Multiparametric magnetic resonance imaging of the prostate with computer-aided detection: experienced observer performance study

Valentina Giannini, Simone Mazzetti, Enrico Armando, Silvia Carabalona, Filippo Russo, Alessandro Giacobbe, Giovanni Muto, Daniele Regge

Research output: Contribution to journalArticle

Abstract

Objectives: To compare the performance of experienced readers in detecting prostate cancer (PCa) using likelihood maps generated by a CAD system with that of unassisted interpretation of multiparametric magnetic resonance imaging (mp-MRI). Methods: Three experienced radiologists reviewed mp-MRI prostate cases twice. First, readers observed CAD marks on a likelihood map and classified as positive those suspicious for cancer. After 6 weeks, radiologists interpreted mp-MRI examinations unassisted, using their favourite protocol. Sensitivity, specificity, reading time and interobserver variability were compared for the two reading paradigms. Results: The dataset comprised 89 subjects of whom 35 with at least one significant PCa. Sensitivity was 80.9% (95% CI 72.1–88.0%) and 87.6% (95% CI 79.8–93.2; p = 0.105) for unassisted and CAD paradigm respectively. Sensitivity was higher with CAD for lesions with GS > 6 (91.3% vs 81.2%; p = 0.046) or diameter ≥10 mm (95.0% vs 80.0%; p = 0.006). Specificity was not affected by CAD. The average reading time with CAD was significantly lower (220 s vs 60 s; p < 0.001). Conclusions: Experienced readers using likelihood maps generated by a CAD scheme can detect more patients with ≥10 mm PCa lesions than unassisted MRI interpretation; overall reporting time is shorter. To gain more insight into CAD–human interaction, different reading paradigms should be investigated. Key points: • With CAD, sensitivity increases in patients with prostate tumours ≥10 mm and/or GS > 6. • CAD significantly reduces reporting time of multiparametric MRI. • When using CAD, a marginal increase of inter-reader agreement was observed.

Original languageEnglish
Pages (from-to)4200-4208
Number of pages9
JournalEuropean Radiology
Volume27
Issue number10
DOIs
Publication statusPublished - Oct 1 2017

Fingerprint

Reading
Prostate
Magnetic Resonance Imaging
Prostatic Neoplasms
Observer Variation
Sensitivity and Specificity
Neoplasms
Radiologists
Datasets

Keywords

  • Computer-aided detection
  • Diagnostic performance
  • Magnetic resonance imaging
  • Observer study
  • Prostate cancer

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Multiparametric magnetic resonance imaging of the prostate with computer-aided detection : experienced observer performance study. / Giannini, Valentina; Mazzetti, Simone; Armando, Enrico; Carabalona, Silvia; Russo, Filippo; Giacobbe, Alessandro; Muto, Giovanni; Regge, Daniele.

In: European Radiology, Vol. 27, No. 10, 01.10.2017, p. 4200-4208.

Research output: Contribution to journalArticle

@article{2f36cb1afd424e86b681fe64828d823c,
title = "Multiparametric magnetic resonance imaging of the prostate with computer-aided detection: experienced observer performance study",
abstract = "Objectives: To compare the performance of experienced readers in detecting prostate cancer (PCa) using likelihood maps generated by a CAD system with that of unassisted interpretation of multiparametric magnetic resonance imaging (mp-MRI). Methods: Three experienced radiologists reviewed mp-MRI prostate cases twice. First, readers observed CAD marks on a likelihood map and classified as positive those suspicious for cancer. After 6 weeks, radiologists interpreted mp-MRI examinations unassisted, using their favourite protocol. Sensitivity, specificity, reading time and interobserver variability were compared for the two reading paradigms. Results: The dataset comprised 89 subjects of whom 35 with at least one significant PCa. Sensitivity was 80.9{\%} (95{\%} CI 72.1–88.0{\%}) and 87.6{\%} (95{\%} CI 79.8–93.2; p = 0.105) for unassisted and CAD paradigm respectively. Sensitivity was higher with CAD for lesions with GS > 6 (91.3{\%} vs 81.2{\%}; p = 0.046) or diameter ≥10 mm (95.0{\%} vs 80.0{\%}; p = 0.006). Specificity was not affected by CAD. The average reading time with CAD was significantly lower (220 s vs 60 s; p < 0.001). Conclusions: Experienced readers using likelihood maps generated by a CAD scheme can detect more patients with ≥10 mm PCa lesions than unassisted MRI interpretation; overall reporting time is shorter. To gain more insight into CAD–human interaction, different reading paradigms should be investigated. Key points: • With CAD, sensitivity increases in patients with prostate tumours ≥10 mm and/or GS > 6. • CAD significantly reduces reporting time of multiparametric MRI. • When using CAD, a marginal increase of inter-reader agreement was observed.",
keywords = "Computer-aided detection, Diagnostic performance, Magnetic resonance imaging, Observer study, Prostate cancer",
author = "Valentina Giannini and Simone Mazzetti and Enrico Armando and Silvia Carabalona and Filippo Russo and Alessandro Giacobbe and Giovanni Muto and Daniele Regge",
year = "2017",
month = "10",
day = "1",
doi = "10.1007/s00330-017-4805-0",
language = "English",
volume = "27",
pages = "4200--4208",
journal = "European Radiology",
issn = "0938-7994",
publisher = "Springer Verlag",
number = "10",

