The Impact of Social Deprivation on Anterior Urethral Stricture Recurrence After Urethroplasty - A Trauma and Urologic Reconstructive Network of Surgeons (TURNS) Analysis.

INTRODUCTION AND OBJECTIVES: Several factors influence recurrence after urethral stricture repair. The impact of socioeconomic factors on stricture recurrence after urethroplasty is poorly understood. This study aims to assess the impact that social deprivation, an area-level measure of disadvantage, has on urethral stricture recurrence after urethroplasty. METHODS: We performed a retrospective review of patients undergoing urethral reconstruction by surgeons participating in a collaborative research group. Home zip code was used to calculate Social Deprivation Indices (SDI; 0-100), which quantifies the level of disadvantage across several sociodemographic domains collected in the American Community Survey. Patients without zip code data were excluded from the analysis. The Cox Proportional Hazards model was used to study the association between SDI and the hazard of functional recurrence, adjusting for stricture characteristics as well as age and body mass index. RESULTS: Median age was 46.0 years with a median follow up of 367 days for the 1452 men included in the study. Patients in the fourth SDI quartile (worst social deprivation) were more likely to be active smokers with traumatic and infectious strictures compared to the first SDI quartile. Patients in the fourth SDI quartile had 1.64 times the unadjusted hazard of functional stricture recurrence vs patients in the first SDI quartile (95% CI 1.04-2.59). Compared to anastomotic ± excision, substitution only repair had 1.90 times the unadjusted hazard of recurrence. The adjusted hazard of recurrence was 1.08 per 10-point increase in SDI (95% CI 1.01-1.15, P = .027). CONCLUSIONS: Patient social deprivation identifies those at higher risk for functional recurrence after anterior urethral stricture repair, offering an opportunity for preoperative counseling and postoperative surveillance. Addressing these social determinants of health can potentially improve outcomes in reconstructive surgery.

Fully Automated Deep Learning Model to Detect Clinically Significant Prostate Cancer at MRI.

Background Multiparametric MRI can help identify clinically significant prostate cancer (csPCa) (Gleason score ≥7) but is limited by reader experience and interobserver variability. In contrast, deep learning (DL) produces deterministic outputs. Purpose To develop a DL model to predict the presence of csPCa by using patient-level labels without information about tumor location and to compare its performance with that of radiologists. Materials and Methods Data from patients without known csPCa who underwent MRI from January 2017 to December 2019 at one of multiple sites of a single academic institution were retrospectively reviewed. A convolutional neural network was trained to predict csPCa from T2-weighted images, diffusion-weighted images, apparent diffusion coefficient maps, and T1-weighted contrast-enhanced images. The reference standard was pathologic diagnosis. Radiologist performance was evaluated as follows: Radiology reports were used for the internal test set, and four radiologists' PI-RADS ratings were used for the external (ProstateX) test set. The performance was compared using areas under the receiver operating characteristic curves (AUCs) and the DeLong test. Gradient-weighted class activation maps (Grad-CAMs) were used to show tumor localization. Results Among 5735 examinations in 5215 patients (mean age, 66 years ± 8 [SD]; all male), 1514 examinations (1454 patients) showed csPCa. In the internal test set (400 examinations), the AUC was 0.89 and 0.89 for the DL classifier and radiologists, respectively (P = .88). In the external test set (204 examinations), the AUC was 0.86 and 0.84 for the DL classifier and radiologists, respectively (P = .68). DL classifier plus radiologists had an AUC of 0.89 (P < .001). Grad-CAMs demonstrated activation over the csPCa lesion in 35 of 38 and 56 of 58 true-positive examinations in internal and external test sets, respectively. Conclusion The performance of a DL model was not different from that of radiologists in the detection of csPCa at MRI, and Grad-CAMs localized the tumor. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Johnson and Chandarana in this issue.

A Multi-View Deep Learning Model for Thyroid Nodules Detection and Characterization in Ultrasound Imaging.

