Transperineal versus Transrectal MRI/TRUS fusion-guided prostate biopsy in a large, ethnically diverse, and multiracial cohort.

PURPOSE: To compare transperineal (TP) vs transrectal (TR) magnetic resonance imaging (MRI) and transrectal ultrasound (TRUS) fusion-guided prostate biopsy (PBx) in a large, ethnically diverse and multiracial cohort. MATERIALS AND METHODS: Consecutive patients who underwent multiparametric (mp) MRI followed by TP or TR TRUS-fusion guided PBx, were identified from a prospective database (IRB #HS-13-00663). All patients underwent mpMRI followed by 12-14 core systematic PBx. A minimum of two additional target-biopsy cores were taken per PIRADS≥3 lesion. The endpoint was the detection of clinically significant prostate cancer (CSPCa; Grade Group, GG≥2). Statistical significance was defined as p<0.05. RESULTS: A total of 1491 patients met inclusion criteria, with 480 undergoing TP and 1011 TR PBx. Overall, 11% of patients were Asians, 5% African Americans, 14% Hispanic, 14% Others, and 56% White, similar between TP and TR (p=0.4). For PIRADS 3-5, the TP PBx CSPCa detection was significantly higher (61% vs 54%, p=0.03) than TR PBx, but not for PIRADS 1-2 (13% vs 13%, p=1.0). After adjusting for confounders on multivariable analysis, Black race, but not the PBx approach (TP vs TR), was an independent predictor of CSPCa detection. The median maximum cancer core length (11 vs 8mm; p<0.001) and percent (80% vs 60%; p<0.001) were greater for TP PBx even after adjusting for confounders. CONCLUSIONS: In a large and diverse cohort, Black race, but not the biopsy approach, was an independent predictor for CSPCa detection. TP and TR PBx yielded similar CSPCa detection rates; however the TP PBx was histologically more informative.

The learning curve for transperineal MRI/TRUS fusion prostate biopsy: A prospective evaluation of a stepwise approach.

OBJECTIVE: To evaluate the learning curve of a transperineal (TP) magnetic resonance imaging (MRI) and transrectal ultrasound (TRUS) fusion prostate biopsy (PBx). MATERIALS AND METHODS: Consecutive patients undergoing MRI followed by TP PBx from May/2017 to January/2023, were prospectively enrolled (IRB# HS-13-00663). All participants underwent MRI followed by 12 to 14 core systematic PBx (SB), with at least 2 additional targeted biopsy (TB) cores per PIRADS ≥3. The biopsies were performed transperineally using an organ tracking image-fusion system. The cohort was divided into chronological quintiles. An inflection point analysis was performed to determine proficiency. Operative time was defined from insertion to removal of the TRUS probe from the patient's rectum. Grade Group ≥2 defined clinically significant prostate cancer (CSPCa). Statistically significant if P < 0.05. RESULTS: A total of 370 patients were included and divided into quintiles of 74 patients. MRI findings and PIRADS distribution were similar between quintiles (P = 0.08). The CSPCa detection with SB+TB was consistent across quintiles: PIRADS 1 and 2 (range, 0%-18%; P = 0.25); PIRADS 3 to 5 (range, 46%-70%; P = 0.12). The CSPCa detection on PIRADS 3 to 5 TB alone, for quintiles 1 to 5, was respectively 44%, 58%, 66%, 41%, and 53% (P = 0.08). The median operative time significantly decreased for PIRADS 1 and 2 (33 min to 13 min; P < 0.01) and PIRADS 3 to 5 (48 min to 19 min; P < 0.01), reaching a plateau after 156 cases. Complications were not significantly different across quintiles (range, 0-5.4%; P = 0.3). CONCLUSIONS: The CSPCa detection remained consistently satisfactory throughout the learning curve of the Transperineal MRI/TRUS fusion prostate biopsy. However, the operative time significantly decreased with proficiency achieved after 156 cases.

Perioperative and extended outcomes of patients undergoing parastomal hernia repair following cystectomy and ileal conduit.

