Quantitative analysis of multiparametric prostate MR images: differentiation between prostate cancer and normal tissue and correlation with Gleason score--a computer-aided diagnosis development study

Repeatability, reproducibility, and the effects of radiotherapy on radiomic features of lowfield MR-LINAC images of the prostate

Definitive radiotherapy (RT) has been shown to be a successful method of treating prostate cancer (PCa) patients. Through radiomics, a quantitative analysis of medical images, it is possible to adapt treatment early on, which may prevent or mitigate future adverse events. During RT of PCa, low-field magnetic resonance (MR) images, taken with a LINAC onboard imaging system in a process known as magnetic resonance-guided radiotherapy (MRgRT), are used to improve treatment accuracy via superior setup compared to x-ray methods. This work investigated baseline repeatability of radiomic features (RFs) by comparing planning MR images (pMR) with first-fraction setup images (FX1) taken with onboard MRI. The changes in RFs following RT were also looked at with the use of last-fraction setup images (FX5). Earlier research has investigated the use of planning images from cone beam CT (CBCT), but to our knowledge no research has previously shown the relationship with onboard MRI. The correlation between FX1 images and 3T diagnostic MR (dT2) images was also studied. Forty-three first and second order radiomic features extracted from these images were compared by calculating Lin's concordance correlation coefficient (with Benjamini-Hochberg correction for multiple comparisons) between the modalities. FX1 and pMR images were correlated (p<0.05) for all but one RF. 12 RFs correlated between pMR and dT2 images. There was a noticeable change in correlation values for RFs when looking at FX1 and FX5 images, with only 15 correlating significantly. The change in correlation values between pMR and FX5 images was comparable to that between FX1 and FX5 images, with 33 features having a CCC value deviation of less than 0.1. These results demonstrate that RF features are repeatable across different images of the same modality without treatment intervention. This study has also shown a noticeable, reproducible change in RFs as RT goes on. Reproducibility of RFs between different modalities was not strong. This study demonstrated that we can reliably use onboard MRI to observe day-to-day feature changes as a result of RT.

Quantitative 3-T Multiparametric MRI Parameters as Predictors of Aggressive Prostate Cancer

Purpose To determine which quantitative 3-T multiparametric MRI (mpMRI) parameters correlate with and help predict the presence of aggressive large cribriform pattern (LCP) and intraductal carcinoma (IDC) prostate cancer (PCa) at whole-mount histopathology (WMHP). Materials and Methods This retrospective study included 130 patients (mean age ± SD, 62.6 years ± 7.2; 100% male) with 141 PCa lesions who underwent preoperative prostate 3-T mpMRI, radical prostatectomy, and WMHP between January 2019 and December 2022. Lesions at WMHP were matched to 3-T mpMRI lesions with American College of Radiology Prostate Imaging Reporting and Data System version 2.1 scores of at least 3 or higher, and the following parameters were derived: apparent diffusion coefficient (ADC), volume transfer constant, rate constant, and initial area under the curve (iAUC). Each lesion was categorized into three subcohorts with increasing aggressiveness: LCP negative and IDC negative (subcohort 1), LCP positive and IDC negative (subcohort 2), and LCP positive and IDC negative (subcohort 3). Analysis of variance was performed to assess differences, Jonckheere test was performed to establish trends, and a classification and regression tree (CART) was used to establish a prediction model. Results Of the 141 total lesions, there were 41 (29.1%), 49 (34.8%), and 51 (36.2%) lesions in subcohorts 1, 2, and 3, with mean ADCs of 892 × 10-6 mm2/sec ± 20, 826 × 10-6 mm2/sec ± 209, and 763 × 10-6 mm2/sec ± 163 (P = .007) and mean iAUCs of 5.4 mmol/L/sec ± 2.5, 6.7 mmol/L/sec ± 3.0, and 6.9 mmol/L/sec ± 3.5 (P = .04), respectively. ADC was negatively correlated (P = .004), and rate constant and iAUC were positively correlated (P = .048 and P = .04, respectively) with increasing histologic PCa aggressiveness. The CART model correctly allocated 39.0%, 24.5%, and 84.3% of PCa lesions to subcohorts 1, 2, and 3, respectively. Conclusion Quantitative 3-T mpMRI parameters significantly correlated with and helped predict aggressive LCP and IDC PCa at WMHP. Keywords: Prostate, MRI, Pathology © RSNA, 2025.

Differential Diagnosis of Prostate Cancer Grade to Augment Clinical Diagnosis Based on Classifier Models with Tuned Hyperparameters

We developed a novel machine-learning algorithm to augment the clinical diagnosis of prostate cancer utilizing first and second-order texture analysis metrics in a novel application of machine-learning radiomics analysis. We successfully discriminated between significant prostate cancers versus non-tumor regions and provided accurate prediction between Gleason score cohorts with statistical sensitivity of 0.82, 0.81 and 0.91 in three separate pathology classifications. Tumor heterogeneity and prediction of the Gleason score were quantified using two feature selection approaches and two separate classifiers with tuned hyperparameters. There was a total of 71 patients analyzed in this study. Multiparametric MRI, incorporating T2WI and ADC maps, were used to derive radiomics features. Recursive feature elimination (RFE), the least absolute shrinkage and selection operator (LASSO), and two classification approaches, incorporating a support vector machine (SVM) (with randomized search) and random forest (RF) (with grid search), were utilized to differentiate between non-tumor regions and significant cancer while also predicting the Gleason score. In T2WI images, the RFE feature selection approach combined with RF and SVM classifiers outperformed LASSO with SVM and RF classifiers. The best performance was achieved by combining LASSO and SVM into a model that used both T2WI and ADC images. This model had an area under the curve (AUC) of 0.91. Radiomic features computed from ADC and T2WI images were used to predict three groups of Gleason score using two kinds of feature selection methods (RFE and LASSO), RF and SVM classifier models with tuned hyperparameters. Using combined sequences (T2WI and ADC map images) and combined radiomics (1st and GLCM features), LASSO, with a feature selection method with RF, was able to predict G3 with the highest sensitivity at a level AUC of 0.92. To predict G3 for single sequence (T2WI images) using GLCM features, LASSO with SVM achieved the highest sensitivity with an AUC of 0.92.

A PSMA PET/CT-based risk model for prediction of concordance between targeted biopsy and combined biopsy in detecting prostate cancer

PURPOSE

This study is to investigate the diagnostic value of 68Ga-PSMA-11 in improving the concordance between mpMRI-TB and combined biopsy (CB) in detecting PCa.

METHODS

115 consecutive men with 68Ga-PSMA-11 PET/CT prior to prostate biopsy were included for analysis. PSMA intensity, quantified as maximum standard uptake value (SUVmax), minimum apparent diffusion coefficient (ADCmin) and other clinical characteristics were evaluated relative to biopsy concordance using univariate and multivariate logistic regression analyses. A prediction model was developed based on the identified parameters, and a dynamic online diagnostic nomogram was constructed, with its discrimination evaluated through the area under the ROC curve (AUC) and consistency assessed using calibration plots. To assess its clinical applicability, a decision curve analysis (DCA) was performed, while internal validation was conducted using bootstrapping methods.

RESULTS

Concordance between mpMRI-TB and CB occurred in 76.5% (88/115) of the patients. Multivariate logistic regression analyses performed that SUVmax (OR= 0.952; 95% CI 0.917-0.988; P= 0.010) and ADCmin (OR= 1.006; 95% CI 1.003-1.010; P= 0.001) were independent risk factors for biopsy concordance. The developed model showed a sensitivity, specificity, accuracy and AUC of 0.67, 0.78, 0.81 and 0.78 in the full sample. The calibration curve demonstrated that the nomogram's predicted outcomes closely resembled the ideal curve, indicating consistency between predicted and actual outcomes. Furthermore, the decision curve analysis (DCA) highlighted the clinical net benefit achievable across various risk thresholds. These findings were reinforced by internal validation.

CONCLUSIONS

The developed prediction model based on SUVmax and ADCmin showed practical value in guiding the optimization of prostate biopsy pattern. Lower SUVmax and Higher ADCmin values are associated with greater confidence in implementing mono-TB and safely avoiding SB, effectively balancing benefits and risks.

Development and validation of a nomogram based on biparametric MRI PI-RADS v2.1 and clinical parameters to avoid unnecessary prostate biopsies

BACKGROUND

Biparametric MRI (bpMRI) is a faster, contrast-free, and less expensive MRI protocol that facilitates the detection of prostate cancer. The aim of this study is to determine whether a biparametric MRI PI-RADS v2.1 score-based model could reduce unnecessary biopsies in patients with suspected prostate cancer (PCa).

METHODS

The patients who underwent MRI-guided biopsies and systematic biopsies between January 2020 and January 2022 were retrospectively analyzed. The development cohort used to derive the prediction model consisted of 275 patients. Two validation cohorts included 201 patients and 181 patients from 2 independent institutions. Predictive models based on the bpMRI PI-RADS v2.1 score (bpMRI score) and clinical parameters were used to detect clinically significant prostate cancer (csPCa) and compared by analyzing the area under the curve (AUC) and decision curves. Spearman correlation analysis was utilized to determine the relationship between International Society of Urological Pathology (ISUP) grade and clinical parameters/bpMRI score.

RESULTS

Logistic regression models were constructed using data from the development cohort to generate nomograms. By applying the models to the all cohorts, the AUC for csPCa was significantly higher for the bpMRI PI-RADS v2.1 score-based model than for the clinical model in both cohorts (p < 0.001). Considering the test trade-offs, urologists would agree to perform 10 fewer bpMRIs to avoid one unnecessary biopsy, with a risk threshold of 10-20% in practice. Correlation analysis showed a strong correlation between the bpMRI score and ISUP grade.