Thyroid Ultrasound (US) is the primary method to evaluate thyroid nodules. Deep learning (DL) has been playing a significant role in evaluating thyroid cancer. We propose a DL-based pipeline to detect and classify thyroid nodules into benign or malignant groups relying on two views of US imaging. Transverse and longitudinal US images of thyroid nodules from 983 patients were collected retrospectively. Eighty-one cases were held out as a testing set, and the rest of the data were used in five-fold cross-validation (CV). Two You Look Only Once (YOLO) v5 models were trained to detect nodules and classify them. For each view, five models were developed during the CV, which was ensembled by using non-max suppression (NMS) to boost their collective generalizability. An extreme gradient boosting (XGBoost) model was trained on the outputs of the ensembled models for both views to yield a final prediction of malignancy for each nodule. The test set was evaluated by an expert radiologist using the American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS). The ensemble models for each view achieved a mAP0.5 of 0.797 (transverse) and 0.716 (longitudinal). The whole pipeline reached an AUROC of 0.84 (CI 95%: 0.75-0.91) with sensitivity and specificity of 84% and 63%, respectively, while the ACR-TIRADS evaluation of the same set had a sensitivity of 76% and specificity of 34% (p-value = 0.003). Our proposed work demonstrated the potential possibility of a deep learning model to achieve diagnostic performance for thyroid nodule evaluation.

Whole-body low-dose computed tomography in patients with newly diagnosed multiple myeloma predicts cytogenetic risk: a deep learning radiogenomics study.

OBJECTIVE: To develop a whole-body low-dose CT (WBLDCT) deep learning model and determine its accuracy in predicting the presence of cytogenetic abnormalities in multiple myeloma (MM). MATERIALS AND METHODS: WBLDCTs of MM patients performed within a year of diagnosis were included. Cytogenetic assessments of clonal plasma cells via fluorescent in situ hybridization (FISH) were used to risk-stratify patients as high-risk (HR) or standard-risk (SR). Presence of any of del(17p), t(14;16), t(4;14), and t(14;20) on FISH was defined as HR. The dataset was evenly divided into five groups (folds) at the individual patient level for model training. Mean and standard deviation (SD) of the area under the receiver operating curve (AUROC) across the folds were recorded. RESULTS: One hundred fifty-one patients with MM were included in the study. The model performed best for t(4;14), mean (SD) AUROC of 0.874 (0.073). The lowest AUROC was observed for trisomies: AUROC of 0.717 (0.058). Two- and 5-year survival rates for HR cytogenetics were 87% and 71%, respectively, compared to 91% and 79% for SR cytogenetics. Survival predictions by the WBLDCT deep learning model revealed 2- and 5-year survival rates for patients with HR cytogenetics as 87% and 71%, respectively, compared to 92% and 81% for SR cytogenetics. CONCLUSION: A deep learning model trained on WBLDCT scans predicted the presence of cytogenetic abnormalities used for risk stratification in MM. Assessment of the model's performance revealed good to excellent classification of the various cytogenetic abnormalities.

Transition-zone PSA-density calculated from MRI deep learning prostate zonal segmentation model for prediction of clinically significant prostate cancer.

PURPOSE: To develop a deep learning (DL) zonal segmentation model of prostate MR from T2-weighted images and evaluate TZ-PSAD for prediction of the presence of csPCa (Gleason score of 7 or higher) compared to PSAD. METHODS: 1020 patients with a prostate MRI were randomly selected to develop a DL zonal segmentation model. Test dataset included 20 cases in which 2 radiologists manually segmented both the peripheral zone (PZ) and TZ. Pair-wise Dice index was calculated for each zone. For the prediction of csPCa using PSAD and TZ-PSAD, we used 3461 consecutive MRI exams performed in patients without a history of prostate cancer, with pathological confirmation and available PSA values, but not used in the development of the segmentation model as internal test set and 1460 MRI exams from PI-CAI challenge as external test set. PSAD and TZ-PSAD were calculated from the segmentation model output. The area under the receiver operating curve (AUC) was compared between PSAD and TZ-PSAD using univariate and multivariate analysis (adjusts age) with the DeLong test. RESULTS: Dice scores of the model against two radiologists were 0.87/0.87 and 0.74/0.72 for TZ and PZ, while those between the two radiologists were 0.88 for TZ and 0.75 for PZ. For the prediction of csPCa, the AUCs of TZPSAD were significantly higher than those of PSAD in both internal test set (univariate analysis, 0.75 vs. 0.73, p < 0.001; multivariate analysis, 0.80 vs. 0.78, p < 0.001) and external test set (univariate analysis, 0.76 vs. 0.74, p < 0.001; multivariate analysis, 0.77 vs. 0.75, p < 0.001 in external test set). CONCLUSION: DL model-derived zonal segmentation facilitates the practical measurement of TZ-PSAD and shows it to be a slightly better predictor of csPCa compared to the conventional PSAD. Use of TZ-PSAD may increase the sensitivity of detecting csPCa by 2-5% for a commonly used specificity level.

