Thomas Sanford1, Stephanie A Harmon2, Evrim B Turkbey3, Deepak Kesani1, Sena Tuncer1, Manuel Madariaga1, Chris Yang1, Jonathan Sackett1, Sherif Mehralivand1, Pingkun Yan4, Sheng Xu3,5, Bradford J Wood3,5, Maria J Merino6, Peter A Pinto7, Peter L Choyke1, Baris Turkbey1. 1. Molecular Imaging Program, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA. 2. Clinical Research Directorate, Frederick National Laboratory for Cancer Research sponsored by the National Cancer Institute, Frederick, Maryland, USA. 3. Department of Radiology, Clinical Center, National Institutes of Health, Bethesda, Maryland, USA. 4. Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, New York, USA. 5. Center for Interventional Oncology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA. 6. Laboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA. 7. Urologic Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA.
Abstract
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.
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.
Authors: Baris Turkbey; Haresh Mani; Vijay Shah; Ardeshir R Rastinehad; Marcelino Bernardo; Thomas Pohida; Yuxi Pang; Dagane Daar; Compton Benjamin; Yolanda L McKinney; Hari Trivedi; Celene Chua; Gennady Bratslavsky; Joanna H Shih; W Marston Linehan; Maria J Merino; Peter L Choyke; Peter A Pinto Journal: J Urol Date: 2011-09-25 Impact factor: 7.450
Authors: Andrew B Rosenkrantz; Luke A Ginocchio; Daniel Cornfeld; Adam T Froemming; Rajan T Gupta; Baris Turkbey; Antonio C Westphalen; James S Babb; Daniel J Margolis Journal: Radiology Date: 2016-04-01 Impact factor: 11.105
Authors: Sherif Mehralivand; Sandra Bednarova; Joanna H Shih; Francesca V Mertan; Sonia Gaur; Maria J Merino; Bradford J Wood; Peter A Pinto; Peter L Choyke; Baris Turkbey Journal: J Urol Date: 2017-03-31 Impact factor: 7.450
Authors: Berrend G Muller; Joanna H Shih; Sandeep Sankineni; Jamie Marko; Soroush Rais-Bahrami; Arvin Koruthu George; Jean J M C H de la Rosette; Maria J Merino; Bradford J Wood; Peter Pinto; Peter L Choyke; Baris Turkbey Journal: Radiology Date: 2015-06-18 Impact factor: 11.105
Authors: Jelle O Barentsz; Jeffrey C Weinreb; Sadhna Verma; Harriet C Thoeny; Clare M Tempany; Faina Shtern; Anwar R Padhani; Daniel Margolis; Katarzyna J Macura; Masoom A Haider; Francois Cornud; Peter L Choyke Journal: Eur Urol Date: 2015-09-08 Impact factor: 20.096
Authors: Baris Turkbey; Andrew B Rosenkrantz; Masoom A Haider; Anwar R Padhani; Geert Villeirs; Katarzyna J Macura; Clare M Tempany; Peter L Choyke; Francois Cornud; Daniel J Margolis; Harriet C Thoeny; Sadhna Verma; Jelle Barentsz; Jeffrey C Weinreb Journal: Eur Urol Date: 2019-03-18 Impact factor: 20.096
Authors: Geoffrey A Sonn; Richard E Fan; Pejman Ghanouni; Nancy N Wang; James D Brooks; Andreas M Loening; Bruce L Daniel; Katherine J To'o; Alan E Thong; John T Leppert Journal: Eur Urol Focus Date: 2017-12-07
Authors: Michelle Bardis; Roozbeh Houshyar; Chanon Chantaduly; Karen Tran-Harding; Alexander Ushinsky; Chantal Chahine; Mark Rupasinghe; Daniel Chow; Peter Chang Journal: Radiol Imaging Cancer Date: 2021-05
Authors: Teodora Telecan; Iulia Andras; Nicolae Crisan; Lorin Giurgiu; Emanuel Darius Căta; Cosmin Caraiani; Andrei Lebovici; Bianca Boca; Zoltan Balint; Laura Diosan; Monica Lupsor-Platon Journal: J Pers Med Date: 2022-06-16
Authors: Francesco Giganti; Sydney Lindner; Jonathan W Piper; Veeru Kasivisvanathan; Mark Emberton; Caroline M Moore; Clare Allen Journal: Eur Radiol Exp Date: 2021-11-05
Authors: Sarah M Ayyad; Mohamed A Badawy; Mohamed Shehata; Ahmed Alksas; Ali Mahmoud; Mohamed Abou El-Ghar; Mohammed Ghazal; Moumen El-Melegy; Nahla B Abdel-Hamid; Labib M Labib; H Arafat Ali; Ayman El-Baz Journal: Sensors (Basel) Date: 2022-02-26 Impact factor: 3.576