CONCLUSION

A predictive model based on the bpMRI score and clinical parameters significantly improved csPCa risk stratification, and the bpMRI score can be used to determine the aggressiveness of PCa prior to biopsy.

Prostate lesion segmentation based on a 3D end-to-end convolution neural network with deep multi-scale attention

Prostate cancer is one of the deadest cancers among human beings. To better diagnose the prostate cancer, prostate lesion segmentation becomes a very important work, but its progress is very slow due to the prostate lesions small in size, irregular in shape, and blurred in contour. Therefore, automatic prostate lesion segmentation from mp-MRI is a great significant work and a challenging task. However, the most existing multi-step segmentation methods based on voxel-level classification are time-consuming, may introduce errors in different steps and lead to error accumulation. To decrease the computation time, harness richer 3D spatial features, and fuse the multi-level contextual information of mp-MRI, we present an automatic segmentation method in which all steps are optimized conjointly as one step to form our end-to-end convolutional neural network. The proposed end-to-end network DMSA-V-Net consists of two parts: (1) a 3D V-Net is used as the backbone network, it is the first attempt in employing 3D convolutional neural network for CS prostate lesion segmentation, (2) a deep multi-scale attention mechanism is introduced into the 3D V-Net which can highly focus on the ROI while suppressing the redundant background. As a merit, the attention can adaptively re-align the context information between the feature maps at different scales and the saliency maps in high-levels. We performed experiments based on five cross-fold validation with data including 97 patients. The results show that the Dice and sensitivity are 0.7014 and 0.8652 respectively, which demonstrates that our segmentation approach is more significant and accurate compared to other methods.

Mutually communicated model based on multi-parametric MRI for automated segmentation and classification of prostate cancer

BACKGROUND

Multiparametric magnetic resonance imaging (mp-MRI) is introduced and established as a noninvasive alternative for prostate cancer (PCa) detection and characterization.

PURPOSE

To develop and evaluate a mutually communicated deep learning segmentation and classification network (MC-DSCN) based on mp-MRI for prostate segmentation and PCa diagnosis.

METHODS

The proposed MC-DSCN can transfer mutual information between segmentation and classification components and facilitate each other in a bootstrapping way. For classification task, the MC-DSCN can transfer the masks produced by the coarse segmentation component to the classification component to exclude irrelevant regions and facilitate classification. For segmentation task, this model can transfer the high-quality localization information learned by the classification component to the fine segmentation component to mitigate the impact of inaccurate localization on segmentation results. Consecutive MRI exams of patients were retrospectively collected from two medical centers (referred to as center A and B). Two experienced radiologists segmented the prostate regions, and the ground truth of the classification refers to the prostate biopsy results. MC-DSCN was designed, trained, and validated using different combinations of distinct MRI sequences as input (e.g., T2-weighted and apparent diffusion coefficient) and the effect of different architectures on the network's performance was tested and discussed. Data from center A were used for training, validation, and internal testing, while another center's data were used for external testing. The statistical analysis is performed to evaluate the performance of the MC-DSCN. The DeLong test and paired t-test were used to assess the performance of classification and segmentation, respectively.

RESULTS

In total, 134 patients were included. The proposed MC-DSCN outperforms the networks that were designed solely for segmentation or classification. Regarding the segmentation task, the classification localization information helped to improve the IOU in center A: from 84.5% to 87.8% (p < 0.01) and in center B: from 83.8% to 87.1% (p < 0.01), while the area under curve (AUC) of PCa classification was improved in center A: from 0.946 to 0.991 (p < 0.02) and in center B: from 0.926 to 0.955 (p < 0.01) as a result of the additional information provided by the prostate segmentation.

CONCLUSION

The proposed architecture could effectively transfer mutual information between segmentation and classification components and facilitate each other in a bootstrapping way, thus outperforming the networks designed to perform only one task.

Radiomics-based machine-learning method to diagnose prostate cancer using mp-MRI: a comparison between conventional and fused models

OBJECTIVES

Multiparametric MRI (mp-MRI) has been significantly used for detection, localization and staging of Prostate cancer (PCa). However, all the assessment suffers from poor reproducibility among the readers. The aim of this study was to evaluate radiomics models to diagnose PCa using high-resolution T2-weighted (T2-W) and dynamic contrast-enhanced (DCE) MRI.

MATERIALS AND METHODS

Thirty two patients who had high prostate specific antigen level were recruited. The prostate biopsies considered as the reference to differentiate between 66 benign and 36 malignant prostate lesions. 181 features were extracted from each modality. K-nearest neighbors, artificial neural network, decision tree, and linear discriminant analysis were used for machine-learning study. The leave-one-out cross-validation method was used to prevent overfitting and build robust models.

RESULTS

Radiomics analysis showed that T2-W images were more effective in PCa detection compare to DCE images. Local binary pattern features and speeded up robust features had the highest ability for prediction in T2-W and DCE images, respectively. The classifier fusion using decision template method showed the highest performance with accuracy, specificity, and sensitivity of 100%.

DISCUSSION

The findings of this framework provide researchers on PCa with a promising method for reliable detection of prostate lesions in MR images by fused model.

Incremental prognostic value of ADC histogram analysis in patients with high-risk prostate cancer receiving adjuvant hormonal therapy after radical prostatectomy

Purpose

To investigate the incremental prognostic value of preoperative apparent diffusion coefficient (ADC) histogram analysis in patients with high-risk prostate cancer (PCa) who received adjuvant hormonal therapy (AHT) after radical prostatectomy (RP).

Methods

Sixty-two PCa patients in line with the criteria were enrolled in this study. The 10^th, 50^th, and 90^th percentiles of ADC (ADC₁₀, ADC₅₀, ADC₉₀), the mean value of ADC (ADC_mean), kurtosis, and skewness were obtained from the whole-lesion ADC histogram. The Kaplan-Meier method and Cox regression analysis were used to analyze the relationship between biochemical recurrence-free survival (BCR-fs) and ADC parameters and other clinicopathological factors. Prognostic models were constructed with and without ADC parameters.

Results

The median follow-up time was 53.4 months (range, 41.1-79.3 months). BCR was found in 19 (30.6%) patients. Kaplan-Meier curves showed that lower ADC_mean, ADC₁₀, ADC₅₀, and ADC₉₀ and higher kurtosis could predict poorer BCR-fs (all p<0.05). After adjusting for clinical parameters, ADC₅₀ and kurtosis remained independent prognostic factors for BCR-fs (HR: 0.172, 95% CI: 0.055-0.541, p=0.003; HR: 7.058, 95% CI: 2.288-21.773, p=0.001, respectively). By adding ADC parameters to the clinical model, the C index and diagnostic accuracy for the 24- and 36-month BCR-fs were improved.

Conclusion

ADC histogram analysis has incremental prognostic value in patients with high-risk PCa who received AHT after RP. Combining ADC₅₀, kurtosis and clinical parameters can improve the accuracy of BCR-fs prediction.

Contribution of Dynamic Contrast-enhanced and Diffusion MRI to PI-RADS for Detecting Clinically Significant Prostate Cancer

Background Prostate Imaging Reporting and Data System (PI-RADS) version 2.0 requires multiparametric MRI of the prostate, including diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) imaging sequences; however, the contribution of DCE imaging remains unclear. Purpose To assess whether DCE imaging in addition to apparent diffusion coefficient (ADC) and normalized T2 values improves PI-RADS version 2.0 for prediction of clinically significant prostate cancer (csPCa). Materials and Methods In this retrospective study, clinically reported PI-RADS lesions in consecutive men who underwent 3-T multiparametric MRI (T2-weighted, DWI, and DCE MRI) from May 2015 to September 2016 were analyzed quantitatively and compared with systematic and targeted MRI-transrectal US fusion biopsy. The normalized T2 signal (nT2), ADC measurement, mean early-phase DCE signal (mDCE), and heuristic DCE parameters were calculated. Logistic regression analysis indicated the most predictive DCE parameters for csPCa (Gleason grade group ≥2). Receiver operating characteristic parameter models were compared using the Obuchowski test. Recursive partitioning analysis determined ADC and mDCE value ranges for combined use with PI-RADS. Results Overall, 260 men (median age, 64 years [IQR, 58-69 years]) with 432 lesions (csPCa [n = 152] and no csPCa [n = 280]) were included. The mDCE parameter was predictive of csPCa when accounting for the ADC and nT2 parameter in the peripheral zone (odds ratio [OR], 1.76; 95% CI: 1.30, 2.44; P = .001) but not the transition zone (OR, 1.17; 95% CI: 0.81, 1.69; P = .41). Recursive partitioning analysis selected an ADC cutoff of 0.897 × 10^-3 mm²/sec (P = .04) as a classifier for peripheral zone lesions with a PI-RADS score assessed on the ADC map (hereafter, ADC PI-RADS) of 3. The mDCE parameter did not differentiate ADC PI-RADS 3 lesions (P = .11), but classified lesions with ADC PI-RADS scores greater than 3 with low ADC values (less than 0.903 × 10^-3 mm²/sec, P < .001) into groups with csPCa rates of 70% and 97% (P = .008). A lesion size cutoff of 1.5 cm and qualitative DCE parameters were not defined as classifiers according to recursive partitioning (P > .05). Conclusion Quantitative or qualitative dynamic contrast-enhanced MRI was not relevant for Prostate Imaging Reporting and Data System (PI-RADS) 3 lesion risk stratification, while quantitative apparent diffusion coefficient (ADC) values were helpful in upgrading PI-RADS 3 and PI-RADS 4 lesions. Quantitative ADC measurement may be more important for risk stratification than current methods in future versions of PI-RADS. © RSNA, 2022 Online supplemental material is available for this article See also the editorial by Goh in this issue.

