Evaluation of a Cascaded Deep Learning-based Algorithm for Prostate Lesion Detection at Biparametric MRI.

Background Multiparametric MRI (mpMRI) improves prostate cancer (PCa) detection compared with systematic biopsy, but its interpretation is prone to interreader variation, which results in performance inconsistency. Artificial intelligence (AI) models can assist in mpMRI interpretation, but large training data sets and extensive model testing are required. Purpose To evaluate a biparametric MRI AI algorithm for intraprostatic lesion detection and segmentation and to compare its performance with radiologist readings and biopsy results. Materials and Methods This secondary analysis of a prospective registry included consecutive patients with suspected or known PCa who underwent mpMRI, US-guided systematic biopsy, or combined systematic and MRI/US fusion-guided biopsy between April 2019 and September 2022. All lesions were prospectively evaluated using Prostate Imaging Reporting and Data System version 2.1. The lesion- and participant-level performance of a previously developed cascaded deep learning algorithm was compared with histopathologic outcomes and radiologist readings using sensitivity, positive predictive value (PPV), and Dice similarity coefficient (DSC). Results A total of 658 male participants (median age, 67 years [IQR, 61-71 years]) with 1029 MRI-visible lesions were included. At histopathologic analysis, 45% (294 of 658) of participants had lesions of International Society of Urological Pathology (ISUP) grade group (GG) 2 or higher. The algorithm identified 96% (282 of 294; 95% CI: 94%, 98%) of all participants with clinically significant PCa, whereas the radiologist identified 98% (287 of 294; 95% CI: 96%, 99%; P = .23). The algorithm identified 84% (103 of 122), 96% (152 of 159), 96% (47 of 49), 95% (38 of 40), and 98% (45 of 46) of participants with ISUP GG 1, 2, 3, 4, and 5 lesions, respectively. In the lesion-level analysis using radiologist ground truth, the detection sensitivity was 55% (569 of 1029; 95% CI: 52%, 58%), and the PPV was 57% (535 of 934; 95% CI: 54%, 61%). The mean number of false-positive lesions per participant was 0.61 (range, 0-3). The lesion segmentation DSC was 0.29. Conclusion The AI algorithm detected cancer-suspicious lesions on biparametric MRI scans with a performance comparable to that of an experienced radiologist. Moreover, the algorithm reliably predicted clinically significant lesions at histopathologic examination. ClinicalTrials.gov Identifier: NCT03354416 © RSNA, 2024 Supplemental material is available for this article.

Multiparametric MRI and 18F-PSMA-1007 PET/CT for the Detection of Clinically Significant Prostate Cancer.

Background Multiparametric MRI (mpMRI) is effective for detecting prostate cancer (PCa); however, there is a high rate of equivocal Prostate Imaging Reporting and Data System (PI-RADS) 3 lesions and false-positive findings. Purpose To investigate whether fluorine 18 (18F) prostate-specific membrane antigen (PSMA) 1007 PET/CT after mpMRI can help detect localized clinically significant PCa (csPCa), particularly for equivocal PI-RADS 3 lesions. Materials and Methods This prospective study included participants with elevated prostate-specific antigen (PSA) levels referred for prostate mpMRI between September 2020 and February 2022. 18F-PSMA-1007 PET/CT was performed within 30 days of mpMRI and before biopsy. PI-RADS category and level of suspicion (LOS) were assessed. PI-RADS 3 or higher lesions at mpMRI and/or LOS 3 or higher lesions at 18F-PSMA-1007 PET/CT underwent targeted biopsies. PI-RADS 2 or lower and LOS 2 or lower lesions were considered nonsuspicious and were monitored during a 1-year follow-up by means of PSA testing. Diagnostic accuracy was assessed, with histologic examination serving as the reference standard. International Society of Urological Pathology (ISUP) grade 2 or higher was considered csPCa. Results Seventy-five participants (median age, 67 years [range, 52-77 years]) were assessed, with PI-RADS 1 or 2, PI-RADS 3, and PI-RADS 4 or 5 groups each including 25 participants. A total of 102 lesions were identified, of which 80 were PI-RADS 3 or higher and/or LOS 3 or higher and therefore underwent targeted biopsy. The per-participant sensitivity for the detection of csPCa was 95% and 91% for mpMRI and 18F-PSMA-1007 PET/CT, respectively, with respective specificities of 45% and 62%. 18F-PSMA-1007 PET/CT was used to correctly differentiate 17 of 26 PI-RADS 3 lesions (65%), with a negative and positive predictive value of 93% and 27%, respectively, for ruling out or detecting csPCa. One additional significant and one insignificant PCa lesion (PI-RADS 1 or 2) were found at 18F-PSMA-1007 PET/CT that otherwise would have remained undetected. Two participants had ISUP 2 tumors without PSMA uptake that were missed at PET/CT. Conclusion 18F-PSMA-1007 PET/CT showed good sensitivity and moderate specificity for the detection of csPCa and ruled this out in 93% of participants with PI-RADS 3 lesions. Clinical trial registration no. NCT04487847 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Turkbey in this issue.