}

TY - JOUR

T1 - Multiparametric magnetic resonance imaging of the prostate with computer-aided detection

T2 - experienced observer performance study

AU - Giannini, Valentina

AU - Mazzetti, Simone

AU - Armando, Enrico

AU - Carabalona, Silvia

AU - Russo, Filippo

AU - Giacobbe, Alessandro

AU - Muto, Giovanni

AU - Regge, Daniele

PY - 2017/10/1

Y1 - 2017/10/1

N2 - Objectives: To compare the performance of experienced readers in detecting prostate cancer (PCa) using likelihood maps generated by a CAD system with that of unassisted interpretation of multiparametric magnetic resonance imaging (mp-MRI). Methods: Three experienced radiologists reviewed mp-MRI prostate cases twice. First, readers observed CAD marks on a likelihood map and classified as positive those suspicious for cancer. After 6 weeks, radiologists interpreted mp-MRI examinations unassisted, using their favourite protocol. Sensitivity, specificity, reading time and interobserver variability were compared for the two reading paradigms. Results: The dataset comprised 89 subjects of whom 35 with at least one significant PCa. Sensitivity was 80.9% (95% CI 72.1–88.0%) and 87.6% (95% CI 79.8–93.2; p = 0.105) for unassisted and CAD paradigm respectively. Sensitivity was higher with CAD for lesions with GS > 6 (91.3% vs 81.2%; p = 0.046) or diameter ≥10 mm (95.0% vs 80.0%; p = 0.006). Specificity was not affected by CAD. The average reading time with CAD was significantly lower (220 s vs 60 s; p < 0.001). Conclusions: Experienced readers using likelihood maps generated by a CAD scheme can detect more patients with ≥10 mm PCa lesions than unassisted MRI interpretation; overall reporting time is shorter. To gain more insight into CAD–human interaction, different reading paradigms should be investigated. Key points: • With CAD, sensitivity increases in patients with prostate tumours ≥10 mm and/or GS > 6. • CAD significantly reduces reporting time of multiparametric MRI. • When using CAD, a marginal increase of inter-reader agreement was observed.

AB - Objectives: To compare the performance of experienced readers in detecting prostate cancer (PCa) using likelihood maps generated by a CAD system with that of unassisted interpretation of multiparametric magnetic resonance imaging (mp-MRI). Methods: Three experienced radiologists reviewed mp-MRI prostate cases twice. First, readers observed CAD marks on a likelihood map and classified as positive those suspicious for cancer. After 6 weeks, radiologists interpreted mp-MRI examinations unassisted, using their favourite protocol. Sensitivity, specificity, reading time and interobserver variability were compared for the two reading paradigms. Results: The dataset comprised 89 subjects of whom 35 with at least one significant PCa. Sensitivity was 80.9% (95% CI 72.1–88.0%) and 87.6% (95% CI 79.8–93.2; p = 0.105) for unassisted and CAD paradigm respectively. Sensitivity was higher with CAD for lesions with GS > 6 (91.3% vs 81.2%; p = 0.046) or diameter ≥10 mm (95.0% vs 80.0%; p = 0.006). Specificity was not affected by CAD. The average reading time with CAD was significantly lower (220 s vs 60 s; p < 0.001). Conclusions: Experienced readers using likelihood maps generated by a CAD scheme can detect more patients with ≥10 mm PCa lesions than unassisted MRI interpretation; overall reporting time is shorter. To gain more insight into CAD–human interaction, different reading paradigms should be investigated. Key points: • With CAD, sensitivity increases in patients with prostate tumours ≥10 mm and/or GS > 6. • CAD significantly reduces reporting time of multiparametric MRI. • When using CAD, a marginal increase of inter-reader agreement was observed.

KW - Computer-aided detection

KW - Diagnostic performance

KW - Magnetic resonance imaging

KW - Observer study

KW - Prostate cancer

UR - http://www.scopus.com/inward/record.url?scp=85017178046&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017178046&partnerID=8YFLogxK

U2 - 10.1007/s00330-017-4805-0

DO - 10.1007/s00330-017-4805-0

M3 - Article

AN - SCOPUS:85017178046

VL - 27

SP - 4200

EP - 4208

JO - European Radiology

JF - European Radiology

SN - 0938-7994

IS - 10

ER -