Conventional vs. reduced field of view diffusion weighted imaging of the prostate: Comparison of image quality, correlation with histology, and inter-reader agreement

Brent A. Warndahl, Eric A. Borisch, Akira Kawashima, Stephen J Riederer, Adam T. Froemming

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Purpose To evaluate if Field of view Optimized and Constrained Undistorted Single shot (FOCUS) (GE Healthcare, Waukesha, WI) diffusion weighted images (DWI) provide more reliable imaging than conventional DWI, with non-inferior quantitative apparent diffusion coefficient (ADC) results. Material and methods IRB approval was obtained for this study of 43 patients (44 exams, one patient with two visits) that underwent multiparametric prostate MRI with two DWI sequences and subsequent radical prostatectomy with histology as the gold standard. Randomized DWI sequence images were graded independently by two blinded experienced prostate MRI radiologists with a period of memory extinction between the two separate reading sessions. Blinded images were also reviewed head to head in a later session for direct comparison. Multiple parameters were measured from a region of interest in a dominant lesion as well as two control areas. Patient characteristics were collected by chart review. Results There was good correlation between the mean ADC value for lesions obtained by conventional and FOCUS DWI (ρ = 0.85), with no trend toward any systematic difference, and equivalent correlation between ADC measurements and Gleason score. Agreement between the two readers was significantly higher for lesion ROI analysis with the FOCUS DWI derived ADC values (CCC 0.839) compared with the conventional ADC values (CCC 0.618; difference 0.221, 95% CI 0.01–0.46). FOCUS showed significantly better image quality scores (separate review: mean 2.17 ± 0.6, p < 0.001) compared to the conventional sequence (mean 2.65 ± 0.6, p < 0.001). In 13 cases the image quality was improved from grade of 3 + with conventional DWI to < 3 with FOCUS DWI, a clinically meaningful improvement. Head-to-head blinded review found 61 ratings showed strong to slight preference for FOCUS, 13 no preference, and 14 slight preference for the conventional sequence. There was also a strong and equivalent correlation between both sequences and PIRADS version 2 grading (ρ = − 0.56 and − 0.58 for FOCUS and conventional, respectively, p < 0.001 for both). Conclusion FOCUS DWI of the prostate shows significant improvement in inter-reader agreement and image quality. As opposed to previous conflicting smaller studies, we found equivalent ADC metrics compared with the conventional DWI sequence, and preserved correlation with Gleason score. In 52% of patients the improved image quality with FOCUS had the potential to salvage exams with otherwise limited to non-diagnostic DWI.

Original languageEnglish (US)
Pages (from-to)67-76
Number of pages10
JournalMagnetic Resonance Imaging
Volume47
DOIs
StatePublished - Apr 1 2018

Fingerprint

Histology
Image quality
Prostate
Imaging techniques
Neoplasm Grading
Magnetic resonance imaging
Salvaging
Research Ethics Committees
Prostatectomy

Keywords

  • Apparent diffusion coefficient
  • Diffusion weighted imaging
  • MRI
  • Prostate

ASJC Scopus subject areas

  • Biophysics
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

@article{7fa0eb0f5ee8426abd2f96669daed3ba,
title = "Conventional vs. reduced field of view diffusion weighted imaging of the prostate: Comparison of image quality, correlation with histology, and inter-reader agreement",
abstract = "Purpose To evaluate if Field of view Optimized and Constrained Undistorted Single shot (FOCUS) (GE Healthcare, Waukesha, WI) diffusion weighted images (DWI) provide more reliable imaging than conventional DWI, with non-inferior quantitative apparent diffusion coefficient (ADC) results. Material and methods IRB approval was obtained for this study of 43 patients (44 exams, one patient with two visits) that underwent multiparametric prostate MRI with two DWI sequences and subsequent radical prostatectomy with histology as the gold standard. Randomized DWI sequence images were graded independently by two blinded experienced prostate MRI radiologists with a period of memory extinction between the two separate reading sessions. Blinded images were also reviewed head to head in a later session for direct comparison. Multiple parameters were measured from a region of interest in a dominant lesion as well as two control areas. Patient characteristics were collected by chart review. Results There was good correlation between the mean ADC value for lesions obtained by conventional and FOCUS DWI (ρ = 0.85), with no trend toward any systematic difference, and equivalent correlation between ADC measurements and Gleason score. Agreement between the two readers was significantly higher for lesion ROI analysis with the FOCUS DWI derived ADC values (CCC 0.839) compared with the conventional ADC values (CCC 0.618; difference 0.221, 95{\%} CI 0.01–0.46). FOCUS showed significantly better image quality scores (separate review: mean 2.17 ± 0.6, p < 0.001) compared to the conventional sequence (mean 2.65 ± 0.6, p < 0.001). In 13 cases the image quality was improved from grade of 3 + with conventional DWI to < 3 with FOCUS DWI, a clinically meaningful improvement. Head-to-head blinded review found 61 ratings showed strong to slight preference for FOCUS, 13 no preference, and 14 slight preference for the conventional sequence. There was also a strong and equivalent correlation between both sequences and PIRADS version 2 grading (ρ = − 0.56 and − 0.58 for FOCUS and conventional, respectively, p < 0.001 for both). Conclusion FOCUS DWI of the prostate shows significant improvement in inter-reader agreement and image quality. As opposed to previous conflicting smaller studies, we found equivalent ADC metrics compared with the conventional DWI sequence, and preserved correlation with Gleason score. In 52{\%} of patients the improved image quality with FOCUS had the potential to salvage exams with otherwise limited to non-diagnostic DWI.",
keywords = "Apparent diffusion coefficient, Diffusion weighted imaging, MRI, Prostate",
author = "Warndahl, {Brent A.} and Borisch, {Eric A.} and Akira Kawashima and Riederer, {Stephen J} and Froemming, {Adam T.}",
year = "2018",
month = "4",
day = "1",
doi = "10.1016/j.mri.2017.10.011",
language = "English (US)",
volume = "47",
pages = "67--76",
journal = "Magnetic Resonance Imaging",
issn = "0730-725X",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Conventional vs. reduced field of view diffusion weighted imaging of the prostate

T2 - Comparison of image quality, correlation with histology, and inter-reader agreement

AU - Warndahl, Brent A.