Development and Validation of a Multimodality Model Based on Whole-Slide Imaging and Biparametric MRI for Predicting Postoperative Biochemical Recurrence in Prostate Cancer.

Purpose To develop and validate a machine learning multimodality model based on preoperative MRI, surgical whole-slide imaging (WSI), and clinical variables for predicting prostate cancer (PCa) biochemical recurrence (BCR) following radical prostatectomy (RP). Materials and Methods In this retrospective study (September 2015 to April 2021), 363 male patients with PCa who underwent RP were divided into training (n = 254; median age, 69 years [IQR, 64-74 years]) and testing (n = 109; median age, 70 years [IQR, 65-75 years]) sets at a ratio of 7:3. The primary end point was biochemical recurrence-free survival. The least absolute shrinkage and selection operator Cox algorithm was applied to select independent clinical variables and construct the clinical signature. The radiomics signature and pathomics signature were constructed using preoperative MRI and surgical WSI data, respectively. A multimodality model was constructed by combining the radiomics signature, pathomics signature, and clinical signature. Using Harrell concordance index (C index), the predictive performance of the multimodality model for BCR was assessed and compared with all single-modality models, including the radiomics signature, pathomics signature, and clinical signature. Results Both radiomics and pathomics signatures achieved good performance for BCR prediction (C index: 0.742 and 0.730, respectively) on the testing cohort. The multimodality model exhibited the best predictive performance, with a C index of 0.860 on the testing set, which was significantly higher than all single-modality models (all P ≤ .01). Conclusion The multimodality model effectively predicted BCR following RP in patients with PCa and may therefore provide an emerging and accurate tool to assist postoperative individualized treatment. Keywords: MR Imaging, Urinary, Pelvis, Comparative Studies Supplemental material is available for this article. © RSNA, 2024.

Whole-orbit-based multiparametric assessment of disease activity of thyroid eye disease on Dixon MRI.

PURPOSE: This study aimed to explore the diagnostic value of whole-orbit-based multiparametric assessment on Dixon MRI for the evaluation of the thyroid eye disease (TED) activity. METHODS: The retrospective study enrolled patients diagnosed as TED and obtained their axial and coronal Dixon MRI scans. Multiparameters were assessed, including water fraction (WF), fat fraction (FF) of extraocular muscles (EOMs), orbital fat (OF), and lacrimal gland (LG). The thickness of OF and herniation of LG were also measured. Univariable and multivariable logistic regression was applied to construct prediction models based on single or multiple structures. Receiver operating characteristic (ROC) curve analysis was also implemented. RESULTS: Univariable logistic analysis revealed significant differences in water fraction (WF) of the superior rectus (P = 0.018), fat fraction (FF) of the medial rectus (P = 0.029), WF of OF (P = 0.004), and herniation of LG (P = 0.012) between the active and inactive TED phases. Multivariable logistic analysis and corresponding receiver operating characteristic curve (ROC) analysis of each structure attained the area under the curve (AUC) values of 0.774, 0.771, and 0.729 for EOMs, OF, and LG, respectively, while the combination of the four imaging parameters generated a final AUC of 0.909. CONCLUSIONS: Dixon MRI may be used for fine multiparametric assessment of multiple orbital structures. The whole-orbit-based model improves the diagnostic performance of TED activity evaluation.

Early therapy evaluation of intra-arterial trastuzumab injection in a human breast cancer xenograft model using multiparametric MR imaging.