Deep Phenotyping the Anterior Urethral Stricture: Characterizing the Relationship Between Inflammation, Fibrosis, Patient History, and Disease Pathophysiology.

PURPOSE: Anterior urethral stricture disease (aUSD) is a complex, heterogeneous condition that is idiopathic in origin for most men. This gap in knowledge rarely affects the current management strategy for aUSD, as urethroplasty does not generally consider etiology. However, as we transition towards personalized, minimally invasive treatments for aUSD and begin to consider aUSD prevention strategies, disease pathophysiology will become increasingly important. The purpose of this study was to perform a deep phenotype of men undergoing anterior urethroplasty for aUSD. We hypothesized that unique biologic signatures and potential targets for intervention would emerge based on stricture presence/absence, stricture etiology, and the presence/absence of stricture inflammation. MATERIALS AND METHODS: Men with aUSD undergoing urethroplasty were recruited from one of 5 participating centers. Enrollees provided urethral stricture tissue and blood/serum on the day of surgery and completed patient-reported outcome measure questionnaires both pre- and postoperatively. The initial study had 3 aims: (1) to determine pediatric and adult subacute and repeated perineal trauma (SRPT) exposures using a study-specific SRPT questionnaire, (2) to determine the degree of inflammation and fibrosis in aUSD and peri-aUSD (normal urethra) tissue, and (3) to determine levels of systemic inflammatory and fibrotic cytokines. Two controls groups provided serum (normal vasectomy patients) and urethral tissue (autopsy patients). Cohorts were based on the presence/absence of stricture, by presumed stricture etiology (idiopathic, traumatic/iatrogenic, lichen sclerosus [LS]), and by the presence/absence of stricture inflammation. RESULTS: Of 138 enrolled men (120 tissue/serum; 18 stricture tissue only), 78 had idiopathic strictures, 33 had trauma-related strictures, and 27 had LS-related strictures. BMI, stricture length, and stricture location significantly differed between cohorts (P < .001 for each). The highest BMIs and the longest strictures were observed in the LS cohort. SRPT exposures did not significantly differ between etiology cohorts, with > 60% of each reporting low/mild risk. Stricture inflammation significantly differed between cohorts, with mild to severe inflammation present in 27% of trauma-related strictures, 54% of idiopathic strictures, and 48% of LS strictures (P = .036). Stricture fibrosis did not significantly differ between cohorts (P = .7). Three serum cytokines were significantly higher in patients with strictures compared to stricture-free controls: interleukin-9 (IL-9; P = .001), platelet-derived growth factor-BB (P = .004), and CCL5 (P = .01). No differences were observed in the levels of these cytokines based on stricture etiology. However, IL-9 levels were significantly higher in patients with inflamed strictures than in patients with strictures lacking inflammation (P = .019). Degree of stricture inflammation positively correlated with serum levels of IL-9 (Spearman's rho 0.224, P = .014). CONCLUSIONS: The most common aUSD etiology is idiopathic. Though convention has implicated SRPT as causative for idiopathic strictures, here we found that patients with idiopathic strictures had low SRPT rates that were similar to rates in patients with a known stricture etiology. Stricture and stricture-adjacent inflammation in idiopathic stricture were similar to LS strictures, suggesting shared pathophysiologic mechanisms. IL-9, platelet-derived growth factor-BB, and CCL5, which were elevated in patients with strictures, have been implicated in fibrotic conditions elsewhere in the body. Further work will be required to determine if this shared biologic signature represents a potential mechanism for an aUSD predisposition.

Proposed revision of the American Association for Surgery of Trauma Renal Organ Injury Scale: Secondary analysis of the Multi-institutional Genitourinary Trauma Study.

