Background
The Prostate Imaging Reporting and Data System (PI‐RADS) provides guidelines for risk stratification of lesions detected on multiparametric MRI (mpMRI) of the prostate but suffers from high intra/interreader variability.
Purpose
To develop an artificial intelligence (AI) solution for PI‐RADS classification and compare its performance with an expert radiologist using targeted biopsy results.
Study Type
Retrospective study including data from our institution and the publicly available ProstateX dataset.
Population
In all, 687 patients who underwent mpMRI of the prostate and had one or more detectable lesions (PI‐RADS score >1) according to PI‐RADSv2.
Field Strength/Sequence
T2‐weighted, diffusion‐weighted imaging (DWI; five evenly spaced b values between b = 0–750 s/mm2) for apparent diffusion coefficient (ADC) mapping, high b‐value DWI (b = 1500 or 2000 s/mm2), and dynamic contrast‐enhanced T1‐weighted series were obtained at 3.0T.
Assessment
PI‐RADS lesions were segmented by a radiologist. Bounding boxes around the T2/ADC/high‐b value segmentations were stacked and saved as JPEGs. These images were used to train a convolutional neural network (CNN). The PI‐RADS scores obtained by the CNN were compared with radiologist scores. The cancer detection rate was measured from a subset of patients who underwent biopsy.
Statistical Tests
Agreement between the AI and the radiologist‐driven PI‐RADS scores was assessed using a kappa score, and differences between categorical variables were assessed with a Wald test.
Results
For the 1034 detection lesions, the kappa score for the AI system vs. the expert radiologist was moderate, at 0.40. However, there was no significant difference in the rates of detection of clinically significant cancer for any PI‐RADS score in 86 patients undergoing targeted biopsy (P = 0.4–0.6).
Data Conclusion
We developed an AI system for assignment of a PI‐RADS score on segmented lesions on mpMRI with moderate agreement with an expert radiologist and a similar ability to detect clinically significant cancer.
Level of Evidence
4
Technical Efficacy Stage
2
Introduction:The aim of this study was to perform a quantitative assessment of the prostate anatomy with a focus on the relation of prostatic urethral anatomic variation to urinary symptoms.Methods: This retrospective study involved patients undergoing magnetic resonance imaging for prostate cancer who were also assessed for lower urinary tract symptoms. Volumetric segmentations were utilized to derive the in vivo prostatic urethral length and urethral trajectory in coronal and sagittal planes using a piece-wise cubic spline function to derive the angle of the urethra within the prostate. Association of anatomical factors with urinary symptoms was evaluated using ordinal univariable and multivariable logistic regression with IPSS score cutoffs of ≤7, 8-19, and >20 to define mild, moderate, and severe symptoms, respectively.Results: A total of 423 patients were included. On univariable analysis, whole prostate volume, transition zone volume, prostatic urethral length, urethral angle, and retrourethral volume were all significantly associated with worse urinary symptoms. On multivariable analysis prostatic urethral length was associated with urinary symptoms with a normalized odds ratio of 1.5 (95% confidence interval 1.0-2.2, p = 0.04). In a subset analysis of patients on alpha blockers, maximal urethral angle, transition zone volume as well as urethral length were all associated with worse urinary symptoms.
Conclusion:Multiple parameters were associated with worse urinary symptoms on univariable analysis, but only prostatic urethral length was associated with worse urinary symptoms on
Testicular cancer represents the most common solid tumors in young men and malignant germ cell tumors constitute the majority of these masses. Orchiectomy is the treatment of choice for intratesticular masses. However, it is important to recognize those benign conditions for which orchiectomy is unwarranted because approximately 5-10% of all testicular masses are postoperatively identified as benign testicular lesions. Ultrasound features of solid scrotal tumors are often non-specific. US, however, identifies the lesion in the largest number of cases. Here, we evaluate the histopatologic features which determine echogenicity, echotexture, stiffness, and vascularity of different testicular masses.METHODS: 98 testicular lesions investigated with colour Doppler US for which histological specimens were available for review were enclosed. 47 lesions had also CEUS, 34 elastography. Imaging features were correlated with histological characteristics.RESULTS: Most testicular cancers were hypervascular at colour Doppler interrogation and with increased consistency at elastography. 6/22 hypovascular lesions with small vessels at colour Doppler interrogation were hypervascular at CEUS. Seminomas presented with lower echogenicity compared to other cancers, mixed tumors were heterogeneous. Irrespective of histotype, abundance of stroma was the main factor determining increased echogenicity. Other factors were hyalization, interleaved necrotic areas and tumor nests, athrophic changes of the surrounding parenchyma. Heterogeous appearance was observed in presence of different histotypes, calcifications, necrotic or fibrotic areas. Lesions were avascular at colour Doppler interrogation if vessels were lacking or were very small. In the latter case, they were vascularized at CEUS. Lesions with abundant stroma were hard at elastography, while necrotic changes caused soft or mixed appearance CONCLUSIONS: The different histological features of testicular lesions determine their sonographic appearance. US is highly sensitive for detection of testicular lesions, but specificity is low.The different US modes, if taken individually, are non-specific but if used together improve lesion characterization.Preoperative imaging and correct characterization of focal testicular lesions play a quite important role in guiding the surgical approach to these patients allowing the identification of testicular lesions amenable to treatment with testicular-sparing surgery
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.