PURPOSE: To investigate the treatment efficacy of intra-arterial (IA) trastuzumab treatment using multiparametric magnetic resonance imaging (MRI) in a human breast cancer xenograft model. MATERIALS AND METHODS: Human breast cancer cells (BT474) were stereotaxically injected into the brains of nude mice to obtain a xenograft model. The mice were divided into four groups and subjected to different treatments (IA treatment [IA-T], intravenous treatment [IV-T], IA saline injection [IA-S], and the sham control group). MRI was performed before and at 7 and 14 d after treatment to assess the efficacy of the treatment. The tumor volume, apparent diffusion coefficient (ADC), and dynamic contrast-enhanced (DCE) MRI parameters (Ktrans, Kep, Ve, and Vp) were measured. RESULTS: Tumor volumes in the IA-T group at 14 d after treatment were significantly lower than those in the IV-T group (13.1 mm3 [interquartile range 8.48-16.05] vs. 25.69 mm3 [IQR 20.39-30.29], p = 0.005), control group (IA-S, 33.83 mm3 [IQR 32.00-36.30], p<0.01), and sham control (39.71 mm3 [IQR 26.60-48.26], p <0.001). The ADC value in the IA-T group was higher than that in the control groups (IA-T, 7.62 [IQR 7.23-8.20] vs. IA-S, 6.77 [IQR 6.48-6.87], p = 0.044 and vs. sham control, 6.89 [IQR 4.93-7.48], p = 0.004). Ktrans was significantly decreased following the treatment compared to that in the control groups (p = 0.002 and p<0.001 for vs. IA-S and sham control, respectively). Tumor growth was decreased in the IV-T group compared to that in the sham control group (25.69 mm3 [IQR 20.39-30.29] vs. 39.71 mm3 [IQR 26.60-48.26], p = 0.27); there was no significant change in the MRI parameters. CONCLUSION: IA treatment with trastuzumab potentially affects the early response to treatment, including decreased tumor growth and decrease of Ktrans, in a preclinical brain tumor model.

Nomograms based on multiparametric MRI radiomics integrated with clinical-radiological features for predicting the response to induction chemotherapy in nasopharyngeal carcinoma.

OBJECTIVE: To establish nomograms integrating multiparametric MRI radiomics with clinical-radiological features to identify the responders and non-responders to induction chemotherapy (ICT) in nasopharyngeal carcinoma (NPC). METHODS: We retrospectively analyzed the clinical and MRI data of 168 NPC patients between December 2015 and April 2022. We used 3D-Slicer to segment the regions of interest (ROIs) and the "Pyradiomic" package to extract radiomics features. We applied the least absolute shrinkage and selection operator regression to select radiomics features. We developed clinical-only, radiomics-only, and the combined clinical-radiomics nomograms using logistic regression analysis. The receiver operating characteristic curves, DeLong test, calibration, and decision curves were used to assess the discriminative performance of the models. The model was internally validated using 10-fold cross-validation. RESULTS: A total of 14 optimal features were finally selected to develop a radiomic signature, with an AUC of 0.891 (95 % CI, 0.825-0.946) in the training cohort and 0.837 (95 % CI, 0.723-0.932) in the testing cohort. The nomogram based on the Rad-Score and clinical-radiological factors for evaluating tumor response to ICT yielded an AUC of 0.926 (95 % CI, 0.875-0.965) and 0.901 (95 % CI, 0.815-0.979) in the two cohorts, respectively. Decision curves demonstrated that the combined clinical-radiomics nomograms were clinically useful. CONCLUSION: Nomograms integrating multiparametric MRI-based radiomics and clinical-radiological features could non-invasively discriminate ICT responders from non-responders in NPC patients.

MRI-based monitoring of prostate cancer after HIFU: Inter-reader agreement and diagnostic performance of the PI-FAB score.

PURPOSE: To investigate inter-reader agreement, and diagnostic performance of the Prostate Imaging after Focal Ablation (PI-FAB) score applied to multiparametric MRI (mpMRI) in patients who underwent focal high-intensity focused ultrasound (HIFU) therapy for localized prostate cancer. METHODS: In this retrospective, IRB-approved, single-center study, 73 men, who underwent focal HIFU treatment and received follow-up mpMRIs with subsequent prostate biopsies, were included. The PI-FAB score was applied to follow-up MRIs at 6, 12, and 36 months post-HIFU by two radiologists with different experience levels. Inter-reader agreement was assessed using Gwet's AC1, and the diagnostic performance of the PI-FAB score was assessed in relation to histopathologic results of subsequent prostate biopsies for each reader. RESULTS: PI-FAB scores showed substantial to almost perfect inter-reader agreement (AC1: 0.80-0.95) and demonstrated high specificity (Reader 1: 90-98 %, Reader 2: 87-98 %) and NPVs (Reader 1: 91-100 %, Reader 2: 88-97 %) in ruling out residual or recurrent in-field prostate cancer post-HIFU. Sensitivity (Reader 1: ≥43 %, Reader 2: ≥14 %) and PPVs (Reader 1: ≥33 %, Reader 2: ≥14 %) were mostly relatively lower, with notable disparities between the two readers, indicating the potential influence of radiologist experience. CONCLUSIONS: The PI-FAB score provides a consistent and reliable tool for post-HIFU monitoring of prostate cancer using mpMRI. It demonstrates substantial to almost perfect inter-reader agreement and is particularly effective in excluding in-field residual or recurrent prostate cancer post-HIFU treatment. Its application can potentially enhance post-treatment patient care, emphasizing its value as a non-invasive MRI-based monitoring approach after focal ablative therapy of the prostate.

