.

Can Urol Assoc J. 2026 Jun;20(6Suppl1):S64–S65.

MP 6.16. Out-of-the-box ChatGPT summarization of urology CaRMS applications: Reliability and risks

Joshua S Cheruvathur 1,✉, Tiange Li 2, William Wu 3, Jason Y Lee 2,3

Introduction

Each year, urology residency selection committees face a significant review burden from CaRMS application materials. Large language models (LLMs) are now widely accessible through conversational interfaces such as ChatGPT, and present an opportunity to reduce reviewer burden given their strong natural language processing capabilities. In practice, ChatGPT’s default consumer interface is often employed with minimal or no custom configuration, but the accuracy and reliability of this “out-of-the-box” use are unknown in CaRMS selection settings. We evaluated ChatGPT’s performance for summarizing urology residency applications using default settings.

Methods

For the 2025 CaRMS match cycle, all urology applicants granted interviews at the University of Toronto were included. For each applicant, de-identified application materials were submitted to ChatGPT (GPT-4o) via the consumer web interface with default application settings for conversation history and user memory. ChatGPT was prompted to generate a summary of each applicant’s personal letter, reference letters, and academic performance transcripts. Three human evaluators rated the ChatGPT summaries across seven domains, assessing summary accuracy, fabrications, and completeness. Discordant ratings were reviewed until consensus was reached. Inter-rater reliability was evaluated with Krippendorff ’s alpha. We additionally deployed the LLM-as-a-judge technique, via three frontier LLMs using zero-shot prompting and structured outputs, to perform the same evaluations of ChatGPT summaries.

Results

Thirty-two applicant files were included, yielding 469 human ratings across seven domains. Human inter-rater reliability was substantial (Krippendorff’s α=0.79). GPT-5, Claude Opus 4.5, and Gemini-3-Pro performed independent LLM-as-a-judge evaluations; pooled LLM-judge consensus was obtained via majority voting. LLM-judges had moderate agreement with human consensus (75.4%; Krippendorff ’s α=0.58). Per human consensus, ChatGPT summaries were fully accurate in 73.7% (95% CI 67.9–79.5) of ratings (Figure 1). Per LLM-as-a-judge consensus, ChatGPT summaries were accurate in 67.0% (95% CI 61.2–72.3%) of ratings. ChatGPT errors were categorized as: 1) misinterpretation of summarization task (53% of errors); 2) fabrications/hallucinations (30%); or 3) data leakage between individual applicant materials (17%).

MP 6.16. Figure 1.

MP 6.16. Figure 1

Open in a new tab

Human consensus ratings of ChatGPT summaries across 7 domains.

Conclusions

Under a default, “out-of-the-box” configuration, ChatGPT-generated summaries of CaRMS applications were often rated acceptable, but exhibited non-trivial errors, including substantial fabrications, omission of merit-based applicant differentiators, and cross-applicant data leakage. These findings support caution in using consumer LLM interfaces to inform selection decisions. Careful implementation of custom LLM pipelines, including rigorous applicant data isolation, disabling conversation memory features, structured outputs and tracing, and end-user training on LLM limitations, will be important in this setting.

Articles from Canadian Urological Association Journal are provided here courtesy of Canadian Urological Association

RESOURCES