PURPOSE: To report perioperative and long-term postoperative outcomes of cystectomy patients with ileal conduit (IC) urinary diversion undergoing parastomal hernia (PSH) repair. METHOD: We reviewed patients who underwent cystectomy and IC diversion between 2003 and 2022 in our center. Baseline variables, including surgical approach of PSH repair and repair technique, were captured. Multivariable Cox regressionanalysis was performed to test for the associations between different variables and PSH recurrence. RESULTS: Thirty-six patients with a median (IQR) age of 79 (73-82) years were included. The median time between cystectomy and PSH repair was 30 (14-49) months. Most PSH repairs (32/36, 89%) were performed electively, while 4 were due to small bowel obstruction. Hernia repairs were performed through open (n=25), robotic (10), and laparoscopic approaches (1). Surgical techniques included direct repair with mesh (20), direct repair without mesh (4), stoma relocation with mesh (5), and stomarelocation without mesh (7). The 90-day complication rate was 28%. In a median follow-up of 24 (7-47) months, 17 patients (47%) had a recurrence. The median time to recurrence was 9 (7-24) months. On multivariable analysis, 90-day complication following PSH repair was associated with an increased risk of recurrence. CONCLUSIONS: In this report of one of the largest series of PSH repair in the Urology literature, 47% of patients had a recurrence following hernia repair with a median follow-up time of 2 years. There was no significant difference in recurrence rates when comparing repair technique or the use of open or minimally invasive approaches.

PCa-RadHop: A transparent and lightweight feed-forward method for clinically significant prostate cancer segmentation.

Prostate Cancer is one of the most frequently occurring cancers in men, with a low survival rate if not early diagnosed. PI-RADS reading has a high false positive rate, thus increasing the diagnostic incurred costs and patient discomfort. Deep learning (DL) models achieve a high segmentation performance, although require a large model size and complexity. Also, DL models lack of feature interpretability and are perceived as "black-boxes" in the medical field. PCa-RadHop pipeline is proposed in this work, aiming to provide a more transparent feature extraction process using a linear model. It adopts the recently introduced Green Learning (GL) paradigm, which offers a small model size and low complexity. PCa-RadHop consists of two stages: Stage-1 extracts data-driven radiomics features from the bi-parametric Magnetic Resonance Imaging (bp-MRI) input and predicts an initial heatmap. To reduce the false positive rate, a subsequent stage-2 is introduced to refine the predictions by including more contextual information and radiomics features from each already detected Region of Interest (ROI). Experiments on the largest publicly available dataset, PI-CAI, show a competitive performance standing of the proposed method among other deep DL models, achieving an area under the curve (AUC) of 0.807 among a cohort of 1,000 patients. Moreover, PCa-RadHop maintains orders of magnitude smaller model size and complexity.

A nomogram to predict the absence of clinically significant prostate cancer in males with negative MRI

ABSTRACT Purpose: To create a nomogram to predict the absence of clinically significant prostate cancer (CSPCa) in males with non-suspicion multiparametric magnetic resonance imaging (mpMRI) undergoing prostate biopsy (PBx). Materials and Methods: We identified consecutive patients who underwent 3T mpMRI followed by PBx for suspicion of PCa or surveillance follow-up. All patients had Prostate Imaging Reporting and Data System score 1-2 (negative mpMRI). CSPCa was defined as Grade Group ≥2. Multivariate logistic regression analysis was performed via backward elimination. Discrimination was evaluated with area under the receiver operating characteristic (AUROC). Internal validation with 1,000x bootstrapping for estimating the optimism corrected AUROC. Results: Total 327 patients met inclusion criteria. The median (IQR) age and PSA density (PSAD) were 64 years (58-70) and 0.10 ng/mL2 (0.07-0.15), respectively. Biopsy history was as follows: 117 (36%) males were PBx-naive, 130 (40%) had previous negative PBx and 80 (24%) had previous positive PBx. The majority were White (65%); 6% of males self-reported Black. Overall, 44 (13%) patients were diagnosed with CSPCa on PBx. Black race, history of previous negative PBx and PSAD ≥0.15ng/mL2 were independent predictors for CSPCa on PBx and were included in the nomogram. The AUROC of the nomogram was 0.78 and the optimism corrected AUROC was 0.75. Conclusions: Our nomogram facilitates evaluating individual probability of CSPCa on PBx in males with PIRADS 1-2 mpMRI and may be used to identify those in whom PBx may be safely avoided. Black males have increased risk of CSPCa on PBx, even in the setting of PIRADS 1-2 mpMRI