Prostate Health Index (PHI) as a triage tool for reducing unnecessary magnetic resonance imaging (MRI) in patients at risk of prostate cancer.

INTRODUCTION: The aim of this study is to assess the usefulness of the Prostate Health Index (PHI) as a triage tool for selecting patients at risk of prostate cancer (PCa) who should undergo multiparametric Magnetic Resonance Imaging (mpMRI). MATERIAL AND METHODS: We enrolled 204 patients with suspected PCa. For each patient, a blood sample was collected before mpMRI to measure PHI. Findings on mpMRI were assessed according to the Prostate Imaging Reporting & Data System version 2.0 (PI-RADSv2) category scale. RESULTS: According to PI-RADSv2, patients were classified into two groups: PI-RADS < 3 (48 %) and ≥ 3 (52 %). PHI showed the best performance for predicting PI-RADS ≥ 3 [AUC: 0,747 (0,679-0,815), 0,680(0,607-0,754), and 0,613 (0,535-0,690) for PHI, PSA ratio, and total PSA, respectively]. The best PHI cut-off was 30, with a sensitivity of 90%. At the univariate logistic regression, total PSA (p = 0.007), PSA ratio (p = 0.001), [-2]proPSA (p = 0.019) and PHI (p < 0.001) were associated with PI-RADS ≥ 3; however, at the multivariate analysis, only PHI (p < 0.001) was found to be an independent predictor of PI-RADS ≥ 3. CONCLUSION: PHI could represent a reliable noninvasive tool for selecting patients to undergo mpMRI.

Oncological Outcomes in Men With Favorable Intermediate Risk Prostate Cancer Enrolled in Active Surveillance.

BACKGROUND/AIM: To evaluate the long-term oncological outcomes in men with intermediate risk prostate cancer (PCa) enrolled in active surveillance (AS). PATIENTS AND METHODS: From April 2015 to December 2022, 30 men with Gleason score 3+4/ISUP Grade Group2 (GG2), greatest percentage of cancer (GPC) ≤50%, Gleason pattern 4 ≤10%, ≤3 positive biopsy cores were enrolled in AS. All patients underwent confirmatory transperineal saturation biopsy (SPBx: 20 cores) 12 months from diagnosis plus multiparametric magnetic resonance (mpMRI) evaluation. At the last follow-up, 68Ga prostate-specific membrane antigen (PSMA) positron-emission tomography (PET)/computed tomography (CT) was added: lesions with PIRADS score ≥3 and/or standardized uptake value (SUVmax) >5 were submitted to four targeted cores. RESULTS: Three out of 30 (10%) men with GG2 PCa were reclassified at confirmatory biopsy. At the last follow-up (median 5.2 years), only 2 of 27 (7.4%) men were reclassified and 23/30 (76.6%) continued AS. CONCLUSION: Men with favorable GG2 PCa enrolled in AS have good long-term oncological results. The use of selective criteria (i.e., SPBx, mpMRI, PSMA PET/CT) reduces the risk of reclassification.

Metabolic habitat imaging with hemodynamic heterogeneity predicts individual progression-free survival in high-grade glioma.

