Diffusion in prostate cancer detection on a 3T scanner: How many b-values are needed?

Using deep learning to optimize the prostate MRI protocol by assessing the diagnostic efficacy of MRI sequences

PURPOSE

To explore diagnostic deep learning for optimizing the prostate MRI protocol by assessing the diagnostic efficacy of MRI sequences.

METHOD

This retrospective study included 840 patients with a biparametric prostate MRI scan. The MRI protocol included a T2-weighted image, three DWI sequences (b50, b400, and b800 s/mm2), a calculated ADC map, and a calculated b1400 sequence. Two accelerated MRI protocols were simulated, using only two acquired b-values to calculate the ADC and b1400. Deep learning models were trained to detect prostate cancer lesions on accelerated and full protocols. The diagnostic performances of the protocols were compared on the patient-level with the area under the receiver operating characteristic (AUROC), using DeLong's test, and on the lesion-level with the partial area under the free response operating characteristic (pAUFROC), using a permutation test. Validation of the results was performed among expert radiologists.

RESULTS

No significant differences in diagnostic performance were found between the accelerated protocols and the full bpMRI baseline. Omitting b800 reduced 53% DWI scan time, with a performance difference of + 0.01 AUROC (p = 0.20) and -0.03 pAUFROC (p = 0.45). Omitting b400 reduced 32% DWI scan time, with a performance difference of -0.01 AUROC (p = 0.65) and + 0.01 pAUFROC (p = 0.73). Multiple expert radiologists underlined the findings.

CONCLUSIONS

This study shows that deep learning can assess the diagnostic efficacy of MRI sequences by comparing prostate MRI protocols on diagnostic accuracy. Omitting either the b400 or the b800 DWI sequence can optimize the prostate MRI protocol by reducing scan time without compromising diagnostic quality.

Computer-aided detection of prostate cancer in early stages using multi-parameter MRI: A promising approach for early diagnosis

BACKGROUND

Transrectal ultrasound-guided prostate biopsy is the gold standard diagnostic test for prostate cancer, but it is an invasive examination of non-targeted puncture and has a high false-negative rate.

OBJECTIVE

In this study, we aimed to develop a computer-assisted prostate cancer diagnosis method based on multiparametric MRI (mpMRI) images.

METHODS

We retrospectively collected 106 patients who underwent radical prostatectomy after diagnosis with prostate biopsy. mpMRI images, including T2 weighted imaging (T2WI), diffusion weighted imaging (DWI), and dynamic-contrast enhanced (DCE), and were accordingly analyzed. We extracted the region of interest (ROI) about the tumor and benign area on the three sequential MRI axial images at the same level. The ROI data of 433 mpMRI images were obtained, of which 202 were benign and 231 were malignant. Of those, 50 benign and 50 malignant images were used for training, and the 333 images were used for verification. Five main feature groups, including histogram, GLCM, GLGCM, wavelet-based multi-fractional Brownian motion features and Minkowski function features, were extracted from the mpMRI images. The selected characteristic parameters were analyzed by MATLAB software, and three analysis methods with higher accuracy were selected.

RESULTS

Through prostate cancer identification based on mpMRI images, we found that the system uses 58 texture features and 3 classification algorithms, including Support Vector Machine (SVM), K-nearest Neighbor (KNN), and Ensemble Learning (EL), performed well. In the T2WI-based classification results, the SVM achieved the optimal accuracy and AUC values of 64.3% and 0.67. In the DCE-based classification results, the SVM achieved the optimal accuracy and AUC values of 72.2% and 0.77. In the DWI-based classification results, the ensemble learning achieved optimal accuracy as well as AUC values of 75.1% and 0.82. In the classification results based on all data combinations, the SVM achieved the optimal accuracy and AUC values of 66.4% and 0.73.

CONCLUSION

The proposed computer-aided diagnosis system provides a good assessment of the diagnosis of the prostate cancer, which may reduce the burden of radiologists and improve the early diagnosis of prostate cancer.

