Multicenter Multireader Evaluation of an Artificial Intelligence–Based Attention Mapping System for the Detection of Prostate Cancer With Multiparametric MRI

Sherif Mehralivand, Stephanie A. Harmon, Joanna H. Shih, Clayton Smith, Nathan Lay, Burak Argun, Sandra Bednarova, Ronaldo Hueb Baroni, Abdullah Erdem Canda, Karabekır Ercan, Rossano Girometti, Ercan Karaarslan, Ali Rıza Kural, Andrei S. Purysko, Soroush Rais‐Bahrami, Victor Martins Tonso, Cristina Magi‐Galluzzi, Jennifer Gordetsky, Ricardo Macarenco, Maria J. Merino, Berrak Gümüşkaya, Yeşim Sağlıcan, Stefano Sioletic, Anne Y. Warren, Tristan Barrett, Leonardo Kayat Bittencourt, Mehmet Coşkun, Chris Knauss, Yan Mee Law, Ashkan A. Malayeri, Daniel Margolis, Jamie Marko, Derya Yakar, Bradford J. Wood, Peter A. Pinto, Peter L. Choyke, Ronald M. Summers, Barış TürkbeyAmerican 2020 · DOI

Abstract

<b>OBJECTIVE.</b> The purpose of this study was to evaluate in a multicenter dataset the performance of an artificial intelligence (AI) detection system with attention mapping compared with multiparametric MRI (mpMRI) interpretation in the detection of prostate cancer. <b>MATERIALS AND METHODS.</b> MRI examinations from five institutions were included in this study and were evaluated by nine readers. In the first round, readers evaluated mpMRI studies using the Prostate Imaging Reporting and Data System version 2. After 4 weeks, images were again presented to readers along with the AI-based detection system output. Readers accepted or rejected lesions within four AI-generated attention map boxes. Additional lesions outside of boxes were excluded from detection and categorization. The performances of readers using the mpMRI-only and AI-assisted approaches were compared. <b>RESULTS.</b> The study population included 152 case patients and 84 control patients with 274 pathologically proven cancer lesions. The lesion-based AUC was 74.9% for MRI and 77.5% for AI with no significant difference (<i>p</i> = 0.095). The sensitivity for overall detection of cancer lesions was higher for AI than for mpMRI but did not reach statistical significance (57.4% vs 53.6%, <i>p</i> = 0.073). However, for transition zone lesions, sensitivity was higher for AI than for MRI (61.8% vs 50.8%, <i>p</i> = 0.001). Reading time was longer for AI than for MRI (4.66 vs 4.03 minutes, <i>p</i> < 0.001). There was moderate interreader agreement for AI and MRI with no significant difference (58.7% vs 58.5%, <i>p</i> = 0.966). <b>CONCLUSION.</b> Overall sensitivity was only minimally improved by use of the AI system. Significant improvement was achieved, however, in the detection of transition zone lesions with use of the AI system at the cost of a mean of 40 seconds of additional reading time.
BibTeX
@inproceedings{mehralivand2020multicenter,
  author = {Sherif Mehralivand and Stephanie A. Harmon and Joanna H. Shih and Clayton Smith and Nathan Lay and Burak Argun and Sandra Bednarova and Ronaldo Hueb Baroni and Abdullah Erdem Canda and Karabekır Ercan and Rossano Girometti and Ercan Karaarslan and Ali Rıza Kural and Andrei S. Purysko and Soroush Rais‐Bahrami and Victor Martins Tonso and Cristina Magi‐Galluzzi and Jennifer Gordetsky and Ricardo Macarenco and Maria J. Merino and Berrak Gümüşkaya and Yeşim Sağlıcan and Stefano Sioletic and Anne Y. Warren and Tristan Barrett and Leonardo Kayat Bittencourt and Mehmet Coşkun and Chris Knauss and Yan Mee Law and Ashkan A. Malayeri and Daniel Margolis and Jamie Marko and Derya Yakar and Bradford J. Wood and Peter A. Pinto and Peter L. Choyke and Ronald M. Summers and Barış Türkbey},
  title = {{Multicenter Multireader Evaluation of an Artificial Intelligence–Based Attention Mapping System for the Detection of Prostate Cancer With Multiparametric MRI}},
  booktitle = {American},
  year = {2020},
  doi = {10.2214/ajr.19.22573},
}

Loading related content...

Figure