AIM: We design a feasibility study to obtain a set of metabolic-hemodynamic habitats for tackling tumor spatial metabolic patterns with hemodynamic information. MATERIALS AND METHODS: Preoperative data from 69 high-grade gliomas (HGG) patients with subsequent histologic confirmation of HGG were prospectively collected (January 2016 to March 2020) after concurrent chemoradiotherapy (CCRT). Four vascular habitats were automatically segmented by multiparametric magnetic resonance imaging (MRI). The metabolic information, either at enhancing or edema tumor regions, was obtained by two neuroradiologists. The relative habitat volumes were used for weight estimation procedures for computing the coefficients of a linear regression model using weighted least squares (WLS) for metabolite semiquantifications (i.e. the Cho/NAA ratio and the Cho/Cr ratio) at vascular habitats. Multivariate Cox proportional hazard regression analyses are used to obtain the odds ratio (OR) and develop a nomogram using weighted estimators corresponding to each covariate derived from Cox regression coefficients. RESULTS: There was a strongly correlation between perfusion indexes and the Cho/Cr ratio (rCBV, r=0.71) or Cho/NAA ratio (rCBV, r=0.66) at high-angiogenic enhancing tumor habitats (HAT) habitat. Compared isocitrate dehydrogenase (IDH) mutation to their wild type, the IDH wild type had significantly decreased Cho/Cr ratio (IDH mutation: Cho/Cr ratio = 2.44 ± 0.33, IDH wildtype: Cho/Cr ratio = 2.66 ± 0.36, p=0.02) and Cho/NAA ratio (IDH mutation: Cho/Cr ratio = 4.59 ± 0.61, IDH wildtype: Cho/Cr ratio = 4.99 ± 0.66, p=0.022) at the HAT. The C-index for the median progression-free survival (PFS) prediction was 0.769 for the Cho/NAA nomogram and 0.747 for the Cho/Cr nomogram through 1000 bootstrapping validation. CONCLUSIONS: Our findings suggest that spatial metabolism combined with hemodynamic heterogeneity is associated with individual PFS to HGG patients post-CCRT.

Head-to-head comparison of prostate-specific membrane antigen positron emission tomography/computed tomography and multiparametric magnetic resonance imaging in the detection of biochemical recurrence of prostate cancer: a systematic review and meta-analysis.

AIM: Our main goal of this meta-analytical analysis was to evaluate the diagnostic effectiveness of prostate-specific membrane antigen (PSMA) positron emission tomography (PET)/computed tomography (CT) against multiparametric magnetic resonance imaging (mpMRI) in the context of identifying biochemical recurrence in patients with prostate cancer (PCa). MATERIALS AND METHODS: A thorough search covering articles published until March 2023 was carried out across major databases such as PubMed, Embase, and Web of Science. Studies examining the direct comparison of PSMA PET/CT and mpMRI in patients with PCa suffering biochemical recurrence were included in the inclusion criteria. Using the renowned Quality Assessment of Diagnostic Performance Studies-2 technique, each study's methodological rigor was assessed. RESULTS: We analyzed data from six eligible studies involving 290 patients in total. The combined data showed that for PSMA PET/CT and mpMRI, respectively, the pooled overall detection rates for recurrent PCa after definitive treatment were 0.69 (95% confidence interval [CI]: 0.45-0.89) and 0.70 (95% CI: 0.44-0.91). The detection rates for local recurrence were specifically 0.52 (95% CI: 0.39-0.65) and 0.62 (95% CI: 0.31-0.89), while they were 0.50 (95% CI: 0.26-0.74) and 0.32 (95% CI: 0.18-0.48) for lymph node metastasis. Notably, there was no discernible difference between the two imaging modalities in terms of the overall detection rate (P = 0.95). The detection rates for local recurrence and lymph node metastasis did not differ statistically significantly (P = 0.55, 0.23). CONCLUSION: The performance of PSMA PET/CT and mpMRI in identifying biochemical recurrence in PCa appears to be comparable. However, the meta-analysis' findings came from research with modest sample sizes. In this context, more extensive research should be conducted in the future.

MRI in-bore biopsy following MRI/US fusion-guided biopsy in patients with persistent suspicion of clinically significant prostate cancer.

