Dominik Stefan Schoeb

Background: Domain-specific large language models (LLMs) like Ortho GPT have potential advantages over general-purpose models in medical education, offering improved factual accuracy and contextual relevance. This study evaluates the performance of Ortho GPT against general LLMs and senior medical students on validated orthopedic examination questions. Methods: Six LLMs (Ortho GPT 4o, ChatGPT 4o, ChatGPT 3.5, Perplexity AI, DeepSeek-R1, and Llama 3.3-70B) were tested using multiple-choice items from final-year medical student orthopedic exams in German language. Each model answered identical questions under standardized zero-shot conditions; accuracy rates and item-level results were compared using McNemar’s test, Jaccard similarity, and point-biserial correlation with student difficulty ratings. Results: Ortho GPT achieved the highest accuracy across models. McNemar’s tests revealed the significant superiority of Ortho GPT over DeepSeek (p = 2.33 × 10⁻³⁵), Llama 3.3-70B (p = 1.11 × 10⁻³²), and Perplexity (p = 4.01 × 10⁻⁵). Differences between Ortho GPT and ChatGPT 4o were non-significant (p = 0.065), suggesting near-equivalent performance to the strongest general model. No LLM showed correlation with student item difficulty (|r| < 0.07, p > 0.05), indicating that models solved items independently of human-perceived difficulty. Jaccard indices suggested moderate overlap between Ortho GPT and ChatGPT 4o, but distinct response profiles compared with general LLMs. Conclusions: These findings illustrate the superiority of Ortho GPT in orthopedic exam accuracy and context relevance, attributed to its specialized training data. The domain-specific approach enables performance matching or exceeding top general LLMs in orthopedics, emphasizing the importance of domain specialization for reliable, curriculum-aligned support in medical education.

Background: Up to 50% of patients with benign prostate hyperplasia (BPH) face post-operative complications after transurethral prostate Surgery. This includes nearly 15% of patients suffering from prolonged post-operative pelvic pain (pPPP) consisting of dysuria and prostatodynia for sometimes several months post-surgery. This study elucidates and proposes a definition of the multi-dimensional nature of prolonged post-operative pelvic pain (pPPP) after transurethral prostate surgery by providing real-world data on therapeutic options and their efficacy. Methods: German and international urologic practitioners participated in an online-survey after invitation via social media accounts and newsletters. The survey included questions on the amount of expertise, the therapeutic regimens for prolonged post-operative pelvic pain (pPPP) and the expected therapy outcome. Participation was voluntary and uncompensated. Chi-Square tests and Student’s t-tests were used for descriptive statistics. Results: 67 German urologists participated in a 9-question online Survey. 94,0% treated patients with lower urinary tract symptoms and 85,1% disclosed their therapeutic regime for prolonged post-operative pelvic pain (pPPP). The most common treatments included anti-inflammatory medication (69,6%), anti-cholinergics (53,6%), alpha-blockers (51,8%) and pelvic physiotherapy (50,0%). Over half of the patients responded to the therapeutic approach, but only 5,4% of urologists anticipated full pain relief after one year. These findings are closely aligned with a recent international survey (n=230). Notably, German urologists more frequently prescribed anti-cholinergics (53,6% vs. 28,7%, p = 0,0004), herbal remedies like saw palmetto (25,0% vs. 6,5%, p < 0,0001) and non-pharmacological therapies (82,1% vs. 49,1%, p < 0,0001), but less anti-inflammatory drugs (69,6% vs. 88,7%, p = 0,0003), gabapentin/pregabalin (8,9% vs. 42,2%, p < 0,0001) and opioids (0% vs 5,7%, p = 0,0221). Based on these results a structured definition and therapy plan for prolonged post-operative pelvic pain (pPPP) is proposed. Conclusion: Prolonged post-operative pelvic pain (pPPP) is a common challenge for urologists. Despite various therapeutic options, treatment outcomes and practitioner confidence remain suboptimal. Further research and attention to prolonged post-operative pelvic pain (pPPP) are essential to develop evidence-based guidelines for effective patient management and prevention of chronic pain syndromes.