BACKGROUND: This study updates the American Association for the Surgery of Trauma (AAST) Organ Injury Scale (OIS) for renal trauma using evidence-based criteria for bleeding control intervention. METHODS: This was a secondary analysis of a multicenter retrospective study including patients with high-grade renal trauma from seven level 1 trauma centers from 2013 to 2018. All eligible patients were assigned new renal trauma grades based on revised criteria. The primary outcome used to measure injury severity was intervention for renal bleeding. Secondary outcomes included intervention for urinary extravasation, units of packed red blood cells transfused within 24 hours, and mortality. To test the revised grading system, we performed mixed-effect logistic regression adjusted for multiple baseline demographic and trauma covariates. We determined the area under the curve (AUC) to assess accuracy of predicting bleeding interventions from the revised grading system and compared this to 2018 AAST OIS. RESULTS: Based on the 2018 OIS grading system, we included 549 patients with AAST grades III to V injuries and computed tomography scans (III, 52% [n = 284]; IV, 45% [n = 249]; and V, 3% [n = 16]). Among these patients, 89% experienced blunt injury (n = 491), and 12% (n = 64) underwent intervention for bleeding. After applying the revised grading criteria, 60% (n = 329) of patients were downgraded, and 4% (n = 23) were upgraded; 2.8% (n = 7) downgraded from grade V to IV, and 69.5% (n = 173) downgraded from grade IV to III. The revised renal trauma grading system demonstrated improved predictive ability for bleeding interventions (2018 AUC, 0.805; revised AUC, 0.883; p = 0.001) and number of units of packed red blood cells transfused. When we removed urinary injury from the revised system, there was no difference in its predictive ability for renal hemorrhage intervention. CONCLUSION: A revised renal trauma grading system better delineates the need for hemostatic interventions than the current AAST OIS renal trauma grading system. LEVEL OF EVIDENCE: Diagnostic Test/Criteria; Level III.

Longitudinal Outcomes of Malignant Ureteral Obstruction Secondary to Ovarian Cancer: Predictors of Resolution and the Role of Surgical Management.

OBJECTIVE: To review the management of ovarian cancer (OCa) associated hydronephrosis (HN). Specifically, we aim to identify optimal management of HN in the acute setting, predictors of HN resolution, and the role of surgery (tumor debulking/(+/-)ureterolysis/hysterectomy). MATERIALS/METHODS: The study cohort included OCa patients managed at our institution from 2004-2019 that developed OCa-associated HN. Initial HN management was recorded as none, retrograde ureteral stent (RUS) or percutaneous nephrostomy tube (PCN). Primary outcomes included (1) HN management failure, (2) HN management complications, and (3) HN resolution. Patient, cancer, and treatment predictors of outcomes were assessed using logistic regression and fine-Gray competing risk models. RESULTS: Of 2580 OCa patients, 190 (7.4%) developed HN. HN was treated in 121; 90 (74.4%) with RUS, 31 (25.6%) with PCN. Complication rates were similar between PCN and RUS (83% vs 85.1%; P = .79; all Clavian Grade I/II). Initial HN treatment failure occurred in 28 patients, predicted by renal atrophy (hazard ratios (HR) 3.27, P <.01). HN resolution occurred in only 52 (27%) patients and was predicted by lower International Federation of Gynecology and Obstetrics (FIGO) stage (FIGO III/IV HR 0.42, P <.01) and surgical tumor debulking/ureterolysis (HR 2.83, P = .02). CONCLUSION: Resolution of HN associated with malignant obstruction from OCa is rare and is most closely associated with tumor debulking and International Federation of Gynecology and Obstetrics (FIGO) stage. Initial endoscopic treatment modality was not significantly associated with complications or resolution, though RUS failures were slightly more common. Ureteral reconstruction at time of debulking/ureterolysis is potentially underutilized.

Association of Kidney Cysts With Progressive CKD After Radical Nephrectomy.