A Fully Automated Artificial Intelligence System to Assist Pathologists' Diagnosis to Predict Histologically High-grade Urothelial Carcinoma from Digitized Urine Cytology Slides Using Deep Learning.

BACKGROUND: Urine cytology, although a useful screening method for urothelial carcinoma, lacks sensitivity. As an emerging technology, artificial intelligence (AI) improved image analysis accuracy significantly. OBJECTIVE: To develop a fully automated AI system to assist pathologists in the histological prediction of high-grade urothelial carcinoma (HGUC) from digitized urine cytology slides. DESIGN, SETTING, AND PARTICIPANTS: We digitized 535 consecutive urine cytology slides for AI use. Among these slides, 181 were used for AI development, 39 were used as AI test data to identify HGUC by cell-level classification, and 315 were used as AI test data for slide-level classification. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Out of the 315 slides, 171 were collected immediately prior to bladder biopsy or transurethral resection of bladder tumor, and then outcomes were compared with the histological presence of HGUC in the surgical specimen. The primary aim was to compare AI prediction of the histological presence of HGUC with the pathologist's histological diagnosis of HGUC. Secondary aims were to compare the time required for AI evaluation and concordance between the AI's classification and pathologist's cytology diagnosis. RESULTS AND LIMITATIONS: The AI capability for predicting the histological presence of HGUC was 0.78 for the area under the curve. Comparing the AI predictive performance with pathologists' diagnosis, the AI sensitivity of 63% for histological HGUC prediction was superior to a pathologists' cytology sensitivity of 46% (p = 0.0037). On the contrary, there was no significant difference between the AI specificity of 83% and pathologists' specificity of 89% (p = 0.13), and AI accuracy of 74% and pathologists' accuracy of 68% (p = 0.08). The time required for AI evaluation was 139 s. With respect to the concordance between the AI prediction and pathologist's cytology diagnosis, the accuracy was 86%. Agreements with positive and negative findings were 92% and 84%, respectively. CONCLUSIONS: We developed a fully automated AI system to assist pathologists' histological diagnosis of HGUC using digitized slides. This AI system showed significantly higher sensitivity than a board-certified cytopathologist and may assist pathologists in making urine cytology diagnoses, reducing their workload. PATIENT SUMMARY: In this study, we present a deep learning-based artificial intelligence (AI) system that classifies urine cytology slides according to the Paris system. An automated AI system was developed and validated with 535 consecutive urine cytology slides. The AI predicted histological high-grade urothelial carcinoma from digitized urine cytology slides with superior sensitivity than pathologists, while maintaining comparable specificity and accuracy.

A nomogram to predict the absence of clinically significant prostate cancer in males with negative MRI