AU - Borisch, Eric A.

AU - Kawashima, Akira

AU - Riederer, Stephen J

AU - Froemming, Adam T.

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Purpose To evaluate if Field of view Optimized and Constrained Undistorted Single shot (FOCUS) (GE Healthcare, Waukesha, WI) diffusion weighted images (DWI) provide more reliable imaging than conventional DWI, with non-inferior quantitative apparent diffusion coefficient (ADC) results. Material and methods IRB approval was obtained for this study of 43 patients (44 exams, one patient with two visits) that underwent multiparametric prostate MRI with two DWI sequences and subsequent radical prostatectomy with histology as the gold standard. Randomized DWI sequence images were graded independently by two blinded experienced prostate MRI radiologists with a period of memory extinction between the two separate reading sessions. Blinded images were also reviewed head to head in a later session for direct comparison. Multiple parameters were measured from a region of interest in a dominant lesion as well as two control areas. Patient characteristics were collected by chart review. Results There was good correlation between the mean ADC value for lesions obtained by conventional and FOCUS DWI (ρ = 0.85), with no trend toward any systematic difference, and equivalent correlation between ADC measurements and Gleason score. Agreement between the two readers was significantly higher for lesion ROI analysis with the FOCUS DWI derived ADC values (CCC 0.839) compared with the conventional ADC values (CCC 0.618; difference 0.221, 95% CI 0.01–0.46). FOCUS showed significantly better image quality scores (separate review: mean 2.17 ± 0.6, p < 0.001) compared to the conventional sequence (mean 2.65 ± 0.6, p < 0.001). In 13 cases the image quality was improved from grade of 3 + with conventional DWI to < 3 with FOCUS DWI, a clinically meaningful improvement. Head-to-head blinded review found 61 ratings showed strong to slight preference for FOCUS, 13 no preference, and 14 slight preference for the conventional sequence. There was also a strong and equivalent correlation between both sequences and PIRADS version 2 grading (ρ = − 0.56 and − 0.58 for FOCUS and conventional, respectively, p < 0.001 for both). Conclusion FOCUS DWI of the prostate shows significant improvement in inter-reader agreement and image quality. As opposed to previous conflicting smaller studies, we found equivalent ADC metrics compared with the conventional DWI sequence, and preserved correlation with Gleason score. In 52% of patients the improved image quality with FOCUS had the potential to salvage exams with otherwise limited to non-diagnostic DWI.

AB - Purpose To evaluate if Field of view Optimized and Constrained Undistorted Single shot (FOCUS) (GE Healthcare, Waukesha, WI) diffusion weighted images (DWI) provide more reliable imaging than conventional DWI, with non-inferior quantitative apparent diffusion coefficient (ADC) results. Material and methods IRB approval was obtained for this study of 43 patients (44 exams, one patient with two visits) that underwent multiparametric prostate MRI with two DWI sequences and subsequent radical prostatectomy with histology as the gold standard. Randomized DWI sequence images were graded independently by two blinded experienced prostate MRI radiologists with a period of memory extinction between the two separate reading sessions. Blinded images were also reviewed head to head in a later session for direct comparison. Multiple parameters were measured from a region of interest in a dominant lesion as well as two control areas. Patient characteristics were collected by chart review. Results There was good correlation between the mean ADC value for lesions obtained by conventional and FOCUS DWI (ρ = 0.85), with no trend toward any systematic difference, and equivalent correlation between ADC measurements and Gleason score. Agreement between the two readers was significantly higher for lesion ROI analysis with the FOCUS DWI derived ADC values (CCC 0.839) compared with the conventional ADC values (CCC 0.618; difference 0.221, 95% CI 0.01–0.46). FOCUS showed significantly better image quality scores (separate review: mean 2.17 ± 0.6, p < 0.001) compared to the conventional sequence (mean 2.65 ± 0.6, p < 0.001). In 13 cases the image quality was improved from grade of 3 + with conventional DWI to < 3 with FOCUS DWI, a clinically meaningful improvement. Head-to-head blinded review found 61 ratings showed strong to slight preference for FOCUS, 13 no preference, and 14 slight preference for the conventional sequence. There was also a strong and equivalent correlation between both sequences and PIRADS version 2 grading (ρ = − 0.56 and − 0.58 for FOCUS and conventional, respectively, p < 0.001 for both). Conclusion FOCUS DWI of the prostate shows significant improvement in inter-reader agreement and image quality. As opposed to previous conflicting smaller studies, we found equivalent ADC metrics compared with the conventional DWI sequence, and preserved correlation with Gleason score. In 52% of patients the improved image quality with FOCUS had the potential to salvage exams with otherwise limited to non-diagnostic DWI.

KW - Apparent diffusion coefficient

KW - Diffusion weighted imaging

KW - MRI

KW - Prostate

UR - http://www.scopus.com/inward/record.url?scp=85036545716&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85036545716&partnerID=8YFLogxK

U2 - 10.1016/j.mri.2017.10.011

DO - 10.1016/j.mri.2017.10.011

M3 - Article

C2 - 29103977

AN - SCOPUS:85036545716

VL - 47

SP - 67

EP - 76

JO - Magnetic Resonance Imaging

JF - Magnetic Resonance Imaging

SN - 0730-725X

ER -