The detection of prostate cancer based on ultrasound RF signal

Objective

The diagnosis of prostate cancer has been a challenging task. Compared with traditional diagnosis methods, the radiofrequency (RF) signal is not only non-invasive but also rich in microscopic lesion information. This paper proposes a novel and accurate method for detecting prostate cancer based on the ultrasound RF signal.

Method

Our approach is based on low-dimensional features in the frequency domain and high-throughput features in the spatial domain. The whole process could be divided into two parts: first, we calculate three feature maps from the ultrasound original RF signal, and 1,050 radiomics features are extracted from the three feature maps; second, we extracted 37 spectral features from the normalized frequency spectrum after Fourier transform.

Results

We use LASSO regression as the method for feature selection; moreover, we use support vector machine (SVM) for classification 10-fold cross-validation for examining the classification performance of the SVM. An AUC (area under the receiver operating characteristic curve) of 0.84 was obtained on 71 subjects.

Conclusions

Our method is feasible to detect prostate cancer based on the ultrasound RF signal with superior classification performance.

Cone-beam CT delta-radiomics to predict genitourinary toxicities and international prostate symptom of prostate cancer patients: a pilot study

For prostate cancer (PCa) patients treated with definitive radiotherapy (RT), acute and late RT-related genitourinary (GU) toxicities adversely impact disease-specific quality of life. Early warning of potential RT toxicities can prompt interventions that may prevent or mitigate future adverse events. During intensity modulated RT (IMRT) of PCa, daily cone-beam computed tomography (CBCT) images are used to improve treatment accuracy through image guidance. This work investigated the performance of CBCT-based delta-radiomic features (DRF) models to predict acute and sub-acute International Prostate Symptom Scores (IPSS) and Common Terminology Criteria for Adverse Events (CTCAE) version 5 GU toxicity grades for 50 PCa patients treated with definitive RT. Delta-radiomics models were built using logistic regression, random forest for feature selection, and a 1000 iteration bootstrapping leave one analysis for cross validation. To our knowledge, no prior studies of PCa have used DRF models based on daily CBCT images. AUC of 0.83 for IPSS and greater than 0.7 for CTCAE grades were achieved as early as week 1 of treatment. DRF extracted from CBCT images showed promise for the development of models predictive of RT outcomes. Future studies will include using artificial intelligence and machine learning to expand CBCT sample sizes available for radiomics analysis.

Prostate cancer classification using radiomics and machine learning on mp-MRI validated using co-registered histology

BACKGROUND

Multi-parametric magnetic resonance imaging (mp-MRI) is emerging as a useful tool for prostate cancer (PCa) detection but currently has unaddressed limitations. Computer aided diagnosis (CAD) systems have been developed to address these needs, but many approaches used to generate and validate the models have inherent biases.

METHOD

All clinically significant PCa on histology was mapped to mp-MRI using a previously validated registration algorithm. Shape and size matched non-PCa regions were selected using a proposed sampling algorithm to eliminate biases towards shape and size. Further analysis was performed to assess biases regarding inter-zonal variability.

RESULTS

A 5-feature Naïve-Bayes classifier produced an area under the receiver operating characteristic curve (AUC) of 0.80 validated using leave-one-patient-out cross-validation. As mean inter-class area mismatch increased, median AUC trended towards positively biasing classifiers to producing higher AUCs. Classifiers were invariant to differences in shape between PCa and non-PCa lesions (AUC: 0.82 vs 0.82). Performance for models trained and tested only in the peripheral zone was found to be lower than in the central gland (AUC: 0.75 vs 0.95).

CONCLUSION

We developed a radiomics based machine learning system to classify PCa vs non-PCa tissue on mp-MRI validated on accurately co-registered mid-gland histology with a measured target registration error. Potential biases involved in model development were interrogated to provide considerations for future work in this area.

The role of MRI in prostate cancer: current and future directions

There has been an increasing role of magnetic resonance imaging (MRI) in the management of prostate cancer. MRI already plays an essential role in the detection and staging, with the introduction of functional MRI sequences. Recent advancements in radiomics and artificial intelligence are being tested to potentially improve detection, assessment of aggressiveness, and provide usefulness as a prognostic marker. MRI can improve pretreatment risk stratification and therefore selection of and follow-up of patients for active surveillance. MRI can also assist in guiding targeted biopsy, treatment planning and follow-up after treatment to assess local recurrence. MRI has gained importance in the evaluation of metastatic disease with emerging technology including whole-body MRI and integrated positron emission tomography/MRI, allowing for not only better detection but also quantification. The main goal of this article is to review the most recent advances on MRI in prostate cancer and provide insights into its potential clinical roles from the radiologist's perspective. In each of the sections, specific roles of MRI tailored to each clinical setting are discussed along with its strengths and weakness including already established material related to MRI and the introduction of recent advancements on MRI.

Semi-Automatic Multiparametric MR Imaging Classification Using Novel Image Input Sequences and 3D Convolutional Neural Networks

The role of multi-parametric magnetic resonance imaging (mp-MRI) is becoming increasingly important in the diagnosis of the clinical severity of prostate cancer (PCa). However, mp-MRI images usually contain several unaligned 3D sequences, such as DWI image sequences and T2-weighted image sequences, and there are many images among the entirety of 3D sequence images that do not contain cancerous tissue, which affects the accuracy of large-scale prostate cancer detection. Therefore, there is a great need for a method that uses accurate computer-aided detection of mp-MRI images and minimizes the influence of useless features. Our proposed PCa detection method is divided into three stages: (i) multimodal image alignment, (ii) automatic cropping of the sequence images to the entire prostate region, and, finally, (iii) combining multiple modal images of each patient into novel 3D sequences and using 3D convolutional neural networks to learn the newly composed 3D sequences with different modal alignments. We arrange the different modal methods to make the model fully learn the cancerous tissue features; then, we predict the clinical severity of PCa and generate a 3D cancer response map for the 3D sequence images from the last convolution layer of the network. The prediction results and 3D response map help to understand the features that the model focuses on during the process of 3D-CNN feature learning. We applied our method to Toho hospital prostate cancer patient data; the AUC (=0.85) results were significantly higher than those of other methods.

Dynamic Contrast Enhanced Study in Multiparametric Examination of the Prostate—Can We Make Better Use of It?

We sought to investigate whether quantitative parameters from a dynamic contrast-enhanced study can be used to differentiate cancer from normal tissue and to determine a cut-off value of specific parameters that can predict malignancy more accurately, compared to the obturator internus muscle as a reference tissue. This retrospective study included 56 patients with biopsy proven prostate cancer (PCa) after multiparametric magnetic resonance imaging (mpMRI), with a total of 70 lesions; 39 were located in the peripheral zone, and 31 in the transition zone. The quantitative parameters for all patients were calculated in the detected lesion, morphologically normal prostate tissue and the obturator internus muscle. Increase in the K_trans value was determined in lesion-to-muscle ratio by 3.974368, which is a cut-off value to differentiate between prostate cancer and normal prostate tissue, with specificity of 72.86% and sensitivity of 91.43%. We introduced a model to detect prostate cancer that combines K_trans lesion-to-muscle ratio value and iAUC lesion-to-muscle ratio value, which is of higher accuracy compared to individual variables. Based on this model, we identified the optimal cut-off value with 100% sensitivity and 64.28% specificity. The use of quantitative DCE pharmacokinetic parameters compared to the obturator internus muscle as reference tissue leads to higher diagnostic accuracy for prostate cancer detection.

Blood oxygenation level-dependent MRI at 3T for differentiating prostate cancer from benign tissue: a preliminary experience

OBJECTIVE

Blood oxygenation-level dependent (BOLD) MRI may identify or quantify the regional distribution of hypoxia within a tumor. We aimed to evaluate the feasibility of BOLD MRI at 3 T in differentiating prostate cancer from benign tissue.

METHODS

A total of 145 patients with biopsy-proven prostate cancer underwent BOLD MRI at 3 T. BOLD MRI was performed using a multiple fast field echo sequence to acquire 12 T₂*-weighted images. The R2* value (rate of relaxation, s^-1) was measured in the index tumor, and benign peripheral (PZ) and transition zone (TZ), and the results were compared. The variability of R2* measurements was evaluated.

RESULTS

Tumor R2* values (25.95 s^-1) were significantly different from the benign PZ (27.83 s^-1) and benign TZ (21.66 s^-1) (p < 0.001). For identifying the tumor, the area under the receiver operating characteristic of R2* was 0.606, with an optimal cut-off value of 22.8 s^-1 resulting in 73.8% sensitivity and 52% specificity. In the Bland-Altman test, the mean differences in R2* values were 8.5% for tumors, 13.3% for benign PZ, and 6.8% for benign TZ. No associations between tumor R2* value and Gleason score, age, prostate volume, prostate-specific antigen, or tumor size.

CONCLUSION

BOLD MRI at 3 T appears to be a feasible tool for differentiating between prostate cancer and benign tissue. However, further studies are required for a direct clinical application.

ADVANCES IN KNOWLEDGE

The R2* values are significantly different among prostate cancer, benign PZ, and benign TZ.