Cross-Modality Image Registration Using a Training-Time Privileged Third Modality

In this work, we consider the task of pairwise cross-modality image registration, which may benefit from exploiting additional images available only at training time from an additional modality that is different to those being registered. As an example, we focus on aligning intra-subject multiparametric Magnetic Resonance (mpMR) images, between T2-weighted (T2w) scans and diffusion-weighted scans with high b-value (DWI [Formula: see text]). For the application of localising tumours in mpMR images, diffusion scans with zero b-value (DWI [Formula: see text]) are considered easier to register to T2w due to the availability of corresponding features. We propose a learning from privileged modality algorithm, using a training-only imaging modality DWI [Formula: see text], to support the challenging multi-modality registration problems. We present experimental results based on 369 sets of 3D multiparametric MRI images from 356 prostate cancer patients and report, with statistical significance, a lowered median target registration error of 4.34 mm, when registering the holdout DWI [Formula: see text] and T2w image pairs, compared with that of 7.96 mm before registration. Results also show that the proposed learning-based registration networks enabled efficient registration with comparable or better accuracy, compared with a classical iterative algorithm and other tested learning-based methods with/without the additional modality. These compared algorithms also failed to produce any significantly improved alignment between DWI [Formula: see text] and T2w in this challenging application.

Deep Learning Reconstruction of Diffusion-weighted MRI Improves Image Quality for Prostatic Imaging

Background Deep learning reconstruction (DLR) may improve image quality. However, its impact on diffusion-weighted imaging (DWI) of the prostate has yet to be assessed. Purpose To determine whether DLR can improve image quality of diffusion-weighted MRI at b values ranging from 1000 sec/mm² to 5000 sec/mm² in patients with prostate cancer. Materials and Methods In this retrospective study, images of the prostate obtained at DWI with a b value of 0 sec/mm², DWI with a b value of 1000 sec/mm² (DWI₁₀₀₀), DWI with a b value of 3000 sec/mm² (DWI₃₀₀₀), and DWI with a b value of 5000 sec/mm² (DWI₅₀₀₀) from consecutive patients with biopsy-proven cancer from January to June 2020 were reconstructed with and without DLR. Image quality was assessed using signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) from region-of-interest analysis and qualitatively assessed using a five-point visual scoring system (1 [very poor] to 5 [excellent]) for each high-b-value DWI sequence with and without DLR. The SNR, CNR, and visual score for DWI with and without DLR were compared with the paired t test and the Wilcoxon signed rank test with Bonferroni correction, respectively. Apparent diffusion coefficients (ADCs) from DWI with and without DLR were also compared with the paired t test with Bonferroni correction. Results A total of 60 patients (mean age, 67 years; age range, 49-79 years) were analyzed. DWI with DLR showed significantly higher SNRs and CNRs than DWI without DLR (P < .001); for example, with DWI₁₀₀₀ the mean SNR was 38.7 ± 0.6 versus 17.8 ± 0.6, respectively (P < .001), and the mean CNR was 18.4 ± 5.6 versus 7.4 ± 5.6, respectively (P < .001). DWI with DLR also demonstrated higher qualitative image quality than DWI without DLR (mean score: 4.8 ± 0.4 vs 4.0 ± 0.7, respectively, with DWI₁₀₀₀ [P = .001], 3.8 ± 0.7 vs 3.0 ± 0.8 with DWI₃₀₀₀ [P = .002], and 3.1 ± 0.8 vs 2.0 ± 0.9 with DWI₅₀₀₀ [P < .001]). ADCs derived with and without DLR did not differ substantially (P > .99). Conclusion Deep learning reconstruction improves the image quality of diffusion-weighted MRI scans of prostate cancer with no impact on apparent diffusion coefficient quantitation with a 3.0-T MRI system. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Turkbey in this issue.

Synthetic Apparent Diffusion Coefficient for High b-Value Diffusion-Weighted MRI in Prostate

Purpose

It has been reported that diffusion-weighted imaging (DWI) with ultrahigh b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher Materials and Methods. Fifteen patients (7 malignant and 8 benign) were included in this study retrospectively with the institutional ethical committee approval. All images were acquired at a 3T MR scanner. The ADC values were calculated using a monoexponential model. Synthetic ADC (sADC) for higher b-value increases the diagnostic power of prostate cancer. DWI with higher

Results

No significant difference was observed between actual ADC and sADC for b-value increases the diagnostic power of prostate cancer. DWI with higher p=0.002, paired t-test) in sDWI as compared to DWI. Malignant lesions showed significantly lower sADC as compared to benign lesions (p=0.002, paired t-test) in sDWI as compared to DWI. Malignant lesions showed significantly lower sADC as compared to benign lesions (Discussion/

Conclusion

Our initial investigation suggests that the ADC values corresponding to higher b-value can be computed using log-linear relationship derived from lower b-values (b ≤ 1000). Our method might help clinicians to decide the optimal b-value for prostate lesion identification.b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher b-value increases the diagnostic power of prostate cancer. DWI with higher

Signal Intensity of High B-value Diffusion-weighted Imaging for the Detection of Prostate Cancer

Background:

Diffusion-weighted imaging (DWI) is a main component of multiparametric MRI for prostate cancer detection. Recently, high b value DWI has gained more attention because of its capability for tumor characterization.