PURPOSE: Patients with suspicion of clinically significant prostate cancer (csPC) on multiparametric prostate MRI (mpMRI) but negative or inconclusive MRI/US fusion-guided biopsy (FB) can be challenging in clinical practice. To assess the utility of MRI in-bore biopsy (IB) in patients with discordant imaging and histopathological findings after FB. METHODS: Consecutive patients with Prostate Imaging Reporting and Data System (PI-RADS) category 4 or 5 on mpMRI at 3T after FB without histologically confirmed csPC who underwent IB between 01/2014 and 05/2022, were retrospectively included. The primary objective was to assess the detection rate of csPC. Secondary objectives were to analyze clinical parameters, MRI parameters, and lesion localization. RESULTS: In the final cohort of 51 patients, the IB resulted in an overall detection rate of 71% for PC and 47% for csPC. Furthermore, in 55% of cases with initial low-grade PC, the Gleason score was upgraded after IB. CsPC was often detected apical and/or anterior. The detection rate for PC was 58% in PI-RADS category 4 and 94% in PI-RADS category 5 (csPC 39% and 61%, respectively). Patients with csPC had statistically significant smaller prostate volumes, a higher PI-RADS category, a higher prostate-specific antigen density (PSAD), and were older. CONCLUSIONS: For a relevant proportion of patients with PI-RADS category 4 or 5 and negative or inconclusive findings on previous FB, but with persistent suspicion of csPC, a subsequent IB verified the presence of csPC. Therefore, IB can be a backup in cases of uncertainty.

Multiparametric MRI-based radiomics nomogram for identifying cervix-corpus junction cervical adenocarcinoma from endometrioid adenocarcinoma.

OBJECTIVE: To developed a magnetic resonance imaging (MRI) radiomics nomogram to identify adenocarcinoma at the cervix-corpus junction originating from the endometrium or cervix in order to better guide clinical treatment. METHODS: Between February 2011 and September 2021, the clinicopathological data and MRI in 143 patients with histopathologically confirmed cervical adenocarcinoma (CAC, n = 86) and endometrioid adenocarcinoma (EAC, n = 57) were retrospectively analyzed at the cervix-corpus junction. Radiomics features were extracted from fat-suppressed T2-weighted imaging (FS-T2WI), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) maps, and delayed phase contrast-enhanced T1-weighted imaging (CE-T1WI) sequences. A radiomics nomogram was developed integrating radscore with independent clinical risk factors. The area under the curve (AUC) was used to evaluate the diagnostic efficacy of the radscore, nomogram and two different experienced radiologists in differentiating CAC from EAC at the cervix-corpus junction, and Delong test was applied to compare the differences of their diagnostic performance. RESULTS: In the training cohort, the AUC was 0.93 for radscore; 0.97 for radiomics nomograms; 0.85 and 0.86 for radiologists 1 and 2, respectively. Delong test showed that the differential efficacy of nomogram was significant better than those of radiologists in the training cohort (both P < 0.05). CONCLUSIONS: The nomogram based on radscore and clinical risk factors could better differentiate CAC from EAC at the cervix-corpus junction than radiologists, and preoperatively and non-invasively identify the origin of adenocarcinoma at the cervix-corpus junction, which facilitates clinicians to make individualized treatment decision.

Histogram analysis of MR quantitative parameters: are they correlated with prognostic factors in prostate cancer?

PURPOSE: To investigate the correlation between quantitative MR parameters and prognostic factors in prostate cancer (PCa). METHOD: A total of 186 patients with pathologically confirmed PCa who underwent preoperative multiparametric MRI (mpMRI), including synthetic MRI (SyMRI), were enrolled from two medical centers. The histogram metrics of SyMRI [T1, T2, proton density (PD)] and apparent diffusion coefficient (ADC) values were extracted. The Mann‒Whitney U test or Student's t test was employed to determine the association between these histogram features and the prognostically relevant factors. Receiver operating characteristic (ROC) curve analysis was conducted to evaluate the differentiation performance. Spearman's rank correlation coefficients were calculated to determine the correlations between histogram parameters and the International Society of Urological Pathology (ISUP) grade group as well as pathological T stage. RESULTS: Significant correlations were found between the histogram parameters and the ISUP grade as well as pathological T stage of PCa. Among these histogram parameters, ADC_minimum had the strongest correlation with the ISUP grade (r = - 0.481, p < 0.001), and ADC_Median showed the strongest association with pathological T stage (r = - 0.285, p = 0.008). The ADC_10th percentile exhibited the highest performance in identifying clinically significant prostate cancer (csPCa) (AUC 0.833; 95% CI 0.771-0.883). When discriminating between the status of different prognostically relevant factors, a significant difference was observed between extraprostatic extension-positive and -negative cancers with regard to histogram parameters of the ADC map (10th percentile, 90th percentile, mean, median, minimum) and T1 map (minimum) (p = 0.002-0.032). Moreover, histogram parameters of the ADC map (90th percentile, maximum, mean, median), T2 map (10th percentile, median), and PD map (10th percentile, median) were significantly lower in PCa with perineural invasion (p = 0.009-0.049). The T2 values were significantly lower in patients with seminal vesicle invasion (minimum, p = 0.036) and positive surgical margin (10th percentile, 90th percentile, mean, median, and minimum, p = 0.015-0.025). CONCLUSION: Quantitative histogram parameters derived from synthetic MRI and ADC maps may have great potential for predicting the prognostic features of PCa.