Relationship between Apparent Diffusion Coefficient Distribution and Cancer Grade in Prostate Cancer and Benign Prostatic Hyperplasia

The aim of this paper was to assess the associations between prostate cancer aggressiveness and histogram-derived apparent diffusion coefficient (ADC) parameters and determine which ADC parameters may help distinguish among stromal hyperplasia (SH), glandular hyperplasia (GH), and low-grade, intermediate-grade, and high-grade prostate cancers. The mean, median, minimum, maximum, and 10th and 25th percentile ADC values were determined from the ADC histogram and compared among two benign prostate hyperplasia (BPH) groups and three Gleason score (GS) groups. Seventy lesions were identified in 58 patients who had undergone proctectomy. Thirty-nine lesions were prostate cancers (GS 6 = 7 lesions, GS 7 = 19 lesions, GS 8 = 11 lesions, GS 9 = 2 lesions), and thirty-one lesions were BPH (SH = 15 lesions, GH = 16 lesions). There were statistically significant differences in 10th percentile and 25th percentile ADC values when comparing GS 6 to GS 7 (p < 0.05). The 10th percentile ADC values yielded the highest area under the curve (AUC). Tenth and 25th percentile ADCs can be used to more accurately differentiate lesions with GS 6 from those with GS 7 than other ADC parameters. Our data indicate that the major challenge with ADC mapping is to differentiate between SH and GS 6, and SH and GS 7.

Pulmonary MRI Radiomics and Machine Learning: Effect of Intralesional Heterogeneity on Classification of Lesion

RATIONALE AND OBJECTIVES

To investigate the effect of intralesional heterogeneity on differentiating benign and malignant pulmonary lesions, quantitative magnetic resonance imaging (MRI) radiomics, and machine learning methods were adopted.

MATERIALS AND METHODS

A total of 176 patients with multiparametric MRI were involved in this exploratory study. To investigate the effect of intralesional heterogeneity on lesion classification, a radiomics model called tumor heterogeneity model was developed and compared to the conventional radiomics model based on the entire tumor. In tumor heterogeneity model, each lesion was divided into five sublesions depending on the spatial location through clustering algorithm. From the five sublesions in multi-parametric MRI sequences, 1100 radiomics features were extracted. The recursive feature elimination method was employed to select features and support vector machine classifier was used to distinguish benign and malignant lesion. The performance of classification was evaluated with the receiver operating characteristic curve and the area under the curve (AUC) was the figure of merit. The 3-fold cross-validation (CV) with and without nesting was used to validate the model, respectively.

RESULTS

The tumor heterogeneity model (AUC = 0.74 ± 0.04 and 0.90 ± 0.03, CV with and without nesting, respectively) outperforms conventional model (AUC = 0.68 ± 0.04 and 0.87 ± 0.03). The difference between the two models is statistically significant (p = 0.03) for lesions greater than 18.80 cm³.

CONCLUSION

Intralesional heterogeneity influences the classification of pulmonary lesions. The tumor heterogeneity model tends to perform better than conventional radiomics model.

Integrative Machine Learning Prediction of Prostate Biopsy Results From Negative Multiparametric MRI

BACKGROUND

Multiparametric MRI (mpMRI) is commonly recommended as a triage test prior to any prostate biopsy. However, there exists limited consensus on which patients with a negative prostate mpMRI could avoid prostate biopsy.

PURPOSE

To identify which patient could safely avoid prostate biopsy when the prostate mpMRI is negative, via a radiomics-based machine learning approach.

STUDY TYPE

Retrospective.

SUBJECTS

Three hundred thirty patients with negative prostate 3T mpMRI between January 2016 and December 2018 were included.

FIELD STRENGTH/SEQUENCE

A 3.0 T/T2-weighted turbo spin echo (TSE) imaging (T₂ WI) and diffusion-weighted imaging (DWI).

ASSESSMENT

The integrative machine learning (iML) model was trained to predict negative prostate biopsy results, utilizing both radiomics and clinical features. The final study cohort comprised 330 consecutive patients with negative mpMRI (PI-RADS < 3) who underwent systematic transrectal ultrasound-guided (TRUS) or MR-ultrasound fusion (MRUS) biopsy within 6 months. A secondary analysis of biopsy naïve subcohort (n = 227) was also conducted.

STATISTICAL TESTS

The Mann-Whitney U test and Chi-Squared test were utilized to evaluate the significance of difference of clinical features between prostate biopsy positive and negative groups. The model performance was validated using leave-one-out cross-validation (LOOCV) and measured by AUC, sensitivity, specificity, and negative predictive value (NPV).

RESULTS

Overall, 306/330 (NPV 92.7%) of the final study cohort patients had negative biopsies, and 207/227 (NPV 91.2%) of the biopsy naïve subcohort patients had negative biopsies. Our iML model achieved NPVs of 98.3% and 98.0% for the study cohort and subcohort, respectively, superior to prostate-specific antigen density (PSAD)-based risk assessment with NPVs of 94.9% and 93.9%, respectively.

DATA CONCLUSION

The proposed iML model achieved high performance in predicting negative prostate biopsy results for patients with negative mpMRI. With improved NPVs, the proposed model can be used to stratify patients who in whom we might obviate biopsies, thus reducing the number of unnecessary biopsies.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Identification of prostate cancer using multiparametric MR imaging characteristics of prostate tissues referenced to whole mount histopathology

In this study, the objective was to characterize the MR signatures of the various benign prostate tissues and to differentiate them from cancer. Data was from seventy prostate cancer patients who underwent multiparametric MRI (mpMRI) and subsequent prostatectomy. The scans included T2-weighted imaging (T2W), diffusion weighted imaging, dynamic contrast-enhanced MRI (DCE MRI), and MR spectroscopic imaging. Histopathology tissue information was translated to MRI images. The mpMRI parameters were characterized separately per zone and by tissue type. The tissues were ordered according to trends in tissue parameter means. The peripheral zone tissue order was cystic atrophy, high grade prostatic intraepithelial neoplasia (HGPIN), normal, atrophy, inflammation, and cancer. Decreasing values for tissue order were exhibited by ADC (1.8 10^-3 mm²/s to 1.2 10^-3 mm²/s) and T2W intensity (3447 to 2576). Increasing values occurred for DCE MRI peak (143% to 157%), DCE MRI slope (101%/min to 169%/min), fractional anisotropy (FA) (0.16 to 0.19), choline (7.2 to 12.2), and choline / citrate (0.3 to 0.9). The transition zone tissue order was cystic atrophy, mixed benign prostatic hyperplasia (BPH), normal, atrophy, inflammation, stroma, anterior fibromuscular stroma, and cancer. Decreasing values occurred for ADC (1.6 10^-3 mm²/s to 1.1 10^-3 mm²/s) and T2W intensity (2863 to 2001). Increasing values occurred for DCE MRI peak (143% to 150%), DCE MRI slope (101%/min to 137%/min), FA (0.18 to 0.25), choline (7.9 to 11.7), and choline / citrate (0.3 to 0.7). Logistic regression was used to create parameter model fits to differentiate cancer from benign prostate tissues. The fits achieved AUCs ≥0.91. This study quantified the mpMRI characteristics of benign prostate tissues and demonstrated the capability of mpMRI to discriminate among benign as well as cancer tissues, potentially aiding future discrimination of cancer from benign confounders.

Validation of Prostate Tissue Composition by Using Hybrid Multidimensional MRI: Correlation with Histologic Findings

Background Tissue estimates obtained by using microstructure imaging techniques, such as hybrid multidimensional (HM) MRI, may improve prostate cancer diagnosis but require histologic validation. Purpose To validate prostate tissue composition measured by using HM MRI, with quantitative histologic evaluation from whole-mount prostatectomy as the reference standard. Materials and Methods In this HIPAA-compliant study, from December 2016 to July 2018, prospective participants with biopsy-confirmed prostate cancer underwent 3-T MRI before radical prostatectomy. Axial HM MRI was performed with all combinations of echo times (57, 70, 150, and 200 msec) and b values (0, 150, 750, and 1500 sec/mm²). Data were fitted by using a three-compartment signal model to generate volumes for each tissue component (stroma, epithelium, lumen). Quantitative histologic evaluation was performed to calculate volume fractions for each tissue component for regions of interest corresponding to MRI. Tissue composition measured by using HM MRI and quantitative histologic evaluation were compared (paired t test) and correlated (Pearson correlation coefficient), and agreement (concordance correlation) was assessed. Receiver operating characteristic curve analysis for cancer diagnosis was performed. Results Twenty-five participants (mean age, 60 years ± 7 [standard deviation]; 30 cancers and 45 benign regions of interest) were included. Prostate tissue composition measured with HM MRI and quantitative histologic evaluation did not differ (stroma, 45% ± 11 vs 44% ± 11 [P = .23]; epithelium, 31% ± 15 vs 34% ± 15 [P = .08]; and lumen, 24% ± 13 vs 22% ± 11 [P = .80]). Between HM MRI and histologic evaluation, there was excellent correlation (Pearson r: overall, 0.91; stroma, 0.82; epithelium, 0.93; lumen, 0.90 [all P < .05]) and agreement (concordance correlation coefficient: overall, 0.91; stroma, 0.81; epithelium, 0.90; and lumen, 0.87). High areas under the receiver operating characteristic curve obtained with HM MRI (0.96 for epithelium and 0.94 for lumen, P < .001) and histologic evaluation (0.94 for epithelium and 0.88 for lumen, P < .001) were found for differentiation between benign tissue and prostate cancer. Conclusion Tissue composition measured by using hybrid multidimensional MRI had excellent correlation with quantitative histologic evaluation as the reference standard. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Muglia in this issue.

Combined Transfer Learning and Test-Time Augmentation Improves Convolutional Neural Network-Based Semantic Segmentation of Prostate Cancer from Multi-Parametric MR Images

PURPOSE

Multiparametric MRI (mp-MRI) is a widely used tool for diagnosing and staging prostate cancer. The purpose of this study was to evaluate whether transfer learning, unsupervised pre-training and test-time augmentation significantly improved the performance of a convolutional neural network (CNN) for pixel-by-pixel prediction of cancer vs. non-cancer using mp-MRI datasets.