Introduction: Magnetic resonance imaging‐transrectal ultrasound (MRI‐TRUS)‐fusion biopsy (FBx) of the prostate allows targeted sampling of suspicious lesions within the prostate, identified by multiparametric MRI. Due to its reliable results and feasibility, perineal MRI/TRUS FBx is now the gold standard for prostate cancer (PC) diagnosis. There are various systems for performing FBx on the market, for example, software‐based, semirobotic, or robot‐assisted platform solutions. Their semiautomated workflow promises high process quality independent of the surgeon's experience. The aim of this study was to analyze how the surgeon's experience influences the cancer detection rate (CDR) via targeted biopsy (TB) and the procedure's duration in robot‐assisted FBx.Patients and Methods: A total of 1716 men who underwent robot‐assisted FBx involving a combination of targeted and systematic sampling between October 2015 and April 2022 were analyzed. We extracted data from the patients' electronic medical records retrospectively. Primary endpoints were the CDR by TB and the procedure's duration. For our analysis, surgeons were divided into three levels of experience: ≤20 procedures (little), 21–100 procedures (intermediate), and >100 procedures (high). Statistical analysis was performed via regression analyses and group comparisons.Results: Median age, prostate‐specific antigen level, and prostate volume of the cohort were 67 (±7.7) years, 8.13 (±9.4) ng/mL, and 53 (±34.2) mL, respectively. Median duration of the procedure was 26 (±10.9) min. The duration decreased significantly with the surgeon's increasing experience from 35.1 (little experience) to 28.4 (intermediate experience) to 24.0 min (high experience) (p < 0.001). Using TB only, significant PC (sPC) was diagnosed in 872/1758 (49.6%) of the men. The CDR revealed no significant correlation with the surgeon's experience in either group comparison (p = 0.907) or in regression analysis (p = 0.65).Conclusion: While the duration of this procedure decreases with increasing experience, the detection rate of sPC in TB is not significantly associated with the experience of the surgeon performing robot‐assisted FBx. This robot‐assisted biopsy system's diagnostic accuracy therefore appears to be independent of experience.

Purpose: The question of best surgical treatment for lower urinary tract symptoms (LUTS) due to benign prostate hyperplasia (BPH) remains controversial. We compared the outcomes of aquablation and holmium laser enucleation of the prostate ("HoLEP") in a prospective cohort. Methods: Patients with BPH underwent aquablation or HoLEP according to their preference between June 2020 and April 2022. Prostate volume (“PV”), laboratory results, postvoid residual volume, uroflowmetry, IPSS, ICIQ-SF, MSHQ-EjD, EES and IIEF were evaluated preoperatively and at three, six and 12 months postoperatively. We also analyzed perioperative characteristics and complications via the Clavien Dindo („CD“) classification. Results: We included 40 patients, 16 of whom underwent aquablation and 24 HoLEP. Mean age was 67 years (SD 7.4). Baseline characteristics were balanced across groups, except the HoLEP patients’ larger PV. IPSS fell from 20.3 (SD 7.1) at baseline to 6.3 (SD 4.2) at 12 months (p < 0.001) without differences between aquablation and HoLEP. HoLEP was associated with shorter operation time (59.5 (SD 18.6) vs. 87.2 (SD 14.8) minutes, p < 0.001) and led to better PV reduction over all timepoints. At three months, aquablation’s results were better regarding ejaculatory (p = 0.02, MSHQ-EjD) and continence function (p < 0.001, ICIQ-SF). Beyond three months, erectile, ejaculatory, continence function and LUTS reduction did not differ significantly between aquablation and HoLEP. CD ≥ grade 3b complications were noted in six patients in aquablation group while only one in HoLEP group (p = < 0.01). Conclusions: While aquablation revealed temporary benefits regarding ejaculation and continence at three months, HoLEP was superior concerning operation time, the safety profile and volumetric results.