Multiparametric MRI-based whole-liver radiomics for predicting early-stage liver fibrosis in rabbits.

OBJECTIVES: To develop and validate a whole-liver radiomic model using multiparametric MRI for predicting early-stage liver fibrosis (LF) in rabbits. METHODS: A total of 134 rabbits (early-stage LF, n = 91; advanced-stage LF, n = 43) who underwent liver magnetic resonance elastography (MRE), hepatobiliary phase, dynamic contrast enhanced (DCE), intravoxel incoherent motion (IVIM), diffusion kurtosis imaging, and T2* scanning were enrolled and randomly allocated to either the training or validation cohort. Whole-liver radiomic features were extracted and selected to develop a radiomic model and generate quantitative Rad-scores. Then, multivariable logistic regression was utilized to determine the Rad-scores associated with early-stage LF, and effective features were integrated to establish a combined model. The predictive performance was assessed by the area under the curve (AUC). RESULTS: The MRE model achieved superior AUCs of 0.95 in the training cohort and 0.86 in the validation cohort, followed by the DCE-MRI model (0.93 and 0.82), while the IVIM model had lower AUC values of 0.91 and 0.82, respectively. The Rad-scores of MRE, DCE-MRI and IVIM were identified as independent predictors associated with early-stage LF. The combined model demonstrated AUC values of 0.96 and 0.88 for predicting early-stage LF in the training and validation cohorts, respectively. CONCLUSIONS: Our study highlights the remarkable performance of a multiparametric MRI-based radiomic model for the individualized diagnosis of early-stage LF. ADVANCES IN KNOWLEDGE: This is the first study to develop a combined model by integrating multiparametric radiomic features to improve the accuracy of LF staging.

Comparison of Ga-PSMA PET MRI with mpMRI in localization and regional staging of prostate cancer.

PURPOSE: To compare diagnostic accuracy in localization and detection of extraprostatic extension (EPE), seminal vesicle invasion (SVI), lymph node involvement (LNI) between PSMA PET MRI and multiparametric MRI (mpMRI) in carcinoma prostate. METHODS: We did a prospective study of consecutive men with biopsy-proven prostate cancer who underwent radical prostatectomy between July'2020 and Dec'2021 at our institution. Patients underwent PSMA PET MRI imaging. MpMRI findings were inferred separately by another radiologist who was blinded to the PSMA PET findings. PIRADS > 2 and any standardized uptake value (SUV) were considered positive. Findings were mapped to a 30-region anatomical grid and compared with pathology. The uro-pathologist also marked the presence of the tumor onto the same anatomical grid. The presence of EPE, SVI, and LVI was noted. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The significance in difference: McNemar test. SUVmax and Gleason score: Kruskal-Wallis test. RESULTS: Seventy-five men (mean age 65) with an average PSA of 21.5 ng/ml were included. The sensitivity of PSMA PET MRI for localization was higher [63.6 vs 41.9] (p < 0.001) while specificity was similar [81.5 vs 83.2] (p 0.103). The former had a higher sensitivity to detect SVI [85.7 vs 57.10] (p = 0.03). No difference in the detection of EPE or LNI was noted. SUVmax > 7 was associated with high-risk disease (Gleason score >/= 7). LIMITATIONS: non-randomized nature, higher risk population. CONCLUSION: Ga-PSMA PET MRI improved the localization of prostate cancer and better detection of SVI. Further studies are required. It can act as a single-stop investigation for the primary staging of prostate cancer.

Values of multiparametric and biparametric MRI in diagnosing clinically significant prostate cancer: a multivariate analysis.