METHODS

154 subjects undergoing mp-MRI were prospectively recruited, 16 of whom subsequently underwent radical prostatectomy. Logistic regression, random forest and CNN models were trained on mp-MRI data using histopathology as the gold standard. Transfer learning, unsupervised pre-training and test-time augmentation were used to boost CNN performance. Models were evaluated using Dice score and area under the receiver operating curve (AUROC) with leave-one-subject-out cross validation. Permutation feature importance testing was performed to evaluate the relative value of each MR contrast to CNN model performance. Statistical significance (p<0.05) was determined using the paired Wilcoxon signed rank test with Benjamini-Hochberg correction for multiple comparisons.

RESULTS

Baseline CNN outperformed logistic regression and random forest models. Transfer learning and unsupervised pre-training did not significantly improve CNN performance over baseline; however, test-time augmentation resulted in significantly higher Dice scores over both baseline CNN and CNN plus either of transfer learning or unsupervised pre-training. The best performing model was CNN with transfer learning and test-time augmentation (Dice score of 0.59 and AUROC of 0.93). The most important contrast was apparent diffusion coefficient (ADC), followed by Ktrans and T2, although each contributed significantly to classifier performance.

CONCLUSIONS

The addition of transfer learning and test-time augmentation resulted in significant improvement in CNN segmentation performance in a small set of prostate cancer mp-MRI data. Results suggest that these techniques may be more broadly useful for the optimization of deep learning algorithms applied to the problem of semantic segmentation in biomedical image datasets. However, further work is needed to improve the generalizability of the specific model presented herein.

Differentiating prostate cancer from benign prostatic hyperplasia using whole-lesion histogram and texture analysis of diffusion- and T2-weighted imaging

Background

To explore the usefulness of analyzing histograms and textures of apparent diffusion coefficient (ADC) maps and T2-weighted (T2W) images to differentiate prostatic cancer (PCa) from benign prostatic hyperplasia (BPH) using histopathology as the reference.

Methods

Ninety patients with PCa and 112 patients with BPH were included in this retrospective study. Differences in whole-lesion histograms and texture parameters of ADC maps and T2W images between PCa and BPH patients were evaluated using the independent samples t-test. The diagnostic performance of ADC maps and T2W images in being able to differentiate PCa from BPH was assessed using receiver operating characteristic (ROC) curves.

Results 

The mean, median, 5^th, and 95^th percentiles of ADC values in images from PCa patients were significantly lower than those from BPH patients (p < 0.05). Significant differences were observed in the means, standard deviations, medians, kurtosis, skewness, and 5^th percentile values of T2W image between PCa and BPH patients (p < 0.05). The ADC_5th showed the largest AUC (0.906) with a sensitivity of 83.3 % and specificity of 89.3 %. The diagnostic performance of the T2W image histogram and texture analysis was moderate and had the largest AUC of 0.634 for T2W_Kurtosis with a sensitivity and specificity of 48.9% and 79.5 %, respectively. The diagnostic performance of the combined ADC_5th & T2W_Kurtosis parameters was also similar to that of the ADC_5th & ADC_{Diff−Variance}.

Conclusions

Histogram and texture parameters derived from the ADC maps and T2W images for entire prostatic lesions could be used as imaging biomarkers to differentiate PCa and BPH biologic characteristics, however, histogram parameters outperformed texture parameters in the diagnostic performance.

Making Radiomics More Reproducible across Scanner and Imaging Protocol Variations: A Review of Harmonization Methods

Radiomics converts medical images into mineable data via a high-throughput extraction of quantitative features used for clinical decision support. However, these radiomic features are susceptible to variation across scanners, acquisition protocols, and reconstruction settings. Various investigations have assessed the reproducibility and validation of radiomic features across these discrepancies. In this narrative review, we combine systematic keyword searches with prior domain knowledge to discuss various harmonization solutions to make the radiomic features more reproducible across various scanners and protocol settings. Different harmonization solutions are discussed and divided into two main categories: image domain and feature domain. The image domain category comprises methods such as the standardization of image acquisition, post-processing of raw sensor-level image data, data augmentation techniques, and style transfer. The feature domain category consists of methods such as the identification of reproducible features and normalization techniques such as statistical normalization, intensity harmonization, ComBat and its derivatives, and normalization using deep learning. We also reflect upon the importance of deep learning solutions for addressing variability across multi-centric radiomic studies especially using generative adversarial networks (GANs), neural style transfer (NST) techniques, or a combination of both. We cover a broader range of methods especially GANs and NST methods in more detail than previous reviews.

IVIM Parameters on MRI Could Predict ISUP Risk Groups of Prostate Cancers on Radical Prostatectomy

Purpose

To elucidate the usefulness of intravoxel incoherent motion (IVIM)/apparent diffusion coefficient (ADC) parameters in preoperative risk stratification using International Society of Urological Pathology (ISUP) grades.

Materials and Methods

Forty-five prostate cancer (PCa) patients undergoing radical prostatectomy (RP) after prostate multiparametric magnetic resonance imaging (mpMRI) were included. The ISUP grades were categorized into low-risk (I-II) and high-risk (III-V) groups, and the concordance between the preoperative and postoperative grades was analyzed. The largest region of interest (ROI) of the dominant tumor on each IVIM/ADC image was delineated to obtain its histogram values (i.e., minimum, mean, and kurtosis) of diffusivity (D), pseudodiffusivity (D*), perfusion fraction (PF), and ADC. Multivariable logistic regression analysis of the IVIM/ADC parameters without and with preoperative ISUP grades were performed to identify predictors for the postoperative high-risk group.

Results

Thirty-two (71.1%) of 45 patients had concordant preoperative and postoperative ISUP grades. D_mean, D*_kurtosis, PF_kurtosis, ADC_min, and ADC_mean were significantly associated with the postoperative ISUP risk group (all p < 0.05). D_mean and D*_kurtosis (model I, both p < 0.05) could predict the postoperative ISUP high-risk group with an area under the curve (AUC) of 0.842 and a 95% confidence interval (CI) of 0.726-0.958. The addition of D*_kurtosis to the preoperative ISUP grade (model II) may enhance prediction performance, with an AUC of 0.907 (95% CI 0.822-0.992).

Conclusions

The postoperative ISUP risk group could be predicted by D_mean and D*_kurtosis from mpMRI, especially D*_kurtosis. Obtaining the biexponential IVIM parameters is important for better risk stratification for PCa.

Diagnostic accuracy of high b-value diffusion weighted imaging for patients with prostate cancer: a diagnostic comprehensive analysis

We performed a meta-analysis to assess the diagnostic accuracy of high b-value diffusion-weighted imaging for patients with prostate cancer. A comprehensive literature search of the PubMed, Excerpta Medica Database, Cochrane Library, China National Knowledge Infrastructure, China Biology Medicine disc, and Wanfang databases from January 1, 1995, to April 30, 2021, was conducted. The quality of the retrieved papers was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2. The sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and their 95% confidence intervals (CIs) were evaluated using bivariate mixed effects models. A total of twenty-four articles matched the selection criteria and were finally included after screening the titles, abstracts, and full texts of 641 initial articles. The pooled sensitivity and specificity (95% CI) were 0.84 (0.80-0.87) and 0.87 (0.81-0.91), respectively. The pooled positive and negative likelihood ratios (95% CI) were 6.4 (4.4-9.3) and 0.19 (0.16-0.23), respectively. The diagnostic odds ratio was 34 (95% CI: 22-51). The area under the summary receiver operator characteristic curve was 0.91 (95% CI: 0.88-0.93). Subgroup analysis presents similar results. The diagnostic accuracy of high b-value diffusion-weighted imaging was similarly high in the qualitative and quantitative evaluation of prostate cancer.

Value of MRI texture analysis for predicting new Gleason grade group

OBJECTIVES

To explore the potential value of multiparametric magnetic resonance imaging (mpMRI) texture analysis (TA) to predict new Gleason Grade Group (GGG).

METHODS

Fifty-eight lesions of fifty patients who underwent mpMRI scanning, including T₂-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) prior to trans-rectal ultrasound (TRUS)-guided core prostate biopsy, were retrospectively enrolled. TA parameters were obtained by the postprocessing software, and each lesion was assigned to its corresponding GGG. TA parameters derived from T2WI and DWI were statistically analyzed in detail.

RESULTS

Energy, inertia, and correlation derived from apparent diffusion coefficient (ADC) maps and T2WI had a statistically significant difference among the five groups. Kurtosis, energy, inertia, correlation on ADC maps and Energy, inertia on T2WI were moderately related to the GGG trend. ADC-energy and T2-energy were significant independent predictors of the GGG trend. ADC-energy, T2WI-energy, and T2WI-correlation had a statistically significant difference between GGG1 and GGG2-5. ADC-energy were significant independent predictors of the GGG1. ADC-energy, T2WI-energy, and T2WI-correlation showed satisfactory diagnostic efficiency of GGG1 (area under the curve (AUC) 84.6, 74.3, and 83.5%, respectively), and ADC-energy showed excellent sensitivity and specificity (88.9 and 95.1%, respectively).

CONCLUSION

TA parameters ADC-energy and T2-energy played an important role in predicting GGG trend. Both ADC-energy and T2-correlation produced a high diagnostic power of GGG1, and ADC-energy was perfect predictors of GGG1.

ADVANCES IN KNOWLEDGE

TA parameters were innovatively used to predict new GGG trend, and the predictive factors of GGG1 were screen out.

