Search

Objective: To evaluate the ability to detect clinically significant prostate cancer (PCa) using a novel electromagnetically (EM) tracked transperineal magnetic resonance imaging (MRI)/ultrasonography (US) fusion‐guided targeted biopsy (transperineal TBx) platform and the impact of inter‐reader variability on cancer detection. Materials and Methods: A total of 176 patients with suspicious lesions detected on multiparametric MRI (mpMRI) underwent a systematic modified Barzel template biopsy (12‐core) transperineal biopsy (transperineal SBx) and transperineal TBx with EM tracking (UroNav; Philips Healthcare, Best, the Netherlands) in the same setting. Cancer detection rates (CDRs) were stratified by Prostate Imaging Reporting and Data System (PI‐RADS) v2 scores and compared with Fisher's exact test. Area under the curve (AUC) was calculated for prostate‐specific antigen (PSA), PSA density (PSAD), PI‐RADS score, and subgroup analysis of individual readers' PI‐RADS scores with respect to overall CDR and clinically significant CDR. Results: The overall CDR was 76.7% (135/176), of which 76.3% (103/135) was clinically significant PCa. Among the 135 patients with PCa, transperineal TBx detected 90.4% of cases (122/135), either alone or in combination with transperineal SBx. The remaining 9.6% of cases (13/135) missed by transperineal TBx were diagnosed by transperineal SBx alone, of which three were clinically significant. Conversely, transperineal SBx missed 14% of cases (19/135), 14 of which were clinically significant PCa. Sensitivities for transperineal TBx and transperineal SBx were 90.4% and 85.9%, respectively. On a per‐lesion basis, PI‐RADS score (AUC 0.74) outperformed both PSA (AUC 0.59) and PSAD (AUC 0.63) in discriminating clinically significant from non‐clinically significant PCa on transperineal TBx. Although not formally statistically tested, AUCs amongst different mpMRI readers appeared to display considerable variability. There were no adverse events, including sepsis. Conclusions: Electromagnetically tracked transperineal TBx of MRI‐visible lesions enhanced the ability of transperineal SBx to detect PCa, with greater sensitivity for clinically significant disease. These findings suggest transperineal TBx is a safe, alternative fusion platform for patients with a suspicious lesion on prostate MRI. The assessment of inter‐reader variability, in conjunction with prediction of clinically significant PCa and CDR, is an important first step for quality control in implementing an MRI‐based screening programme. [ABSTRACT FROM AUTHOR]

Purpose: To train a deep learning classification algorithm to predict chest radiograph severity scores and clinical outcomes in patients with coronavirus disease 2019 (COVID-19)., Materials and Methods: In this retrospective cohort study, patients aged 21-50 years who presented to the emergency department (ED) of a multicenter urban health system from March 10 to 26, 2020, with COVID-19 confirmation at real-time reverse-transcription polymerase chain reaction screening were identified. The initial chest radiographs, clinical variables, and outcomes, including admission, intubation, and survival, were collected within 30 days ( n = 338; median age, 39 years; 210 men). Two fellowship-trained cardiothoracic radiologists examined chest radiographs for opacities and assigned a clinically validated severity score. A deep learning algorithm was trained to predict outcomes on a holdout test set composed of patients with confirmed COVID-19 who presented between March 27 and 29, 2020 ( n = 161; median age, 60 years; 98 men) for both younger (age range, 21-50 years; n = 51) and older (age >50 years, n = 110) populations. Bootstrapping was used to compute CIs., Results: The model trained on the chest radiograph severity score produced the following areas under the receiver operating characteristic curves (AUCs): 0.80 (95% CI: 0.73, 0.88) for the chest radiograph severity score, 0.76 (95% CI: 0.68, 0.84) for admission, 0.66 (95% CI: 0.56, 0.75) for intubation, and 0.59 (95% CI: 0.49, 0.69) for death. The model trained on clinical variables produced an AUC of 0.64 (95% CI: 0.55, 0.73) for intubation and an AUC of 0.59 (95% CI: 0.50, 0.68) for death. Combining chest radiography and clinical variables increased the AUC of intubation and death to 0.88 (95% CI: 0.79, 0.96) and 0.82 (95% CI: 0.72, 0.91), respectively., Conclusion: The combination of imaging and clinical information improves outcome predictions. Supplemental material is available for this article. © RSNA, 2020., Competing Interests: Disclosures of Conflicts of Interest: Y.J.F.K. disclosed no relevant relationships. D.T. disclosed no relevant relationships. M.F. disclosed no relevant relationships. M.A.C. disclosed no relevant relationships. S.Z.M. disclosed no relevant relationships. S.M. disclosed no relevant relationships. N.V. disclosed no relevant relationships. C.E. disclosed no relevant relationships. A.J. disclosed no relevant relationships. A.B. disclosed no relevant relationships. Y.S.G. disclosed no relevant relationships. M.S.C. disclosed no relevant relationships. Z.A.F. disclosed no relevant relationships. B.S.G. disclosed no relevant relationships. E.K.O. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: author is a consultant for Google; author’s spouse works at Merck. Other relationships: disclosed no relevant relationships. A.B.C. disclosed no relevant relationships., (2021 by the Radiological Society of North America, Inc.)