PURPOSE: To create a nomogram to predict the absence of clinically significant prostate cancer (CSPCa) in males with non-suspicion multiparametric magnetic resonance imaging (mpMRI) undergoing prostate biopsy (PBx). MATERIALS AND METHODS: We identified consecutive patients who underwent 3T mpMRI followed by PBx for suspicion of PCa or surveillance follow-up. All patients had Prostate Imaging Reporting and Data System score 1-2 (negative mpMRI). CSPCa was defined as Grade Group ≥2. Multivariate logistic regression analysis was performed via backward elimination. Discrimination was evaluated with area under the receiver operating characteristic (AUROC). Internal validation with 1,000x bootstrapping for estimating the optimism corrected AUROC. RESULTS: Total 327 patients met inclusion criteria. The median (IQR) age and PSA density (PSAD) were 64 years (58-70) and 0.10 ng/mL2 (0.07-0.15), respectively. Biopsy history was as follows: 117 (36%) males were PBx-naive, 130 (40%) had previous negative PBx and 80 (24%) had previous positive PBx. The majority were White (65%); 6% of males self-reported Black. Overall, 44 (13%) patients were diagnosed with CSPCa on PBx. Black race, history of previous negative PBx and PSAD ≥0.15ng/mL2 were independent predictors for CSPCa on PBx and were included in the nomogram. The AUROC of the nomogram was 0.78 and the optimism corrected AUROC was 0.75. CONCLUSIONS: Our nomogram facilitates evaluating individual probability of CSPCa on PBx in males with PIRADS 1-2 mpMRI and may be used to identify those in whom PBx may be safely avoided. Black males have increased risk of CSPCa on PBx, even in the setting of PIRADS 1-2 mpMRI.

Transperineal vs transrectal magnetic resonance and ultrasound image fusion prostate biopsy: a pair-matched comparison.

The objective of this study was to compare transperineal (TP) versus transrectal (TR) magnetic resonance imaging (MRI) and transrectal ultrasound (TRUS) fusion prostate biopsy (PBx). Consecutive men who underwent prostate MRI followed by a systematic biopsy. Additional target biopsies were performed from Prostate Imaging Reporting & Data System (PIRADS) 3-5 lesions. Men who underwent TP PBx were matched 1:2 with a synchronous cohort undergoing TR PBx by PSA, Prostate volume (PV) and PIRADS score. Endpoint of the study was the detection of clinically significant prostate cancer (CSPCa; Grade Group ≥ 2). Univariate and multivariable analyses were performed. Results were considered statistically significant if p < 0.05. Overall, 504 patients met the inclusion criteria. A total of 168 TP PBx were pair-matched to 336 TR PBx patients. Baseline demographics and imaging characteristics were similar between the groups. Per patient, the CSPCa detection was 2.1% vs 6.3% (p = 0.4) for PIRADS 1-2, and 59% vs 60% (p = 0.9) for PIRADS 3-5, on TP vs TR PBx, respectively. Per lesion, the CSPCa detection for PIRADS 3 (21% vs 16%; p = 0.4), PIRADS 4 (51% vs 44%; p = 0.8) and PIRADS 5 (76% vs 84%; p = 0.3) was similar for TP vs TR PBx, respectively. However, the TP PBx showed a longer maximum cancer core length (11 vs 9 mm; p = 0.02) and higher cancer core involvement (83% vs 65%; p < 0.001) than TR PBx. Independent predictors for CSPCa detection were age, PSA, PV, abnormal digital rectal examination findings, and PIRADS 3-5. Our study demonstrated transperineal MRI/TRUS fusion PBx provides similar CSPCa detection, with larger prostate cancer core length and percent of core involvement, than transrectal PBx.

Patient-reported Satisfaction and Regret Following Focal Therapy for Prostate Cancer: A Prospective Multicenter Evaluation.