RATIONALE & OBJECTIVE: Simple kidney cysts, which are common and usually considered of limited clinical relevance, are associated with older age and lower glomerular filtration rate (GFR), but little has been known of their association with progressive chronic kidney disease (CKD). STUDY DESIGN: Observational cohort study. SETTING & PARTICIPANTS: Patients with presurgical computed tomography or magnetic resonance imaging who underwent a radical nephrectomy for a tumor; we reviewed the retained kidney images to characterize parenchymal cysts at least 5mm in diameter according to size and location. EXPOSURE: Parenchymal cysts at least 5mm in diameter in the retained kidney. Cyst characteristics were correlated with microstructural findings on kidney histology. OUTCOME: Progressive CKD defined by dialysis, kidney transplantation, a sustained≥40% decline in eGFR for at least 3 months, or an eGFR<10mL/min/1.73m2 that was at least 5mL/min/1.73m2 below the postnephrectomy baseline for at least 3 months. ANALYTICAL APPROACH: Cox models assessed the risk of progressive CKD. Models adjusted for baseline age, sex, body mass index, hypertension, diabetes, eGFR, proteinuria, and tumor volume. Nonparametric Spearman's correlations were used to examine the association of the number and size of the cysts with clinical characteristics, kidney function, and kidney volumes. RESULTS: There were 1,195 patients with 50 progressive CKD events over a median 4.4 years of follow-up evaluation. On baseline imaging, 38% had at least 1 cyst, 34% had at least 1 cortical cyst, and 8.7% had at least 1 medullary cyst. A higher number of cysts was associated with progressive CKD and was modestly correlated with larger nephrons and more nephrosclerosis on kidney histology. The number of medullary cysts was more strongly associated with progressive CKD than the number of cortical cysts. LIMITATIONS: Patients who undergo a radical nephrectomy may differ from the general population. A radical nephrectomy may accelerate the risk of progressive CKD. Genetic testing was not performed. CONCLUSIONS: Cysts in the kidney, particularly the medulla, should be further examined as a potentially useful imaging biomarker of progressive CKD beyond the current clinical evaluation of kidney function and common CKD risk factors. PLAIN-LANGUAGE SUMMARY: Kidney cysts are common and often are considered of limited clinical relevance despite being associated with lower glomerular filtration rate. We studied a large cohort of patients who had a kidney removed due to a tumor to determine whether cysts in the retained kidney were associated with kidney health in the future. We found that more cysts in the kidney and, in particular, cysts in the deepest tissue of the kidney (the medulla) were associated with progressive kidney disease, including kidney failure where dialysis or a kidney transplantation is needed. Patients with cysts in the kidney medulla may benefit from closer monitoring.

Does Rounding Order Bias Discharge Efficiency? Predictors of Discharge Timing on an Academic Urology Service.

ABSTRACT: No previous works have analyzed whether the order in which surgical teams see patients on morning rounds affects discharge efficiency at teaching hospitals. We obtained perioperative urologic surgery timing data at our academic institution from 2014 to 2019. We limited the analysis to routine postoperative day 1 discharges. Univariate and multivariate analyses were performed to determine whether various hospital and patient factors were associated with discharge timing. We analyzed 1,494 patients. Average discharge order time was 11:22 a.m. and hospital discharge 1:24 p.m. Univariate regression revealed earlier discharge order time for patients seen later in rounds by 4 minutes per sequential room cluster (p = .013) and by 12 minutes per cluster when excluding short-stay patients. Multivariate analysis revealed discharge order placement did not vary significantly by rounding order. However, time of hospital discharge did (p < .001), likely due to speed of discharge in the designated short-stay units. Attending physician was the most consistent predictor in variations of discharge timing, with statistical significance across all measured outcomes. Patients seen later in rounding progression received earlier discharge orders, but this relationship does not remain in multivariate modeling or translate to earlier discharge. These findings have helped guide quality improvement efforts focused on discharge efficiency.

Location and Types of Treatment for Prostate Cancer After the Veterans Choice Program Implementation.