Medical image segmentation and reconstruction of prostate tumor based on 3D AlexNet

BACKGROUND

Prostate cancer is a disease with a high incidence of tumors in men. Due to the long incubation time and insidious condition, early diagnosis is difficult; especially imaging diagnosis is more difficult. In actual clinical practice, the method of manual segmentation by medical experts is mainly used, which is time-consuming and labor-intensive and relies heavily on the experience and ability of medical experts. The rapid, accurate and repeatable segmentation of the prostate area is still a challenging problem. It is important to explore the automated segmentation of prostate images based on the 3D AlexNet network.

METHOD

Taking the medical image of prostate cancer as the entry point, the three-dimensional data is introduced into the deep learning convolutional neural network. This paper proposes a 3D AlexNet method for the automatic segmentation of prostate cancer magnetic resonance images, and the general network ResNet 50, Inception -V4 compares network performance.

RESULTS

Based on the training samples of magnetic resonance images of 500 prostate cancer patients, a set of 3D AlexNet with simple structure and excellent performance was established through adaptive improvement on the basis of classic AlexNet. The accuracy rate was as high as 0.921, the specificity was 0.896, and the sensitivity It is 0.902 and the area under the receiver operating characteristic curve (AUC) is 0.964. The Mean Absolute Distance (MAD) between the segmentation result and the medical expert's gold standard is 0.356 mm, and the Hausdorff distance (HD) is 1.024 mm, the Dice similarity coefficient is 0.9768.

CONCLUSION

The improved 3D AlexNet can automatically complete the structured segmentation of prostate magnetic resonance images. Compared with traditional segmentation methods and depth segmentation methods, the performance of the 3D AlexNet network is superior in terms of training time and parameter amount, or network performance evaluation. Compared with the algorithm, it proves the effectiveness of this method.

Quantitative tumor heterogeneity MRI profiling improves machine learning-based prognostication in patients with metastatic colon cancer

OBJECTIVES

Intra-tumor heterogeneity has been previously shown to be an independent predictor of patient survival. The goal of this study is to assess the role of quantitative MRI-based measures of intra-tumor heterogeneity as predictors of survival in patients with metastatic colorectal cancer.

METHODS

In this IRB-approved retrospective study, we identified 55 patients with stage 4 colon cancer with known hepatic metastasis on MRI. Ninety-four metastatic hepatic lesions were identified on post-contrast images and manually volumetrically segmented. A heterogeneity phenotype vector was extracted from each lesion. Univariate regression analysis was used to assess the contribution of 110 extracted features to survival prediction. A random forest-based machine learning technique was applied to the feature vector and to the standard prognostic clinical and pathologic variables. The dataset was divided into a training and test set at a ratio of 4:1. ROC analysis and confusion matrix analysis were used to assess classification performance.

RESULTS

Mean survival time was 39 ± 3.9 months for the study population. A total of 22 texture features were associated with patient survival (p < 0.05). The trained random forest machine learning model that included standard clinical and pathological prognostic variables resulted in an area under the ROC curve of 0.83. A model that adds imaging-based heterogeneity features to the clinical and pathological variables resulted in improved model performance for survival prediction with an AUC of 0.94.

CONCLUSIONS

MRI-based texture features are associated with patient outcomes and improve the performance of standard clinical and pathological variables for predicting patient survival in metastatic colorectal cancer.

KEY POINTS

• MRI-based tumor heterogeneity texture features are associated with patient survival outcomes. • MRI-based tumor texture features complement standard clinical and pathological variables for prognosis prediction in metastatic colorectal cancer. • Agglomerative hierarchical clustering shows that patient survival outcomes are associated with different MRI tumor profiles.

Multiparametric magnetic resonance imaging, 68Ga prostate-specific membrane antigen positron emission tomography–Computed tomography, and respective quantitative parameters in detection and localization of clinically significant prostate cancer in intermediate- and high-risk group patients: An Indian demographic study

Objectives:

The objective of this study was to evaluate the diagnostic accuracy of multiparametric magnetic resonance imaging (mpMRI) and ⁶⁸Ga prostate-specific membrane antigen positron emission tomography–computed tomography (PSMA PET-CT) and respective quantitative parameters (K^trans – influx rate contrast, K_ep – efflux rate constant, ADC – apparent diffusion coefficient, and SUVmax ratio – prostate SUVmax to background SUVmax ratio) in detection and localization of clinically significant prostate cancer (CSPCa) in D’Amico intermediate- and high-risk group patients (prostate-specific antigen [PSA] >10 ng/ml).

Methodology:

The study included thirty-three consecutive adult men with serum prostate specific antigen >10ng/ml, and systematic 12 core prostate biopsy proven prostate cancer. All the 33 patients, were evaluated with mpMRI, and ⁶⁸Ga PSMA PET-CT. The biopsy specimens and imaging were evaluated for 12 sectors per prostate by a predetermined scheme.

Results:

MpMRI Prostate Imaging Reporting and Data System version 2 (PI-RADS v2) score ≥3 showed higher sensitivity than ⁶⁸Ga PSMA PET-CT (96.3% vs. 82.4%), with similar specificity (54.5% vs. 54.5%) (n = 33 patients, 396 sectors). Combined use of MRI and ⁶⁸Ga PSMA PET-CT in parallel increased sensitivity (99.5%) and NPV (98.7%) for detection of CSPCa and combined use of MRI and ⁶⁸Ga PSMA PET-CT in series increased specificity (71.8%) and PPV (71.5%) (n = 33 patients, 396 sectors). ADC showed a strong negative correlation with Gleason score (r = −0.77), and the highest discriminative ability for detection and localization of CSPCa (area under curve [AUC]: 0.91), followed by K^trans (r = 0.74; AUC: 0.89), PI-RADS (0.73; 0.86), SUVmax ratio (0.49; 0.74), and K_ep (0.24; 0.66).

Conclusion:

MpMRI PI-RADS v2 score and ⁶⁸Ga PSMA PET-CT (individually as well as in combination) are reliable tool for detection and localization of CSPCa. Quantitative MRI and ⁶⁸Ga PSMA PET-CT parameters have potential to predict Gleason score and detect CSPCa.

Semi-automated PIRADS scoring via mpMRI analysis

Purpose: Prostate cancer (PCa) is the most common solid organ cancer and second leading cause of death in men. Multiparametric magnetic resonance imaging (mpMRI) enables detection of the most aggressive, clinically significant PCa (csPCa) tumors that require further treatment. A suspicious region of interest (ROI) detected on mpMRI is now assigned a Prostate Imaging-Reporting and Data System (PIRADS) score to standardize interpretation of mpMRI for PCa detection. However, there is significant inter-reader variability among radiologists in PIRADS score assignment and a minimal input semi-automated artificial intelligence (AI) system is proposed to harmonize PIRADS scores with mpMRI data. Approach: The proposed deep learning model (the seed point model) uses a simulated single-click seed point as input to annotate the lesion on mpMRI. This approach is in contrast to typical medical AI-based approaches that require annotation of the complete lesion. The mpMRI data from 617 patients used in this study were prospectively collected at a major tertiary U.S. medical center. The model was trained and validated to classify whether an mpMRI image had a lesion with a PIRADS score greater than or equal to PIRADS 4. Results: The model yielded an average receiver-operator characteristic (ROC) area under the curve (ROC-AUC) of 0.704 over a 10-fold cross-validation, which is significantly higher than the previously published benchmark. Conclusions: The proposed model could aid in PIRADS scoring of mpMRI, providing second reads to promote quality as well as offering expertise in environments that lack a radiologist with training in prostate mpMRI interpretation. The model could help identify tumors with a higher PIRADS for better clinical management and treatment of PCa patients at an early stage.

Harmonization strategies for multicenter radiomics investigations

Carrying out large multicenter studies is one of the key goals to be achieved towards a faster transfer of the radiomics approach in the clinical setting. This requires large-scale radiomics data analysis, hence the need for integrating radiomic features extracted from images acquired in different centers. This is challenging as radiomic features exhibit variable sensitivity to differences in scanner model, acquisition protocols and reconstruction settings, which is similar to the so-called 'batch-effects' in genomics studies. In this review we discuss existing methods to perform data integration with the aid of reducing the unwanted variation associated with batch effects. We also discuss the future potential role of deep learning methods in providing solutions for addressing radiomic multicentre studies.

Differentiation of prostate cancer and benign prostatic hyperplasia: comparisons of the histogram analysis of intravoxel incoherent motion and monoexponential model with in-bore MR-guided biopsy as pathological reference

PURPOSE

To evaluate the diagnostic performance of histogram analysis of intravoxel incoherent motion (IVIM) parameters for differentiating prostate cancer (PCa) from benign prostatic hyperplasia (BPH), and compare with the monoexponential model, with in-bore MR-guided biopsy as pathological reference.

METHODS

Thirty patients were included in this study. DWI images were processed with Matlab R2015b software by IVIM and monoexponential model for quantitation of diffusion coefficient (D), pseudo diffusion coefficient (D*), perfusion fraction (f), and apparent diffusion coefficient (ADC). The multiparametric data were compared between PCa and BPH group. Correlations between parameters and Gleason scores of PCa were assessed with Spearman rank test. ROC analysis was used to evaluate and compare the diagnostic ability of each parameter for discriminating PCa from BPH. Logistic regression model was used to evaluate the diagnostic performance of combination of different histogram parameters.