Objective:

To assess based on histopathological findings of transrectal ultrasound (TRUS)-guided prostate biopsy as a reference, an increase in signal intensity of prostatic lesions in comparison with normal background tissue on high b-value diffusion-weighted images could be a sign of malignancy.

Material and Methods:

Fifty-three consecutive patients retrospectively included in the study. All patients underwent routine TRUS-guided prostate biopsies involving 12 cores after the magnetic resonance imaging (MRI) examinations. In seventeen patients (n =35 lesions), the prostate cancer was histologically confirmed by TRUS-guided prostate biopsy. The biopsy results of other patients were negative. Signal intensities on the high b-value (1600 s/mm²) images of the peripheral zone, the central gland, and the defined lesions were evaluated using region of interest-based measurements. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for prostate cancer detection using signal intensity of high b value diffusion-weighted images were calculated.

Results:

In the patients with confirmed prostate cancer, fourteen had visually increased SI on the high b-value images. The SI of lesions for these patients was higher than the SI of peripheral zone (22±18%) or central gland (31±20%). In patients with a negative biopsy, eight had visually increased SI on the high b-value images. The SI of lesions for these patients was 23±21% and 35±18% higher than the SI in the peripheral zone and the central gland, respectively. The sensitivity, specificity, PPV, and NPV for prostate cancer using SI of high b value DWI were 71, 87, 62, and 87 %, respectively.

Conclusion:

Visually increased SI on the high b-value images can be an indication of malignancy, although some benign lesions also show this increase in signal intensity.

Intravoxel incoherent motion diffusion MR and diffusion kurtosis imaging for discriminating atypical bone metastasis from benign bone lesion

Objectives:

To investigate the feasibility of intravoxel incoherent motion (IVIM) diffusion MR and diffusion kurtosis imaging (DKI) in discriminating atypical bone metastasis from benign bone lesion in patients with tumors.

Methods:

Patients with bone lesions in lower extremity suspected of metastases were enrolled in this prospective study. IVIM diffusion MR and DKI were performed before biopsy. Apparent diffusion coefficient (ADC), true diffusion (D), perfusion fraction (f) and perfusion-related pseudodiffusion (D*) were generated with IVIM, while mean kurtosis (MK) and mean diffusion (MD) generated with DKI. Two radiologists blinded to pathology results separately measured these parameters for each lesion through drawing region of interest. Intraclass correlation coefficient was used to determine the inter-reader viability in measurement. The patients with pathology-confirmed metastasis or benign lesion were analyzed. The Mann–Whitney test was used to compare IVIM and DKI parameters between metastasis group and benign lesion group. Receiver operating characteristic curves were constructed to evaluate the ability of discrimination.

Results:

Bone lesions from 28 patients (metastasis, n = 15; benign lesion, n = 13; mean age = 55 years; age range, 34~77) were analyzed with IVIM and DKI. Intraclass correlation coefficient was greater than 0.8 for all parameters. ADC, D and MD were significantly lower in metastases versus benign lesions (p<0.05). MK and f value were significantly higher in metastases versus benign lesions (p<0.05). D* was not significantly different between the two groups (p>0.05). Areas under curve for ADC, D, f, MK and MD were 0.935, 0.939, 0.891, 0.840 and 0.844 respectively.

Conclusions:

IVIM and DKI derived parameters distinguish between atypical bone metastasis and benign bone lesion in selected patients with tumors.

Advances in knowledge:

Bone metastasis and benign bone lesion differ in water molecular diffusion.

Intravoxel incoherent motion derived true diffusion distinguishes between atypical bone metastasis and benign lesion.

Risk Stratification of Prostate Cancer Using the Combination of Histogram Analysis of Apparent Diffusion Coefficient Across Tumor Diffusion Volume and Clinical Information: A Pilot Study

BACKGROUND

The effectiveness of quantitative MRI and clinical information in the risk stratification of prostate cancer (PCa) patients was evaluated separately in previous research; however, the differentiation power of combining quantitative MRI and clinical information has yet to be investigated.

PURPOSE

To investigate the power of combining histogram analysis of apparent diffusion coefficient (ADC) of tumor diffusion volume (tDv) with clinical information for the differentiation of low-grade (Gleason score [GS] ≤6) and high-grade (GS ≥7) PCa.

STUDY TYPE

Retrospective.

POPULATION

Fifty-nine PCa patients who underwent preoperative diffusion-weighted imaging (DWI) (acquired with b = 0, 1000 mm² /s) and followed by radical prostatectomy within 6 months.