Importance: The Veterans Choice Program (VCP) was implemented in 2014 to help veterans gain broader access to specialized care outside of the Veterans Health Administration (VHA) facilities by providing them with purchased community care (CC). Objective: To describe the prevalence and patterns in VCP-funded purchased CC after the implementation of the VCP among veterans with prostate cancer. Design, Setting, and Participants: This cohort study used VHA administrative data on veterans with prostate cancer diagnosed between January 1, 2015, and December 31, 2018. These veterans were regular VHA primary care users. Analyses were performed from March to July 2023. Exposures: Driving distance (in miles) from residence to nearest VHA tertiary care facility. The location (VHA or purchased CC) in which treatment decisions were made was ascertained by considering 3 factors: (1) location of the diagnostic biopsy, (2) location of most of the postdiagnostic prostate-specific antigen laboratory testing, and (3) location of most of the postdiagnostic urological care encounters. Main Outcomes and Measures: The main outcome was receipt of definitive treatment and proportion of purchased CC by treatment type (radical prostatectomy [RP], radiotherapy [RT], or active surveillance) and by distance to nearest VHA tertiary care facility. Quality was evaluated based on receipt of definitive treatment for Gleason grade group 1 prostate cancer (low risk/limited treatment benefit by guidelines). Results: The cohort included 45 029 veterans (mean [SD] age, 67.1 [6.9] years) with newly diagnosed prostate cancer; of these patients, 28 866 (64.1%) underwent definitive treatment. Overall, 56.8% of patients received definitive treatment from the purchased CC setting, representing 37.5% of all RP care and 66.7% of all RT care received during the study period. Most patients who received active surveillance management (92.5%) remained within the VHA. Receipt of definitive treatment increased over the study period (from 5830 patients in 2015 to 9304 in 2018), with increased purchased CC for patients living farthest from VHA tertiary care facilities. The likelihood of receiving definitive treatment of Gleason grade group 1 prostate cancer was higher in the purchased CC setting (adjusted relative risk ratio, 1.79; 95% CI, 1.65-1.93). Conclusions and Relevance: This cohort study found that the percentage of veterans receiving definitive treatment in VCP-funded purchased CC settings increased significantly over the study period. Increased access, however, may come at the cost of low care quality (overtreatment) for low-risk prostate cancer.

Few-shot biomedical image segmentation using diffusion models: Beyond image generation.

BACKGROUND: Medical image analysis pipelines often involve segmentation, which requires a large amount of annotated training data, which is time-consuming and costly. To address this issue, we proposed leveraging generative models to achieve few-shot image segmentation. METHODS: We trained a denoising diffusion probabilistic model (DDPM) on 480,407 pelvis radiographs to generate 256 âœ• 256 px synthetic images. The DDPM was conditioned on demographic and radiologic characteristics and was rigorously validated by domain experts and objective image quality metrics (Frechet inception distance [FID] and inception score [IS]). For the next step, three landmarks (greater trochanter [GT], lesser trochanter [LT], and obturator foramen [OF]) were annotated on 45 real-patient radiographs; 25 for training and 20 for testing. To extract features, each image was passed through the pre-trained DDPM at three timesteps and for each pass, features from specific blocks were extracted. The features were concatenated with the real image to form an image with 4225 channels. The feature-set was broken into random patches, which were fed to a U-Net. Dice Similarity Coefficient (DSC) was used to compare the performance with a vanilla U-Net trained on radiographs. RESULTS: Expert accuracy was 57.5 % in determining real versus generated images, while the model reached an FID = 7.2 and IS = 210. The segmentation UNet trained on the 20 feature-sets achieved a DSC of 0.90, 0.84, and 0.61 for OF, GT, and LT segmentation, respectively, which was at least 0.30 points higher than the naively trained model. CONCLUSION: We demonstrated the applicability of DDPMs as feature extractors, facilitating medical image segmentation with few annotated samples.

Application of Natural Language Processing in Total Joint Arthroplasty: Opportunities and Challenges.

Total joint arthroplasty is becoming one of the most common surgeries within the United States, creating an abundance of analyzable data to improve patient experience and outcomes. Unfortunately, a large majority of this data is concealed in electronic health records only accessible by manual extraction, which takes extensive time and resources. Natural language processing (NLP), a field within artificial intelligence, may offer a viable alternative to manual extraction. Using NLP, a researcher can analyze written and spoken data and extract data in an organized manner suitable for future research and clinical use. This article will first discuss common subtasks involved in an NLP pipeline, including data preparation, modeling, analysis, and external validation, followed by examples of NLP projects. Challenges and limitations of NLP will be discussed, closing with future directions of NLP projects, including large language models.

Educational Overview of the Concept and Application of Computer Vision in Arthroplasty.

Image data has grown exponentially as systems have increased their ability to collect and store it. Unfortunately, there are limits to human resources both in time and knowledge to fully interpret and manage that data. Computer Vision (CV) has grown in popularity as a discipline for better understanding visual data. Computer Vision has become a powerful tool for imaging analytics in orthopedic surgery, allowing computers to evaluate large volumes of image data with greater nuance than previously possible. Nevertheless, even with the growing number of uses in medicine, literature on the fundamentals of CV and its implementation is mainly oriented toward computer scientists rather than clinicians, rendering CV unapproachable for most orthopedic surgeons as a tool for clinical practice and research. The purpose of this article is to summarize and review the fundamental concepts of CV application for the orthopedic surgeon and musculoskeletal researcher.