RESULTS

Sixteen PCa lesions and 20 BPH nodules were analyzed in this study. For IVIM-derived D, the histogram mean, 75th, 90th, and max of PCa were significantly lower than BPH. PCa had significantly lower min and 10th D* than BPH. For f, histogram mean, min, 10th, 25th, 50th, 75th, 90th, max and skew showed significant differences between PCa and BPH. For ADC, PCa were significantly lower than BPH in terms of histogram mean, min, 10th, 25th, 50th, 75th, 90th, max and kurtosis. Histogram mean D and min, 25th D* show significantly negative correlation with Gleason score (r = - 0.582, - 0.534, - 0.554, respectively). Histogram max D and mean f and min ADC showed higher diagnostic performance than other parameters (AUC = 0.925, 0.881, 0.969, respectively). The IVIM model (combined with max D, min D* and mean f) (AUC = 0.950 [0.821, 0.995]) didn't show significant difference from the monoexponential model (AUC = 0.969 [0.849, 0.999], p = 0.23). Besides, combination of the IVIM and monoexponential model didn't improve diagnostic performance compared with the single model (p = 0.362 and 0.763, respectively).

CONCLUSIONS

Histogram analyses of IVIM and monoexponential model were both useful methods for discriminating PCa from BPH. The diagnostic performance of IVIM model seemed to be not superior to that of monoexponential model. Combination of IVIM and monoexponential model did not add significant information to the single model alone.

The role of radiomics in prostate cancer radiotherapy

"Radiomics," as it refers to the extraction and analysis of a large number of advanced quantitative radiological features from medical images using high-throughput methods, is perfectly suited as an engine for effectively sifting through the multiple series of prostate images from before, during, and after radiotherapy (RT). Multiparametric (mp)MRI, planning CT, and cone beam CT (CBCT) routinely acquired throughout RT and the radiomics pipeline are developed for extraction of thousands of variables. Radiomics data are in a format that is appropriate for building descriptive and predictive models relating image features to diagnostic, prognostic, or predictive information. Prediction of Gleason score, the histopathologic cancer grade, has been the mainstay of the radiomic efforts in prostate cancer. While Gleason score (GS) is still the best predictor of treatment outcome, there are other novel applications of quantitative imaging that are tailored to RT. In this review, we summarize the radiomics efforts and discuss several promising concepts such as delta-radiomics and radiogenomics for utilizing image features for assessment of the aggressiveness of prostate cancer and its outcome. We also discuss opportunities for quantitative imaging with the advance of instrumentation in MRI-guided therapies.

Computer-Aided Diagnosis in Multiparametric MRI of the Prostate: An Open-Access Online Tool for Lesion Classification with High Accuracy

Computer-aided diagnosis (CADx) approaches could help to objectify reporting on prostate mpMRI, but their use in many cases is hampered due to common-built algorithms that are not publicly available. The aim of this study was to develop an open-access CADx algorithm with high accuracy for classification of suspicious lesions in mpMRI of the prostate. This retrospective study was approved by the local ethics commission, with waiver of informed consent. A total of 124 patients with 195 reported lesions were included. All patients received mpMRI of the prostate between 2014 and 2017, and transrectal ultrasound (TRUS)-guided and targeted biopsy within a time period of 30 days. Histopathology of the biopsy cores served as a standard of reference. Acquired imaging parameters included the size of the lesion, signal intensity (T2w images), diffusion restriction, prostate volume, and several dynamic parameters along with the clinical parameters patient age and serum PSA level. Inter-reader agreement of the imaging parameters was assessed by calculating intraclass correlation coefficients. The dataset was stratified into a train set and test set (156 and 39 lesions in 100 and 24 patients, respectively). Using the above parameters, a CADx based on an Extreme Gradient Boosting algorithm was developed on the train set, and tested on the test set. Performance optimization was focused on maximizing the area under the Receiver Operating Characteristic curve (ROC_AUC). The algorithm was made publicly available on the internet. The CADx reached an ROC_AUC of 0.908 during training, and 0.913 during testing (p = 0.93). Additionally, established rule-in and rule-out criteria allowed classifying 35.8% of the malignant and 49.4% of the benign lesions with error rates of <2%. All imaging parameters featured excellent inter-reader agreement. This study presents an open-access CADx for classification of suspicious lesions in mpMRI of the prostate with high accuracy. Applying the provided rule-in and rule-out criteria might facilitate to further stratify the management of patients at risk.

Dynamic contrast-enhanced (DCE) MR imaging: the role of qualitative and quantitative parameters for evaluating prostate tumors stratified by Gleason score and PI-RADS v2

PURPOSE

To investigate the role of qualitative and quantitative DCE-MRI parameters in prostate cancer (PCa) stratified by whole-mount histopathology (WMHP) Gleason score (GS) and PI-RADSv2.

METHODS

This retrospective study included 323 PCa tumors in 254 men, who underwent 3T MRI prior to prostatectomy, 7/2009-12/2016. Qualitative DCE curve types included type 1 (progressive), type 2 (plateau) and type 3 (washout). Quantitative DCE-MRI pharmacokinetic (PK) parameters included K^trans (influx volume transfer coefficient), K_ep (efflux reflux rate constant) and iAUC (initial area under the curve). DCE-MRI features of true positive lesions were evaluated for overall, index, transition zone (TZ) and peripheral zone (PZ), based on GS grade (low = 6, high > 6) and PI-RADSv2 score using SPSSv24.

RESULTS

There were 57 (17.6%) low-grade and 266 (82.4%) high-grade PCa lesions. PI-RADSv2 3, 4 and 5 included 106, 120 and 97 lesions, respectively. 251 (77.7%) and 72 (22.3%) lesions were located in PZ and TZ, respectively. High-grade lesions had significantly higher proportion of Type 3 curves compared to low-grade lesions in overall (70.3% vs. 54.4%) and TZ (73.5% vs. 43.5%). As PI-RADSv2 increased, the proportion of type 3 curve significantly increased for overall (80.4-51.9%), index (80.4-54.7%) and PZ (78.7-52.1%) lesions. Among PK parameters, K^trans (0.43 vs 0.32) and iAUC (8.99 vs 6.9) for overall PCa, K^trans (0.43 vs 0.31) and iAUC (9 vs 6.67) for PZ PCa, and iAUC (8.94 vs 7.42) for index PCa were significantly higher for high-grade versus low-grade lesions. Also, K^trans (0.51-0.34), K_ep (1.75-1.29) and iAUC (9.79-7.6) for overall PCa, K^trans (0.53-0.32), K_ep (1.81-1.26) and iAUC (9.83-7.34) for PZ PCa; and K_ep (1.79-1.17) and iAUC (11.3-8.45) for index PCa increased significantly with a higher PI-RADSv2 score.

CONCLUSIONS

The results of study show the possible utility of qualitative and quantitative DCE-MRI parameters for assessment of PCa GS and PI-RADSv2 categorization.

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Significant prostate carcinoma (sPCa) classification based on MRI using radiomics or deep learning approaches has gained much interest, due to the potential application in assisting in clinical decision-making.

OBJECTIVE

To systematically review the literature (i) to determine which algorithms are most frequently used for sPCa classification, (ii) to investigate whether there exists a relation between the performance and the method or the MRI sequences used, (iii) to assess what study design factors affect the performance on sPCa classification, and (iv) to research whether performance had been evaluated in a clinical setting Methods: The databases Embase and Ovid MEDLINE were searched for studies describing machine learning or deep learning classification methods discriminating between significant and nonsignificant PCa on multiparametric MRI that performed a valid validation procedure. Quality was assessed by the modified radiomics quality score. We computed the median area under the receiver operating curve (AUC) from overall methods and the interquartile range.

RESULTS

From 2846 potentially relevant publications, 27 were included. The most frequent algorithms used in the literature for PCa classification are logistic regression (22%) and convolutional neural networks (CNNs) (22%). The median AUC was 0.79 (interquartile range: 0.77-0.87). No significant effect of number of included patients, image sequences, or reference standard on the reported performance was found. Three studies described an external validation and none of the papers described a validation in a prospective clinical trial.

CONCLUSIONS

To unlock the promising potential of machine and deep learning approaches, validation studies and clinical prospective studies should be performed with an established protocol to assess the added value in decision-making.

Prediction of prostate cancer aggressiveness using 18F-Fluciclovine (FACBC) PET and multisequence multiparametric MRI

The aim of this prospective single-institution clinical trial (NCT02002455) was to evaluate the potential of advanced post-processing methods for 18F-Fluciclovine PET and multisequence multiparametric MRI in the prediction of prostate cancer (PCa) aggressiveness, defined by Gleason Grade Group (GGG). 21 patients with PCa underwent PET/CT, PET/MRI and MRI before prostatectomy. DWI was post-processed using kurtosis (ADCk, K), mono- (ADCm), and biexponential functions (f, Dp, Df) while Logan plots were used to calculate volume of distribution (VT). In total, 16 unique PET (VT, SUV) and MRI derived quantitative parameters were evaluated. Univariate and multivariate analysis were carried out to estimate the potential of the quantitative parameters and their combinations to predict GGG 1 vs >1, using logistic regression with a nested leave-pair out cross validation (LPOCV) scheme and recursive feature elimination technique applied for feature selection. The second order rotating frame imaging (RAFF), monoexponential and kurtosis derived parameters had LPOCV AUC in the range of 0.72 to 0.92 while the corresponding value for VT was 0.85. The best performance for GGG prediction was achieved by K parameter of kurtosis function followed by quantitative parameters based on DWI, RAFF and 18F-FACBC PET. No major improvement was achieved using parameter combinations with or without feature selection. Addition of 18F-FACBC PET derived parameters (VT, SUV) to DWI and RAFF derived parameters did not improve LPOCV AUC.

Artificial intelligence in multiparametric prostate cancer imaging with focus on deep-learning methods

Prostate cancer represents today the most typical example of a pathology whose diagnosis requires multiparametric imaging, a strategy where multiple imaging techniques are combined to reach an acceptable diagnostic performance. However, the reviewing, weighing and coupling of multiple images not only places additional burden on the radiologist, it also complicates the reviewing process. Prostate cancer imaging has therefore been an important target for the development of computer-aided diagnostic (CAD) tools. In this survey, we discuss the advances in CAD for prostate cancer over the last decades with special attention to the deep-learning techniques that have been designed in the last few years. Moreover, we elaborate and compare the methods employed to deliver the CAD output to the operator for further medical decision making.

