Variability induced by the MR imager in dynamic contrast-enhanced imaging of the prostate

Performance of a Region of Interest-based Algorithm in Diagnosing International Society of Urological Pathology Grade Group ≥2 Prostate Cancer on the MRI-FIRST Database-CAD-FIRST Study

BACKGROUND AND OBJECTIVE

Prostate multiparametric magnetic resonance imaging (MRI) shows high sensitivity for International Society of Urological Pathology grade group (GG) ≥2 cancers. Many artificial intelligence algorithms have shown promising results in diagnosing clinically significant prostate cancer on MRI. To assess a region-of-interest-based machine-learning algorithm aimed at characterising GG ≥2 prostate cancer on multiparametric MRI.

METHODS

The lesions targeted at biopsy in the MRI-FIRST dataset were retrospectively delineated and assessed using a previously developed algorithm. The Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) score assigned prospectively before biopsy and the algorithm score calculated retrospectively in the regions of interest were compared for diagnosing GG ≥2 cancer, using the areas under the curve (AUCs), and sensitivities and specificities calculated with predefined thresholds (PIRADSv2 scores ≥3 and ≥4; algorithm scores yielding 90% sensitivity in the training database). Ten predefined biopsy strategies were assessed retrospectively.

KEY FINDINGS AND LIMITATIONS

After excluding 19 patients, we analysed 232 patients imaged on 16 different scanners; 85 had GG ≥2 cancer at biopsy. At patient level, AUCs of the algorithm and PI-RADSv2 were 77% (95% confidence interval [CI]: 70-82) and 80% (CI: 74-85; p = 0.36), respectively. The algorithm's sensitivity and specificity were 86% (CI: 76-93) and 65% (CI: 54-73), respectively. PI-RADSv2 sensitivities and specificities were 95% (CI: 89-100) and 38% (CI: 26-47), and 89% (CI: 79-96) and 47% (CI: 35-57) for thresholds of ≥3 and ≥4, respectively. Using the PI-RADSv2 score to trigger a biopsy would have avoided 26-34% of biopsies while missing 5-11% of GG ≥2 cancers. Combining prostate-specific antigen density, the PI-RADSv2 and algorithm's scores would have avoided 44-47% of biopsies while missing 6-9% of GG ≥2 cancers. Limitations include the retrospective nature of the study and a lack of PI-RADS version 2.1 assessment.

CONCLUSIONS AND CLINICAL IMPLICATIONS

The algorithm provided robust results in the multicentre multiscanner MRI-FIRST database and could help select patients for biopsy.

PATIENT SUMMARY

An artificial intelligence-based algorithm aimed at diagnosing aggressive cancers on prostate magnetic resonance imaging showed results similar to expert human assessment in a prospectively acquired multicentre test database.

Characterization of high-grade prostate cancer at multiparametric MRI using a radiomic-based computer-aided diagnosis system as standalone and second reader

PURPOSE

The purpose of this study was to develop and test across various scanners a zone-specific region-of-interest (ROI)-based computer-aided diagnosis system (CAD) aimed at characterizing, on MRI, International Society of Urological Pathology (ISUP) grade≥2 prostate cancers.

MATERIALS AND METHODS

ROI-based quantitative models were selected in multi-vendor training (265 pre-prostatectomy MRIs) and pre-test (112 pre-biopsy MRIs) datasets. The best peripheral and transition zone models were combined and retrospectively assessed in internal (158 pre-biopsy MRIs) and external (104 pre-biopsy MRIs) test datasets. Two radiologists (R1/R2) retrospectively delineated the lesions targeted at biopsy in test datasets. The CAD area under the receiver operating characteristic curve (AUC) for characterizing ISUP≥2 cancers was compared to that of the Prostate Imaging-Reporting and Data System version2 (PI-RADSv2) score prospectively assigned to targeted lesions.