An Overview of Machine Learning in Orthopedic Surgery: An Educational Paper.

The growth of artificial intelligence combined with the collection and storage of large amounts of data in the electronic medical record collection has created an opportunity for orthopedic research and translation into the clinical environment. Machine learning (ML) is a type of artificial intelligence tool well suited for processing the large amount of available data. Specific areas of ML frequently used by orthopedic surgeons performing total joint arthroplasty include tabular data analysis (spreadsheets), medical imaging processing, and natural language processing (extracting concepts from text). Previous studies have discussed models able to identify fractures in radiographs, identify implant type in radiographs, and determine the stage of osteoarthritis based on walking analysis. Despite the growing popularity of ML, there are limitations including its reliance on "good" data, potential for overfitting, long life cycle for creation, and ability to only perform one narrow task. This educational article will further discuss a general overview of ML, discussing these challenges and including examples of successfully published models.

Haemorrhagic cystitis: a review of management strategies and emerging treatments.

Haemorrhagic cystitis (HC) is characterised by persistent haematuria and lower urinary tract symptoms following radiotherapy or chemotherapy. Its pathogenesis is poorly understood but thought to be related to acrolein toxicity following chemotherapy or fibrosis/vascular remodelling after radiotherapy. There is no standard of care for patients with HC, although existing strategies including fulguration, hyperbaric oxygen therapy, botulinum toxin A, and other intravesical therapies have demonstrated short-term efficacy in cohort studies. Novel agents including liposomal tacrolimus are promising targets for further research. This review summarises the incidence and pathogenesis of HC as well as current evidence supporting its different management strategies.

Clinical Validation of an Adult-acquired Buried Penis Classification System Based on Standardized Evaluation of the Penis, Abdomen, and Scrotum.

OBJECTIVE: To clinically validate a previously developed adult-acquired buried penis (AABP) classification system that is based on a standardized preoperative physical examination that subtypes patients by their penile skin/escutcheon complex (P), abdominal pannus (A), and scrotal skin (S). METHODS: The Trauma and Urologic Reconstruction Network of Surgeons (TURNS) database was used to create an AABP cohort. Patients were retrospectively classified using the previously described PAS classification system. The frequency of subtypes, surgical methods utilized for AABP repair, and correlations between PAS classification and surgery subtypes were analyzed. RESULTS: The final cohort consisted of 101 patients from 10 institutions. Interrater reliability between two reviewers was excellent (κ = 0.95). The most common subtypes were P2c (contributory escutcheon+insufficient penile skin; 27%) and P2a (contributory escutcheon+sufficient penile skin; 21%) for penile subtypes, A0 (no pannus; 41%) and A1 (noncontributory pannus; 39%) for abdominal subtypes, and S0 (normal scrotal skin with preserved scrotal sulcus; 71%) for scrotal subtypes. AABP repair procedures included escutcheonectomy (n = 59, 55%), scrotoplasty (n = 51, 48%), split-thickness skin grafting (n = 50, 47%), penile skin excision (n = 47, 44%) and panniculectomy (n = 7, 7%). P, A, and S subtypes were strongly associated with specific AABP surgical techniques. CONCLUSION: The PAS classification schema adequately describes AABP heterogeneity, is reproducible among observers, and correlates well with AABP surgery types. Future work will focus on how PAS subtypes affect both surgical and patient-centered outcomes.

Deep learning approach for differentiating indeterminate adrenal masses using CT imaging.