Background: To review and discuss the literature on applying tyrosine kinase inhibitors (TKIs) in the treatment of metastasised renal cell carcinoma (mRCC). Materials and Methods: Medline, PubMed, the Cochrane database, and Embase were screened for randomised controlled trials, clinical trials, and reviews on treating renal cell carcinoma, and the role of TKI. Each substance’s results were summarised descriptively. Results: While TKI monotherapy is not currently recommended as a first-line treatment for metastasized renal cell carcinoma, TKIs are regularly applied to treat treatment-naïve patients in combination with immunotherapy. TKIs depict the first-choice alternative therapy if immunotherapy is not tolerated or inapplicable. Currently, seven different TKIs are available to treat mRCC. Conclusions: The importance of TKIs in a monotherapeutic approach has declined in the past few years. The current trend toward combination therapy for mRCC, however, includes TKIs as one significant component of treatment regimens. We found that to remain applicable to ongoing studies, both when including new substances and when testing novel combinations of established drugs. TKIs are of major importance for the treatment of renal cancer now, as well as for the foreseeable future.

Background: The gold standard for detecting bladder cancer is white light cystoscopy (WLC) and resection of suspicious lesions. In this study, we evaluate two miniaturized Optical Coherence Tomography (OCT) probes for endoscopic use, regarding their applicability in diagnosing urothelial cancer. Materials and methods In total, 33 patients who underwent a radical cystectomy were included. Preoperative oncological staging and determining the indication for the surgical intervention were done following the latest European Association of Urology (EAU) guidelines. Samples were taken from bladder tissue after bladder removal and prepared for OCT measurement. Additionally, porcine bladder samples were used as reference tissue. We took measurements using two miniaturized probes: a bimodal probe and a single modality OCT probe. A non-miniaturized standard OCT scanner was used as a reference. Results: Histopathological examination revealed urothelial cancer in all but three patients. Measurements on porcine tissue revealed a clear distinction between the urothelial layers for all probes. Furthermore, we detected improved image quality thanks to the stretching of the tissue. We took 271 measurements in human samples. While the urothelial layers were well delineated in healthy tissue, all the probes revealed a loss of these structures in cancerous regions. While the single-modality probe delivered an image quality equaling the reference images, it was possible to detect cancerous areas with the bimodal probe. Conclusion: We demonstrate that endoscopic probes for OCT imaging are technologically feasible and deliver acceptable image quality. A distinction between healthy and abnormal tissue is possible. We propose combining different endoscopic imaging modalities as a promising approach for urothelial cancer diagnostics.

We developed a new mobile ultrasound device for long-term and automated bladder monitoring without user interaction consisting of 32 transmit and receive electronics as well as a 32-element phased array 3 MHz transducer. The device architecture is based on data digitization and rapid transfer to a consumer electronics device (e.g., a tablet) for signal reconstruction (e.g., by means of plane wave compounding algorithms) and further image processing. All reconstruction algorithms are implemented in the GPU, allowing real-time reconstruction and imaging. The system and the beamforming algorithms were evaluated with respect to the imaging performance on standard sonographical phantoms (CIRS multipurpose ultrasound phantom) by analyzing the resolution, the SNR and the CNR. Furthermore, ML-based segmentation algorithms were developed and assessed with respect to their ability to reliably segment human bladders with different filling levels. A corresponding CNN was trained with 253 B-mode data sets and 20 B-mode images were evaluated. The quantitative and qualitative results of the bladder segmentation are presented and compared to the ground truth obtained by manual segmentation.

Evaluation of functional parameters, patient-reported outcomes and workload related to continuous urinary bladder irrigation after transurethral surgery Anja Christina Reichelt, Franz Friedrich Dressler, Christian Gratzke, Arkadiusz Miernik, Dominik Stefan Schoeb