RESULTS

The best models used the 25th apparent diffusion coefficient (ADC) percentile in transition zone and the 2nd ADC percentile and normalized wash-in rate in peripheral zone. The PI-RADSv2 AUCs were 82% (95% confidence interval [CI]: 74-87) and 86% (95% CI: 81-91) in the internal and external test datasets respectively. They were not different from the CAD AUCs obtained with R1 and R2 delineations, in the internal (82% [95% CI: 76-89], P = 0.95 and 85% [95% CI: 78-91], P = 0.55) and external (82% [95% CI: 74-91], P = 0.41 and 86% [95% CI:78-95], P = 0.98) test datasets. The CAD yielded sensitivities of 86-89% and 90-91%, and specificities of 64-65% and 69-75% in the internal and external test datasets respectively.

CONCLUSION

The CAD performance for characterizing ISUP grade≥2 prostate cancers on MRI is not different from that of PI-RADSv2 score across two test datasets.

Characterization of high-grade prostate cancer at multiparametric MRI: assessment of PI-RADS version 2.1 and version 2 descriptors across 21 readers with varying experience (MULTI study)

OBJECTIVE

To assess PI-RADSv2.1 and PI-RADSv2 descriptors across readers with varying experience.

METHODS

Twenty-one radiologists (7 experienced (≥ 5 years) seniors, 7 less experienced seniors and 7 juniors) assessed 240 'predefined' lesions from 159 pre-biopsy multiparametric prostate MRIs. They specified their location (peripheral, transition or central zone) and size, and scored them using PI-RADSv2.1 and PI-RADSv2 descriptors. They also described and scored 'additional' lesions if needed. Per-lesion analysis assessed the 'predefined' lesions, using targeted biopsy as reference; per-lobe analysis included 'predefined' and 'additional' lesions, using combined systematic and targeted biopsy as reference. Areas under the curve (AUCs) quantified the performance in diagnosing clinically significant cancer (csPCa; ISUP ≥ 2 cancer). Kappa coefficients (κ) or concordance correlation coefficients (CCC) assessed inter-reader agreement.

RESULTS

At per-lesion analysis, inter-reader agreement on location and size was moderate-to-good (κ = 0.60-0.73) and excellent (CCC ≥ 0.80), respectively. Agreement on PI-RADSv2.1 scoring was moderate (κ = 0.43-0.47) for seniors and fair (κ = 0.39) for juniors. Using PI-RADSv2.1, juniors obtained a significantly lower AUC (0.74; 95% confidence interval [95%CI]: 0.70-0.79) than experienced seniors (0.80; 95%CI 0.76-0.84; p = 0.008) but not than less experienced seniors (0.74; 95%CI 0.70-0.78; p = 0.75). As compared to PI-RADSv2, PI-RADSv2.1 downgraded 17 lesions/reader (interquartile range [IQR]: 6-29), of which 2 (IQR: 1-3) were csPCa; it upgraded 4 lesions/reader (IQR: 2-7), of which 1 (IQR: 0-2) was csPCa. Per-lobe analysis, which included 60 (IQR: 25-73) 'additional' lesions/reader, yielded similar results.

CONCLUSIONS

Experience significantly impacted lesion characterization using PI-RADSv2.1 descriptors. As compared to PI-RADSv2, PI-RADSv2.1 tended to downgrade non-csPCa lesions, but this effect was small and variable across readers.

MRI combined with PSA density in detecting clinically significant prostate cancer in patients with PSA serum levels of 4∼10ng/mL: Biparametric versus multiparametric MRI

PURPOSE

To compare the performance of biparametric magnetic resonance imaging (bpMRI) to that of multiparametric MRI (mpMRI) in combination with prostate-specific antigen density (PSAD) in detecting clinically significant prostate cancer (csPCa) in patients with PSA serum levels of 4∼10ng/mL.

MATERIALS AND METHODS

A total of 123 men (mean age, 66.3±8.9 [SD]; range: 42-83 years) with PSA serum levels of 4∼10ng/mL with suspected csPCa were included. All patients underwent mpMRI at 3 Tesla and transrectal ultrasound-guided prostate biopsy in their clinical workup and were followed-up for >1 year when no csPCa was found at initial biopsy. The mpMRI images were reinterpreted according to the Prostate Imaging Reporting and Data System (PI-RADS, v2.1) twice in two different sessions using either mpMRI sequences or bpMRI sequences. The patients were divided into 2 groups according to whether csPCa was detected. The PI-RADS (mpMRI or bpMRI) categories and PSAD were used in combination to detect csPCa. Receiver operating characteristic (ROC) curve and decision curve analyses were performed to compare the efficacy of the different models (mpMRI, bpMRI, PSAD, mpMRI+PSAD and bpMRI+PSAD).

