Original Article
Low-to-high b value DWI ratio approaches in multiparametric MRI of the prostate: feasibility, optimal combination of b values, and comparison with ADC maps for the visual presentation of prostate cancer
Abstract
Background: Diffusion-weighted imaging (DWI) is considered by experts as one of the key elements in multi-parametric magnetic resonance imaging (mpMRI) employed in oncological studies outside the brain. A low-to-high b value ratio DWI has been proposed as an approach to decrease acquisition time and simplify the analysis of DWI data without the need to use a mathematical model.
Methods: Forty-three men with biopsy-proven prostate cancer (PCa) who underwent mpMRI of the prostate were included. Apparent diffusion coefficient (ADC) maps were created in the MRI scanner using a mono-exponential algorithm [b value (× number of averages) =0 (×1), 10 (×1), 25 (×1), 50 (×1), 100 (×1), 250 (×1), 450 (×1), 1,000 (×2), 1,500 (×3), and 2,000 (×5) s/mm2]. DWI ratio images were calculated with three previously estimated optimal b value combinations: (I) b=100 and b=1,000 s/mm2 (R1); (II) b=100 and b=1,500 s/mm2 (R2); and (III) b=100 and b=2,000 s/mm2 (R3). For quantitative analysis, contrast-to-noise ratio (CNR) between normal and cancerous tissue was compared between the ADC maps and the DWI ratio images in terms of noninferiority. For qualitative analysis, two radiologists read all images in a randomized order without knowing whether the presented image was an ADC map or a DWI ratio image. All images were scored in terms of artifacts, cancer conspicuity and overall image quality with a 5-point scale. Agreement between the readers was assessed by weighted kappa statistics. Agreement was considered as poor when kappa <0.4, fair to good when kappa >0.4 and <0.75 and excellent when kappa >0.75. Mean scores were compared between ADC and each of the DWI ratio images. Agreement between ADC maps and DWI ratio based synthetic ADC were assessed by intraclass correlation (ICC). Values less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater than 0.90 were indicative of poor, moderate, good, and excellent reliability, respectively. Median difference between low and intermediate/high risk were tested.
Results: Quantitative analysis shows DWI ratio images were not inferior to ADC maps quantitatively [P=0.0298 (ADC vs. R1), <0.0001 (ADC vs. R2) and <0.0001 (ADC vs. R3)]. Qualitatively, DWI ratio images were no more than 0.5 point on Likert scale lower than ADC in overall quality [P=0.0043 (ADC vs. R1), <0.0001 (ADC vs. R2), <0.0001 (ADC vs. R3)]. Reader agreement for the qualitative analysis was good to excellent (weighted kappa =0.4–0.7). Agreement between ADC maps and the synthetic ADC’s were excellent. Significant difference between low and intermediate/high risk were found in all measurements on average (all P values <0.05).
Conclusions: We presented an analytical method for searching for the optimal combination of high and low b values for DWI ratio images in terms of minimizing CNR between cancer and surrounding benign tissues. Optimized DWI ratio images are comparable both quantitatively and qualitatively to ADC maps for the interpretation of DWI data in the context of prostate mpMRI.
Methods: Forty-three men with biopsy-proven prostate cancer (PCa) who underwent mpMRI of the prostate were included. Apparent diffusion coefficient (ADC) maps were created in the MRI scanner using a mono-exponential algorithm [b value (× number of averages) =0 (×1), 10 (×1), 25 (×1), 50 (×1), 100 (×1), 250 (×1), 450 (×1), 1,000 (×2), 1,500 (×3), and 2,000 (×5) s/mm2]. DWI ratio images were calculated with three previously estimated optimal b value combinations: (I) b=100 and b=1,000 s/mm2 (R1); (II) b=100 and b=1,500 s/mm2 (R2); and (III) b=100 and b=2,000 s/mm2 (R3). For quantitative analysis, contrast-to-noise ratio (CNR) between normal and cancerous tissue was compared between the ADC maps and the DWI ratio images in terms of noninferiority. For qualitative analysis, two radiologists read all images in a randomized order without knowing whether the presented image was an ADC map or a DWI ratio image. All images were scored in terms of artifacts, cancer conspicuity and overall image quality with a 5-point scale. Agreement between the readers was assessed by weighted kappa statistics. Agreement was considered as poor when kappa <0.4, fair to good when kappa >0.4 and <0.75 and excellent when kappa >0.75. Mean scores were compared between ADC and each of the DWI ratio images. Agreement between ADC maps and DWI ratio based synthetic ADC were assessed by intraclass correlation (ICC). Values less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater than 0.90 were indicative of poor, moderate, good, and excellent reliability, respectively. Median difference between low and intermediate/high risk were tested.
Results: Quantitative analysis shows DWI ratio images were not inferior to ADC maps quantitatively [P=0.0298 (ADC vs. R1), <0.0001 (ADC vs. R2) and <0.0001 (ADC vs. R3)]. Qualitatively, DWI ratio images were no more than 0.5 point on Likert scale lower than ADC in overall quality [P=0.0043 (ADC vs. R1), <0.0001 (ADC vs. R2), <0.0001 (ADC vs. R3)]. Reader agreement for the qualitative analysis was good to excellent (weighted kappa =0.4–0.7). Agreement between ADC maps and the synthetic ADC’s were excellent. Significant difference between low and intermediate/high risk were found in all measurements on average (all P values <0.05).
Conclusions: We presented an analytical method for searching for the optimal combination of high and low b values for DWI ratio images in terms of minimizing CNR between cancer and surrounding benign tissues. Optimized DWI ratio images are comparable both quantitatively and qualitatively to ADC maps for the interpretation of DWI data in the context of prostate mpMRI.