SEQUENCES

T₂ -weighted, DWI, and ADC images at 3.0T.

ASSESSMENT

tDv defined with different ADC thresholds were analyzed for each patient and combined with age and prostate-specific antigen (PSA) level. Binary logistic regression with backward feature selection was applied to determine the best discrimination and corresponding combination of parameters.

STATISTICAL TESTS

Kolmogorov-Smirnov test; independent samples t-test; Mann-Whitney U-test; Spearman's rank correlation; receiver operating characteristic (ROC) analysis; binary logistical regression.

RESULTS

PSA and the 10^th percentile ADC value of tDv defined with different diffusion thresholds were significantly different between low-grade and high-grade PCa groups (P < 0.05 for all). Median ADC of tDv based on a threshold of 1.008 × 10^-3 mm² /s exhibited the best performance (AUC = 0.86, 95% confidence interval [CI]: 0.75-0.94), whereas binary logistic regression with backward feature selection achieved 97.20% accuracy with AUC = 0.978 (95% CI: 0.929-0.997).

DATA CONCLUSION

The discriminatory power of a single histogram variable of ADC in tDv was not significantly superior to that of a single clinical parameter. The combination of histogram analysis of ADC of tDv and clinical information using logistic regression might significantly improve the risk stratification of PCa and achieve reasonably high accuracy.

LEVEL OF EVIDENCE

4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:556-564.

Multiparametric (mp) MRI of prostate cancer

Prostate cancer (PCa) is one of the most prevalent cancers in men. A large number of men are detected with PCa; however, the clinical behavior ranges from low-grade indolent tumors that never develop into a clinically significant disease to aggressive, invasive tumors that may rapidly progress to metastatic disease. The challenges in clinical management of PCa are at levels of screening, diagnosis, treatment, and follow-up after treatment. Magnetic resonance imaging (MRI) methods have shown a potential role in detection, localization, staging, assessment of aggressiveness, targeting biopsies, etc. in PCa patients. Multiparametric MRI (mpMRI) is emerging as a better option compared to the individual imaging methods used in the evaluation of PCa. There are attempts to improve the reproducibility and reliability of mpMRI by using an objective scoring system proposed in the prostate imaging reporting and data system (PIRADS) for standardized reporting. Prebiopsy mpMRI may be used to detect PCa in men with elevated prostate-specific antigen or abnormal digital rectal examination and to enable targeted biopsies. mpMRI can also be used to decide on clinical management of patients, for example active surveillance, and may help in detecting only the pathology that requires detection. It can potentially not only guide patient selection for initial and repeat biopsy but also reduce false-negative biopsies. This review presents a description of the MR methods most commonly applied for investigations of prostate. The anatomical, functional and metabolic parameters obtained from these MR methods are discussed with regard to their physical basis and their contribution to mpMRI investigations of PCa.

Optimized b-value selection for the discrimination of prostate cancer grades, including the cribriform pattern, using diffusion weighted imaging

Multiparametric magnetic resonance imaging (MP-MRI), including diffusion-weighted imaging, is commonly used to diagnose prostate cancer. This radiology–pathology study correlates prostate cancer grade and morphology with common b-value combinations for calculating apparent diffusion coefficient (ADC). Thirty-nine patients undergoing radical prostatectomy were recruited for MP-MRI prior to surgery. Diffusion imaging was collected with seven b-values, and ADC was calculated. Excised prostates were sliced in the same orientation as the MRI using 3-D printed slicing jigs. Whole-mount slides were digitized and annotated by a pathologist. Annotated samples were aligned to the MRI, and ADC values were extracted from annotated peripheral zone (PZ) regions. A receiver operating characteristic (ROC) analysis was performed to determine accuracy of tissue type discrimination and optimal ADC b-value combination. ADC significantly discriminates Gleason (G) G4-5 cancer from G3 and other prostate tissue types. The optimal b-values for discriminating high from low-grade and noncancerous tissue in the PZ are 50 and 2000, followed closely by 100 to 2000 and 0 to 2000. Optimal ADC cut-offs are presented for dichotomized discrimination of tissue types according to each b-value combination. Selection of b-values affects the sensitivity and specificity of ADC for discrimination of prostate cancer.

PI-RADS version 2: evaluation of diffusion-weighted imaging interpretation between b = 1000 and b = 1500 s mm-2

OBJECTIVE

To investigate the variability of diffusion-weighted imaging (DWI) interpretation of Prostate Imaging Reporting and Data System (PI-RADS) version 2 (v2) in evaluating prostate cancer (PCa).