Classification of pulmonary lesion based on multiparametric MRI: utility of radiomics and comparison of machine learning methods

OBJECTIVES

We develop and validate a radiomics model based on multiparametric magnetic resonance imaging (MRI) in the classification of the pulmonary lesion and identify optimal machine learning methods.

MATERIALS AND METHODS

This retrospective analysis included 201 patients (143 malignancies, 58 benign lesions). Radiomics features were extracted from multiparametric MRI, including T2-weighted imaging (T2WI), T1-weighted imaging (TIWI), and apparent diffusion coefficient (ADC) map. Three feature selection methods, including recursive feature elimination (RFE), t test, and least absolute shrinkage and selection operator (LASSO), and three classification methods, including linear discriminate analysis (LDA), support vector machine (SVM), and random forest (RF) were used to distinguish benign and malignant pulmonary lesions. Performance was compared by AUC, sensitivity, accuracy, precision, and specificity. Analysis of performance differences in three randomly drawn cross-validation sets verified the stability of the results.

RESULTS

For most single MR sequences or combinations of multiple MR sequences, RFE feature selection method with SVM classifier had the best performance, followed by RFE with RF. The radiomics model based on multiple sequences showed a higher diagnostic accuracy than single sequence for every machine learning method. Using RFE with SVM, the joint model of T1WI, T2WI, and ADC showed the highest performance with AUC = 0.88 ± 0.02 (sensitivity 83%; accuracy 82%; precision 91%; specificity 79%) in test set.

CONCLUSION

Quantitative radiomics features based on multiparametric MRI have good performance in differentiating lung malignancies and benign lesions. The machine learning method of RFE with SVM is superior to the combination of other feature selection and classifier methods.

KEY POINTS

• Radiomics approach has the potential to distinguish between benign and malignant pulmonary lesions. • Radiomics model based on multiparametric MRI has better performance than single-sequence models. • The machine learning methods RFE with SVM perform best in the current cohort.

Comparison of double inversion recovery magnetic resonance imaging (DIR-MRI) and dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in detection of prostate cancer: A pilot study

INTRODUCTION

DCE-MRI is established for detecting prostate cancer (PCa). However, it requires a gadolinium contrast agent, with potential risks for patients. The application of DIR-MRI is simple and may allow cancer detection without the use of an intravenous contrast agent by differentially nullifying signal from normal and abnormal prostate tissue, creating contrast between the cancer and background normal prostate. In this pilot study we gathered data from DIR-MRI and DCE-MRI of the prostate for an equivalence trial. We also looked at how the DIR-MRI appearance varies with the aggressiveness of PCa.

METHOD

DIR-MRI and DCE-MRI were acquired. The images were assessed by an experienced Consultant Radiologist and a novice reporter (Radiographer). The potential PCa lesions were quantified using a lesion to normal ratio (LNR). Radiological pathological correlation was made to identify the MRI lesions that represented significant PCa. A Wilcoxon sign rank was used to compare DCE-LNR and DIR-LNR for PCa containing lesions. Pearson's correlation was used to look at the relationship between DIR-LNR and PCa grade group (aggressiveness).

RESULTS

DCE-LNR and DIR-LNR were found to be significantly different (Z = -5.910, p < 0.001). However, a significant correlation was found between PCa grade group and DIR-LNR.

CONCLUSION

DIR and DCE sequences are not equivalent and significant cancer is more conspicuous on the DCE sequence. However, DIR-LNR does correlate with PCa aggressiveness.

IMPLICATIONS FOR PRACTICE

With the correlation of PCa grade group with DIR-LNR this may be a useful sequence in evaluation of the prostate; stratifying the risk of there being clinically significant PCa before biopsy is performed. Furthermore, given that DIR-LNR appears to predict PCa aggressiveness DIR might be used as part of a multiparametric MRI protocol designed to avoid biopsy.

[Artificial intelligence and radiomics in MRI-based prostate diagnostics]

CLINICAL/METHODICAL ISSUE

In view of the diagnostic complexity and the large number of examinations, modern radiology is challenged to identify clinically significant prostate cancer (PCa) with high sensitivity and specificity. Meanwhile overdiagnosis and overtreatment of clinically nonsignificant carcinomas need to be avoided.

STANDARD RADIOLOGICAL METHODS

Increasingly, international guidelines recommend multiparametric magnetic resonance imaging (mpMRI) as first-line investigation in patients with suspected PCa.

METHODICAL INNOVATIONS

Image interpretation according to the PI-RADS criteria is limited by interobserver variability. Thus, rapid developments in the field of automated image analysis tools, including radiomics and artificial intelligence (AI; machine learning, deep learning), give hope for further improvement in patient care.

PERFORMANCE

AI focuses on the automated detection and classification of PCa, but it also attempts to stratify tumor aggressiveness according to the Gleason score. Recent studies present good to very good results in radiomics or AI-supported mpMRI diagnosis. Nevertheless, these systems are not widely used in clinical practice.

ACHIEVEMENTS AND PRACTICAL RECOMMENDATIONS

In order to apply these innovative technologies, a growing awareness for the need of structured data acquisition, development of robust systems and an increased acceptance of AI as diagnostic support are needed. If AI overcomes these obstacles, it may play a key role in the quantitative and reproducible image-based diagnosis of ever-increasing prostate MRI examination volumes.

Diagnosis and Grading of Prostate Cancer by Relaxation Maps From Synthetic MRI

BACKGROUND

The interpretation system for prostate MRI is largely based on qualitative image contrast of different tissue types. Therefore, a fast, standardized, and robust quantitative technique is necessary. Synthetic MRI is capable of quantifying multiple relaxation parameters, which might have potential applications in prostate cancer (PCa).

PURPOSE

To investigate the use of quantitative relaxation maps derived from synthetic MRI for the diagnosis and grading of PCa.

STUDY TYPE

Prospective.

SUBJECTS

In all, 94 men with pathologically confirmed PCa or benign pathological changes.

FIELD STRENGTH/SEQUENCE

T₁ -weighted imaging, T₂ -weighted imaging, diffusion-weighted imaging, and synthetic MRI at 3.0T.

ASSESSMENT

Four kinds of tissue types were identified on pathology, including PCa, stromal hyperplasia (SH), glandular hyperplasia (GH), and noncancerous peripheral zone (PZ). PCa foci were grouped as low-grade (LG, Gleason score ≤6) and intermediate/high-grade (HG, Gleason score ≥7). Regions of interest were manually drawn by two radiologists in consensus on parametric maps according to the pathological results.

STATISTICAL TESTS

Independent sample t-test, Mann-Whitney U-test, and receiver operating characteristic curve analysis.

RESULTS

T₁ and T₂ values of PCa were significantly lower than SH (P = 0.015 and 0.002). The differences of T₁ and T₂ values between PCa and noncancerous PZ were also significant (P ≤ 0.006). The area under the curve (AUC) of the apparent diffusion coefficient (ADC) value was significantly higher than T₁ , T₂ , and proton density (PD) values in discriminating PCa from SH and noncancerous PZ (P ≤ 0.025). T₂ , PD, and ADC values demonstrated similar diagnostic performance in discriminating LG from HG PCa (AUC = 0.806 [0.640-0.918], 0.717 [0.542-0.854], and 0.817 [0.652-0.925], respectively; P ≥ 0.535).

DATA CONCLUSION

Relaxation maps derived from synthetic MRI were helpful for discriminating PCa from other benign pathologies. But the overall diagnostic performance was inferior to the ADC values. T₂ , PD, and ADC values performed similarly in discriminating LG from HG PCa lesions.

LEVEL OF EVIDENCE

2 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;52:552-564.

The Utility of Diffusion-Weighted Imaging and Perfusion Magnetic Resonance Imaging Parameters for Detecting Clinically Significant Prostate Cancer

Introduction

To establish the diagnostic performance of the parameters obtained from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion-weighted imaging at 3T in discriminating between non-clinically significant prostate cancers (ncsPCa, Gleason score [GS] < 7) and clinically significant prostate cancers (csPCa, GS ≥ 7) in the peripheral zone.

Materials and Methods

Twenty-six male patients with peripheral zone prostate cancer (PCa) who had undergone 3T multiparametric magnetic resonance imaging (MRI) scan prior to biopsy were included in the study and evaluated retrospectively. The GS was obtained by both standard 12-core transrectal ultrasound guided biopsy and targeted MRI-US fusion biopsy and then confirmed by prostatectomy, if available. For each confirmed tumour focus, DCE-derived quantitative perfusion metrics (Ktrans, Kep, Ve, initial area under the curve [AUC]), the apparent diffusion coefficient (ADC) value, and normalized versions of quantitative metrics were measured and correlated with the GS.

Results

Ktrans had the highest diagnostic accuracy value of 82% among the DCE-MRI parameters (AUC 0.90), and ADC had the strongest diagnostic accuracy value of 87% among the overall parameters (AUC 0.92). The combination of ADC and Ktrans have higher diagnostic performance with the area under the receiver operating characteristic curve being 0.98 (sensitivity 0.94; specificity 0.89; accuracy 0.92) compared to the individual evaluation of each parameter alone. The GS showed strong negative correlations with ADC (r = −0.72) and normalized ADC (r = −0.69) as well as a significant positive correlation with Ktrans (r = 0.69).

Conclusion

The combination of Ktrans and ADC and their normalized versions may help differentiate between ncsPCa from csPCa in the peripheral zone.