Background: Several reports are available regarding the treatment decision regret of patients receiving conventional treatments for localized prostate cancer (PCa); yet data on patients undergoing focal therapy (FT) are sparse. Objective: To evaluate the treatment decision satisfaction and regret among patients who underwent FT for PCa with high-intensity focused ultrasound (HIFU) or cryoablation (CRYO). Design setting and participants: We identified consecutive patients who underwent HIFU or CRYO FT as the primary treatment for localized PCa at three US institutions. A survey with validated questionnaires, including the five-question Decision Regret Scale (DRS), International Prostate Symptom Score (IPSS), and International Index of Erectile Function (IIEF-5), was mailed to the patients. The regret score was calculated based on the five items of the DRS, and regret was defined as a DRS score of >25. Outcome measurements and statistical analysis: Multivariable logistic regression models were applied to assess the predictors of treatment decision regret. Results and limitations: Of 236 patients, 143 (61%) responded to the survey. Baseline characteristics were similar between responders and nonresponders. During a median (interquartile range) follow-up of 43 (26-68) mo, the treatment decision regret rate was 19.6%. On a multivariable analysis, higher prostate-specific antigen (PSA) at nadir after FT (odds ratio [OR] 1.48, 95% confidence interval [CI] 1.1-2, p = 0.009), presence of PCa on follow-up biopsy (OR 3.98, 95% CI 1.5-10.6, p = 0.006), higher post-FT IPSS (OR 1.18, 95% CI 1.01-1.37, p = 0.03), and newly diagnosed impotence (OR 6.67, 95% CI 1.57-27, p = 0.03) were independent predictors of treatment regret. The type of energy treatment (HIFU/CRYO) was not a predictor of regret/satisfaction. Limitations include retrospective abstraction. Conclusions: FT for localized PCa is well accepted by the patients, with a low regret rate. Higher PSA at nadir, presence of cancer on follow-up biopsy, bothersome postoperative urinary symptoms, and impotence after FT were independent predictors of treatment decision regret. Patient summary: In this report, we looked at the factors affecting satisfaction and regret in patients with prostate cancer undergoing focal therapy. We found that focal therapy is well accepted by the patients, while presence of cancer on follow-up biopsy as well as bothersome urinary symptoms and sexual dysfunction can predict treatment decision regret.

Is Artificial Intelligence Replacing Our Radiology Stars? Not Yet!

Artificial intelligence (AI) is here to stay and will change health care as we know it. The availability of big data and the increasing numbers of AI algorithms approved by the US Food and Drug Administration together will help in improving the quality of care for patients and in overcoming human fatigue barriers. In oncology practice, patients and providers rely on the interpretation of radiologists when making clinical decisions; however, there is considerable variability among readers, and in particular for prostate imaging. AI represents an emerging solution to this problem, for which it can provide a much-needed form of standardization. The diagnostic performance of AI alone in comparison to a combination of an AI framework and radiologist assessment for evaluation of prostate imaging has yet to be explored. Here, we compare the performance of radiologists alone versus a combination of radiologists aided by a modern computer-aided diagnosis (CAD) AI system. We show that the radiologist-CAD combination demonstrates superior sensitivity and specificity in comparison to both radiologists alone and AI alone. Our findings demonstrate that a radiologist + AI combination could perform best for detection of prostate cancer lesions. A hybrid technology-human system could leverage the benefits of AI in improving radiologist performance while also reducing physician workload, minimizing burnout, and enhancing the quality of patient care. Patient summary: Our report demonstrates the potential of artificial intelligence (AI) for improving the interpretation of prostate scans. A combination of AI and evaluation by a radiologist has the best performance in determining the severity of prostate cancer. A hybrid system that uses both AI and radiologists could maximize the quality of care for patients while reducing physician workload and burnout.

Gender Disparities Among Editorial Boards of International Urology Journals.

BACKGROUND: Gender composition among surgical academic leadership, including academic medical journals, disproportionately favors men and may inadvertently introduce a bias. An understanding of the factors associated with gender representation among urologic journals may aid in prioritizing an equitable balance. OBJECTIVE: To evaluate female representation on editorial boards of pre-eminent international urologic journals. DESIGN, SETTING, AND PARTICIPANTS: The names and position descriptions of urologic journal leadership appointees were collected in October 2021. Gender was assessed using gender-api.com or through personal title, as available. Journal characteristics were summarized using SCImago, a bibliometric indicator database extracted from Scopus journal data. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: A multivariate logistic regression analysis was performed to describe associations between SCImago Journal Rank (SJR) quartile and geographic region with female gender representation. Quartile 1 (Q1) was considered the top quartile and Q4 the bottom quartile concordant with journal impact factor. RESULTS AND LIMITATIONS: A total of 105 urology-focused journals were identified with 5989 total editorial board members, including 877 (14.6%) female, 5112 (85.4%) male, and two nonbinary persons. Female representation differed significantly by journal leadership position, SJR quartile, and geographic region. On the multivariate analysis of overall female representation, Q1 journals had higher odds of female representation than Q2 and Q3 journals, and had no significant difference from Q4 journals. Additionally, compared with Western Europe, North American journals had 78% higher odds while Asiatic journals had 50% lower odds of female representation. This study is limited by the inability to account for outside factors that lead to invitation or acceptance of journal leadership positions. CONCLUSIONS: Contemporary female leadership at urology journals is about six times less common than male leadership across all journals, although trends in their proportion were noted when assessed by journal quartile and region. Addressing this gender imbalance represents an important step toward achieving gender equity in the field of urology. PATIENT SUMMARY: In this study, we looked at the gender balance of academic journal leaders who serve as gatekeepers for sharing urologic research with the public. We found that the most prestigious journals and those in western countries tended to have the highest female representation. We hope that these findings help the academic community recognize and improve gender representation.