PURPOSE: Distinguishing stage 1-2 adrenocortical carcinoma (ACC) and large, lipid poor adrenal adenoma (LPAA) via imaging is challenging due to overlapping imaging characteristics. This study investigated the ability of deep learning to distinguish ACC and LPAA on single time-point CT images. METHODS: Retrospective cohort study from 1994 to 2022. Imaging studies of patients with adrenal masses who had available adequate CT studies and histology as the reference standard by method of adrenal biopsy and/or adrenalectomy were included as well as four patients with LPAA determined by stability or regression on follow-up imaging. Forty-eight (48) subjects with pathology-proven, stage 1-2 ACC and 43 subjects with adrenal adenoma >3 cm in size demonstrating a mean non-contrast CT attenuation > 20 Hounsfield Units centrally were included. We used annotated single time-point contrast-enhanced CT images of these adrenal masses as input to a 3D Densenet121 model for classifying as ACC or LPAA with five-fold cross-validation. For each fold, two checkpoints were reported, highest accuracy with highest sensitivity (accuracy focused) and highest sensitivity with the highest accuracy (sensitivity focused). RESULTS: We trained a deep learning model (3D Densenet121) to predict ACC versus LPAA. The sensitivity-focused model achieved mean accuracy: 87.2% and mean sensitivity: 100%. The accuracy-focused model achieved mean accuracy: 91% and mean sensitivity: 96%. CONCLUSION: Deep learning demonstrates promising results distinguishing between ACC and large LPAA using single time-point CT images. Before being widely adopted in clinical practice, multicentric and external validation are needed.

Grade V renal trauma management: results from the multi-institutional genito-urinary trauma study.

PURPOSE: To investigate management trends for American Association for the Surgery of Trauma (AAST) grade V renal trauma with focus on non-operative management. METHODS: We used prospectively collected data as part of the Multi-institutional Genito-Urinary Trauma Study (MiGUTS). We included patients with grade V renal trauma according to the AAST Injury Scoring Scale 2018 update. All cases submitted by participating centers with radiology images available were independently reviewed to confirm renal trauma grade. Management was classified as expectant, conservative (minimally invasive, endoscopic or percutaneous procedures), or operative (renal-related surgery). RESULTS: Eighty patients were included, 25 of whom had complete imaging and had independent confirmation of AAST grade V renal trauma. Median age was 35 years (Interquartile range (IQR) 25-50) and 23 (92%) had blunt trauma. Ten patients (40%) were managed operatively with nephrectomy. Conservative management was used in nine patients (36%) of which six received angioembolization and three had a stent or drainage tube placed. Expectant management was followed in six (24%) patients. Transfusion requirements were progressively higher with groups requiring more aggressive treatment, and injury characteristics differed significantly across management groups in terms of hematoma size and laceration size. Vascular contrast extravasation was more likely in operatively managed patients though a statistically significant association was not found. CONCLUSION: Successful use of nonoperative management for grade V injuries is used for a substantial subset of patients. Lower transfusion requirement and less severe injury radiologic phenotype appear to be important characteristics delineating this group.

Shattered Kidney After Renal Trauma: Should It Be Classified As an American Association for the Surgery of Trauma Grade V Injury?

OBJECTIVE: To study the prevalence and management of shattered kidney and to evaluate if the new description of "loss of identifiable renal anatomy" in the 2018 American Association for the Surgery of Trauma (AAST) organ injury scale (OIS) would improve the ability to predict bleeding control interventions. METHODS: We used high-grade renal trauma data from 21 Level-1 trauma centers from 2013 to 2018. Initial CT scans were reviewed to identify shattered kidneys, defined as a kidney having ≥3 parenchymal fragments displaced by blood or fluid on cross-sectional imaging. We further categorized patients with shattered kidney in two models based on loss of identifiable renal parenchymal anatomy and presence or absence of vascular contrast extravasation (VCE). Bleeding interventions were compared between the groups. RESULTS: From 861 high-grade renal trauma patients, 41 (4.8%) had shattered kidney injury. 25 (61%) underwent a bleeding control intervention including 18 (43.9%) nephrectomies and 11 (26.8%) angioembolizations. 18 (41%) had shattered kidney with "loss of identifiable parenchymal renal anatomy" per 2018 AAST OIS (model-1). 28 (68.3%) had concurrent VCE (model-2). Model-2 had a statistically significant improvement in area under the curve over model-1 in predicting bleeding interventions (0.75 vs 0.72; P = .01). CONCLUSION: Shattered kidney is associated with high rates of active bleeding, urinary extravasation, and interventions including nephrectomy. The definition of shattered kidney is vague and subjective and our definition might be simpler and more reproducible. Loss of identifiable renal anatomy per the 2018 AAST OIS did not provide better distinction for bleeding control interventions over presence of VCE.