RESULTS

Thirty-seven patients (30.1%, 37/123) had csPCa. ROC analysis showed that bpMRI (AUC=0.884 [95% confidence interval (CI): 0.814-0.935]) outperformed mpMRI (AUC=0.867 [95% CI: 0.794-0.921]) (P=0.035) and that bpMRI and mpMRI performed better than PSAD (0.682 [95% CI: 0.592-0.763]) in detecting csPCa; bpMRI+PSAD (AUC=0.907 [95% CI: 0.841-0.952]) performed similarly to mpMRI+PSAD (AUC=0.896 [95% CI: 0.828-0.944]) (P=0.151) and bpMRI (P=0.224). The sensitivity and specificity were 81.1% (95% CI: 64.8-92.0%) and 88.4% (95% CI: 79.7-94.3%), respectively for bpMRI, and 83.8% (95% CI: 68.0-93.8%) and 80.2% (95% CI: 70.2-88.0%), respectively for mpMRI (P>0.999 for sensitivity and P=0.016 for specificity). Among the 5 decision models, the decision curve analysis showed that all models (except for PSAD) achieved a high net benefit.

CONCLUSION

In patients with PSA serum levels of 4∼10ng/mL, bpMRI and bpMRI combined with PSAD achieve better performance than mpMRI in detecting csPCa; bpMRI has a higher specificity than mpMRI, which could decrease unnecessary biopsy, and may serve as a potential alternative to mpMRI to optimize clinical workup.

Differentiation between Fabry disease and hypertrophic cardiomyopathy with cardiac T1 mapping

PURPOSE

To evaluate the potential of non-contrast myocardial T1 mapping on cardiovascular magnetic resonance examination (CMR) in differentiating patients with Fabry disease (FD) from those with hypertrophic cardiomyopathy (HCM) and healthy control subjects.

MATERIALS AND METHODS

Seventeen patients with FD (8 men, 9 women; mean age, 48 ±18 [SD] years; [range: 19-73 years]; 53% with left ventricular hypertrophy [LVH]) were matched with 36 patients with hypertrophic cardiomyopathy (HCM) (22 men, 14 women; mean age, 57±16 [SD] years; [range: 22-85 years]) and 70 healthy control subjects (34 men, 36 women; mean age, 38 ±15 [SD] years; [range: 18-65 years]). Cardiac T1 mapping was performed using the modified Look-Locker inversion (MOLLI®) sequence on a 1.5-T magnet. T1 values were calculated, on midventricular section, for septal left ventricular segments (S8-S9) and all mid-ventricular ones (global T1 values; S7-S12). Statistical analysis included unpaired Mann-Whitney test, receiver operating characteristic curve and likelihood ratios.

RESULTS

Septal native T1 values were significantly decreased in patients with FD (889±61 [SD] ms; range: 784-980ms) compared to those with HCM (995±48 [SD] ms; range: 935-1125ms) (P<0.001) and versus healthy controls (965±29 [SD] ms; range: 910-1028ms) (P<0.001). Global native T1 values were also significantly decreased in patients with FD (891±49 [SD] ms; range 794-970ms) compared to those with HCM (995±34 [SD] ms; range: 952-1086ms) (P<0.001) and versus healthy controls (966±27 [SD] ms; range: 920-1042ms) (P<0.001). A septal left ventricular native T1 cutoff value of 940ms could distinguish FD from HCM with 88% sensitivity (95% CI: 73-100%) and 92% specificity (95% CI: 83-100%). Positive likelihood ratio was 11, negative likelihood ratio was 0.12. Compared to controls, the same threshold could distinguish FD with 88% sensitivity (95% CI: 73-100%) and 86% specificity (95% CI: 78-94%). Positive likelihood ratio was 6.3, negative likelihood ratio was 0.14. T1 value was abnormal in 4 of 8 (50%) of FD patients who did not have LVH.

CONCLUSION

Native T1 values are significantly lower in patients with FD by comparison with those with HCM and healthy volunteers.