METHODS

154 patients with PCa underwent multiparametric 3T MRI, followed by radical prostatectomy. DWI with different b values (b = 0, 100, 1000 and 1500 s mm^-2) was obtained. Using the PI-RADS v2, two radiologists independently scored suspicious lesions in each patient and compared DWI of b = 1000 (DWI₁₀₀₀) with 1500 (DWI₁₅₀₀) s mm^-2.

RESULTS

On DWI₁₀₀₀ and DWI₁₅₀₀, the intermethod and interobserver agreements of DWI scores were excellent in all patients (κ ≥ 0.873). In each peripheral zone and transition zone DWI scores, both observers showed excellent intermethod agreement between DWI₁₀₀₀ and DWI₁₅₀₀ (κ ≥ 0.897), and interobserver agreement for DWI₁₀₀₀ and DWI₁₅₀₀ was good to excellent (κ ≥ 0.796). For estimating clinically significant cancer, the area under receiver operating characteristics curves of DWI₁₀₀₀ and DWI₁₅₀₀ were 0.710 and 0.724 for observer 1 (p = 0.11), and 0.649 and 0.656 for observer 2 (p = 0.12), respectively.

CONCLUSION

The PI-RADS v2 scoring at 3T shows excellent agreement between DWI₁₀₀₀ and DWI₁₅₀₀ in evaluating PCa, with excellent inter-observer agreement. Advance in knowledge: DWI using b = 1000 s mm^-2 instead of b = 1500 s mm^-2 reduces examination time or image distortion, with improved the signal-to-noise ratio.

Diffusion radiomics analysis of intratumoral heterogeneity in a murine prostate cancer model following radiotherapy: Pixelwise correlation with histology

PURPOSE

To investigate the biological meaning of apparent diffusion coefficient (ADC) values in tumors following radiotherapy.

MATERIALS AND METHODS

Five mice bearing TRAMP-C1 tumor were half-irradiated with a dose of 15 Gy. Diffusion-weighted images, using multiple b-values from 0 to 3000 s/mm² , were acquired at 7T on day 6. ADC values calculated by a two-point estimate and monoexponential fitting of signal decay were compared between the irradiated and nonirradiated regions of the tumor. Pixelwise ADC maps were correlated with histological metrics including nuclear counts, nuclear sizes, nuclear spaces, cytoplasmic spaces, and extracellular spaces.

RESULTS

As compared with the nonirradiated region, the irradiated region exhibited significant increases in ADC, extracellular space, and nuclear size, and a significant decrease in nuclear counts (P < 0.001 for all). Optimal ADC to differentiate the irradiated from nonirradiated regions was achieved at a b-value of 800 s/mm² by the two-point method and monoexponential curve fitting. ADC positively correlated with extracellular spaces (r = 0.74) and nuclear sizes (r = 0.72), and negatively correlated with nuclear counts (r = -0.82, P < 0.001 for all).

CONCLUSION

As a radiomic biomarker, ADC maps correlating with histological metrics pixelwise could be a means of evaluating tumor heterogeneity and responses to radiotherapy.

LEVEL OF EVIDENCE

1 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:483-489.

New Magnetic Resonance Imaging Index for Renal Fibrosis Assessment: A Comparison between Diffusion-Weighted Imaging and T1 Mapping with Histological Validation

A need exists to noninvasively assess renal interstitial fibrosis, a common process to all kidney diseases and predictive of renal prognosis. In this translational study, Magnetic Resonance Imaging (MRI) T1 mapping and a new segmented Diffusion-Weighted Imaging (DWI) technique, for Apparent Diffusion Coefficient (ADC), were first compared to renal fibrosis in two well-controlled animal models to assess detection limits. Validation against biopsy was then performed in 33 kidney allograft recipients (KARs). Predictive MRI indices, ΔT1 and ΔADC (defined as the cortico-medullary differences), were compared to histology. In rats, both T1 and ADC correlated well with fibrosis and inflammation showing a difference between normal and diseased kidneys. In KARs, MRI indices were not sensitive to interstitial inflammation. By contrast, ΔADC outperformed ΔT1 with a stronger negative correlation to fibrosis (R² = 0.64 against R² = 0.29 p < 0.001). ΔADC tends to negative values in KARs harboring cortical fibrosis of more than 40%. Using a discriminant analysis method, the ΔADC, as a marker to detect such level of fibrosis or higher, led to a specificity and sensitivity of 100% and 71%, respectively. This new index has potential for noninvasive assessment of fibrosis in the clinical setting.