Microwave focal therapy of prostate cancer: a non-clinical study and exploratory clinical trial.

OBJECTIVE: To examine the safety and efficacy of microwave tissue coagulation (MTC) for prostate cancer and assess its use in lesion-targeted focal therapy in a non-clinical study and a clinical phase II trial. METHODS: In the non-clinical study using Microtaze® -AFM-712 (Alfresa Pharma Corporation, Osaka, Japan) with an MTC needle, MTC was performed using a transperineal approach to targeted canine prostatic tissue under real-time ultrasonography guidance. Using various MTC output and irradiation time combinations, the targeted and surrounding tissues (rectum, bladder and fat) were examined to confirm the extent of coagulative necrosis or potential cell death, and to compare intra-operative ultrasonography and pathology findings. The exploratory clinical trial was conducted to examine the safety and efficacy of MTC. Five selected patients underwent transperineal MTC to clinically single lesion magnetic resonance imaging (MRI)-visible lesions with Gleason score 3 + 4 or 4 + 4. Prostate-specific antigen (PSA), MRI and Expanded Prostate Cancer Index Composite questionnaire findings were compared before and 6 months after surgery. RESULTS: The region of coagulative necrosis was predictable by monitoring of ultrasonically visible vaporization; thus, by placing the MTC needle at a certain distance, we were able to perform a safe procedure without adverse events affecting the surrounding organs. Based on the non-clinical study, which used various combinations of output and irradiation time, MTC with 30-W output for 60-s irradiation was selected for the prostate. Based on the predictable necrosis, the therapeutic plan (where to place the MTC needle to achieve complete ablation of the target and how many sessions) was strictly determined per patient. There were no serious adverse events in any patient and only temporary urinary symptoms related to MTC therapy were observed. Furthermore, post-treatment satisfaction was very high. All preoperative MRI-visible lesions disappeared, and PSA decreased by 55% 6 months after surgery. CONCLUSION: Microwave tissue coagulation may be an option for lesion-targeted focal therapy for prostate cancer.

Artificial intelligence trained with integration of multiparametric MR-US imaging data and fusion biopsy trajectory-proven pathology data for 3D prediction of prostate cancer: A proof-of-concept study.

BACKGROUND: We aimed to develop an artificial intelligence (AI) algorithm that predicts the volume and location of clinically significant cancer (CSCa) using convolutional neural network (CNN) trained with integration of multiparametric MR-US image data and MRI-US fusion prostate biopsy (MRI-US PBx) trajectory-proven pathology data. METHODS: Twenty consecutive patients prospectively underwent MRI-US PBx, followed by robot-assisted radical prostatectomy (RARP). The AI algorithm was trained with the integration of MR-US image data with a MRI-US PBx trajectory-proven pathology. The relationship with the 3D-cancer-mapping of RARP specimens was compared between AI system-suggested 3D-CSCa mapping and an experienced radiologist's suggested 3D-CSCa mapping on MRI alone according to the Prostate Imaging Reporting and Data System (PI-RADS) version 2. The characteristics of detected and undetected tumors at AI were compared in 22,968 image data. The relationships between CSCa volumes and volumes predicted by AI as well as the radiologist's reading based on PI-RADS were analyzed. RESULTS: The concordance of the CSCa center with that in RARP specimens was significantly higher in the AI prediction than the radiologist' reading (83% vs. 54%, p = 0.036). CSCa volumes predicted with AI were more accurate (r = 0.90, p < 0.001) than the radiologist's reading. The limitations include that the elastic fusion technology has its own registration error. CONCLUSIONS: We presented a novel pilot AI algorithm for 3D prediction of PCa. AI was trained by integration of multiparametric MR-US image data and fusion biopsy trajectory-proven pathology data. This deep learning AI model may more precisely predict the 3D mapping of CSCa in its volume and center location than a radiologist's reading based on PI-RADS version 2, and has potential in the planning of focal therapy.