BACKGROUND: To investigate the value of semi-quantitative and quantitative parameters (PI-RADS score, T2WI score, ADC, Ktrans, and Kep) based on multiparametric MRI (mpMRI) or biparametric MRI (bpMRI) combined with prostate specific antigen density (PSAD) in detecting clinically significant prostate cancer (csPCa). METHODS: A total of 561 patients (276 with csPCa; 285 with non-csPCa) with biopsy-confirmed prostate diseases who underwent preoperative mpMRI were included. Prostate volume was measured for calculation of PSAD. Prostate index lesions were scored on a five-point scale on T2WI images (T2WI score) and mpMRI images (PI-RADS score) according to the PI-RADS v2.1 scoring standard. DWI and DCE-MRI images were processed to measure the quantitative parameters of the index lesion, including ADC, Kep, and Ktrans values. The predictors of csPCa were screened by logistics regression analysis. Predictive models of bpMRI and mpMRI were established. ROC curves were used to evaluate the efficacy of parameters and the model in diagnosing csPCa. RESULTS: The independent diagnostic accuracy of PSA density, PI-RADS score, T2WI score, ADCrec, Ktrans, and Kep for csPCa were 80.2%, 89.5%, 88.3%, 84.6%, 58.5% and 61.6%, respectively. The diagnostic accuracy of bpMRI T2WI score and ADC value combined with PSAD was higher than that of PI-RADS score. The combination of mpMRI PI­RADS score, ADC value with PSAD had the highest diagnostic accuracy. CONCLUSIONS: PI-RADS score according to the PI-RADS v2.1 scoring standard was the most accurate independent diagnostic index. The predictive value of bpMRI model for csPCa was slightly lower than that of mpMRI model, but higher than that of PI-RADS score.

Biparametric MRI for Local Staging of Prostate Cancer: Current Status and Future Applications.

BACKGROUND/AIM: Multiparametric magnetic resonance imaging (mpMRI) is the recommended modality for local staging of prostate cancer (PCa). The use of dynamic contrast-enhanced (DCE) imaging alone significantly improves staging performance. However, several studies have revealed that DCE imaging adds no extra benefit for PCa detection. Many authors observed benefits of performing prostate MRI without DCE, so called biparametric MRI (bpMRI), such as the elimination of the toxicity of gadolinium administration, reduction of examination time, costs and better accessibility. This narrative review describes the variety of imaging modalities in Local staging of PCa with bpMRI utilization and its comparison to mpMRI. MATERIALS AND METHODS: A search of medical databases was performed to find eligible articles using the following key words: "prostate cancer", "MRI", "multiparametric", "biparametric" and "staging". MEDLINE, Web of Science, PubMed and Google Scholar were used to search for eligible articles published in the past 5 years and compared the diagnostic accuracy of mpMRI and bp MRI in local staging of PCa. RESULTS: A total of 48 articles were evaluated. Multiple systematic reviews used pooled data to compare the accuracy of biparametric and multiparametric examinations. However, all these studies advise caution on using pooled data for clinical practice, pointing to multiple sources of heterogeneity among the studies evaluated. CONCLUSION: Given the absence of prospective data comparing bpMRI and mpMRI, randomized, prospective, multicenter studies are encouraged. However, mpMRI is the recommended modality for local staging of PCa. It has superior performance compared to traditional staging based on clinical nomograms, and provides additional information on the site and extent of disease.

Multiparametric MRI radiomics improves preoperative diagnostic performance for local staging in patients with endometrial cancer.

PURPOSE: To determine whether multiparametric magnetic resonance imaging (MRI) radiomics-based machine learning methods can improve preoperative local staging in patients with endometrial cancer (EC). METHODS: Data of patients with histologically confirmed EC who underwent preoperative MRI were retrospectively analyzed and divided into a training or test set. Radiomic features extracted from multiparametric MR images were used to train and test the prediction of deep myometrial invasion (DMI) and cervical stromal invasion (CSI). Two radiologists assessed the presence of DMI and CSI on conventional MR images. A combined model incorporating a radiomic signature and conventional MR images was constructed and presented as a nomogram. Performance of the predictive models was assessed using the area under curve (AUC) in the receiver operating curve analysis and pairwise comparison using DeLong's test with Bonferroni correction. RESULTS: This study included 198 women (training set = 138, test set = 60). Conventional MRI achieved AUCs of 0.837 and 0.799 for detecting DMI and 0.825 and 0.858 for detecting CSI in the training and test sets, respectively. The nomogram achieved AUCs of 0.928 and 0.869 for detecting DMI and 0.913 and 0.937 for detecting CSI in the training and test sets, respectively. The ability of the nomogram to detect DMI and CSI in the two sets was superior to that of conventional MRI (adjusted p < 0.05), except for the ability to detect CSI in the test set (adjusted p > 0.05). CONCLUSION: A nomogram incorporating radiomics signature into conventional MRI improved the efficacy of preoperative local staging of EC.