Multiparametric ultrasound of prostate: role in prostate cancer diagnosis.

Recent advances in ultrasonography (US) technology established modalities, such as Doppler-US, HistoScanning, contrast-enhanced ultrasonography (CEUS), elastography, and micro-ultrasound. The early results of these US modalities have been promising, although there are limitations including the need for specialized equipment, inconsistent results, lack of standardizations, and external validation. In this review, we identified studies evaluating multiparametric ultrasonography (mpUS), the combination of multiple US modalities, for prostate cancer (PCa) diagnosis. In the past 5 years, a growing number of studies have shown that use of mpUS resulted in high PCa and clinically significant prostate cancer (CSPCa) detection performance using radical prostatectomy histology as the reference standard. Recent studies have demonstrated the role mpUS in improving detection of CSPCa and guidance for prostate biopsy and therapy. Furthermore, some aspects including lower costs, real-time imaging, applicability for some patients who have contraindication for magnetic resonance imaging (MRI) and availability in the office setting are clear advantages of mpUS. Interobserver agreement of mpUS was overall low; however, this limitation can be improved using standardized and objective evaluation systems such as the machine learning model. Whether mpUS outperforms MRI is unclear. Multicenter randomized controlled trials directly comparing mpUS and multiparametric MRI are warranted.

Urine cell image recognition using a deep-learning model for an automated slide evaluation system.

OBJECTIVES: To develop a classification system for urine cytology with artificial intelligence (AI) using a convolutional neural network algorithm that classifies urine cell images as negative (benign) or positive (atypical or malignant). PATIENTS AND METHODS: We collected 195 urine cytology slides from consecutive patients with a histologically confirmed diagnosis of urothelial cancer (between January 2016 and December 2017). Two certified cytotechnologists independently evaluated and labelled each slide; 4637 cell images with concordant diagnoses were selected, including 3128 benign cells (negative), 398 atypical cells, and 1111 cells that were malignant or suspicious for malignancy (positive). This pathologically confirmed labelled dataset was used to represent the ground truth for AI training/validation/testing. Customized CutMix (CircleCut) and Refined Data Augmentation were used for image processing. The model architecture included EfficientNet B6 and Arcface. We used 80% of the data for training and validation (4:1 ratio) and 20% for testing. Model performance was evaluated with fivefold cross-validation. A receiver-operating characteristic (ROC) analysis was used to evaluate the binary classification model. Bayesian posterior probabilities for the AI performance measure (Y) and cytotechnologist performance measure (X) were compared. RESULTS: The area under the ROC curve was 0.99 (95% confidence interval [CI] 0.98-0.99), the highest accuracy was 95% (95% CI 94-97), sensitivity was 97% (95% CI 95-99), and specificity was 95% (95% CI 93-97). The accuracy of AI surpassed the highest level of cytotechnologists for the binary classification [Pr(Y > X) = 0.95]. AI achieved >90% accuracy for all cell subtypes. In the subgroup analysis based on the clinicopathological characteristics of patients who provided the test cells, the accuracy of AI ranged between 89% and 97%. CONCLUSION: Our novel AI classification system for urine cytology successfully classified all cell subtypes with an accuracy of higher than 90%, and achieved diagnostic accuracy of malignancy superior to the highest level achieved by cytotechnologists.