Improved prostate cancer diagnosis using a modified ResNet50-based deep learning architecture.

Prostate cancer, the most common cancer in men, is influenced by age, family history, genetics, and lifestyle factors. Early detection of prostate cancer using screening methods improves outcomes, but the balance between overdiagnosis and early detection remains debated. Using Deep Learning (DL) algorithms for prostate cancer detection offers a promising solution for accurate and efficient diagnosis, particularly in cases where prostate imaging is challenging. In this paper, we propose a Prostate Cancer Detection Model (PCDM) model for the automatic diagnosis of prostate cancer. It proves its clinical applicability to aid in the early detection and management of prostate cancer in real-world healthcare environments. The PCDM model is a modified ResNet50-based architecture that integrates faster R-CNN and dual optimizers to improve the performance of the detection process. The model is trained on a large dataset of annotated medical images, and the experimental results show that the proposed model outperforms both ResNet50 and VGG19 architectures. Specifically, the proposed model achieves high sensitivity, specificity, precision, and accuracy rates of 97.40%, 97.09%, 97.56%, and 95.24%, respectively.

Value of T2 Mapping MRI for Prostate Cancer Detection and Classification.

BACKGROUND: Currently, multi-parametric prostate MRI (mpMRI) consists of a qualitative T2 , diffusion weighted, and dynamic contrast enhanced imaging. Quantification of T2 imaging might further standardize PCa detection and support artificial intelligence solutions. PURPOSE: To evaluate the value of T2 mapping to detect prostate cancer (PCa) and to differentiate PCa aggressiveness. STUDY TYPE: Retrospective single center cohort study. POPULATION: Forty-four consecutive patients (mean age 67 years; median PSA 7.9 ng/mL) with mpMRI and verified PCa by subsequent targeted plus systematic MR/ultrasound (US)-fusion biopsy from February 2019 to December 2019. FIELD STRENGTH/SEQUENCE: Standardized mpMRI at 3 T with an additionally acquired T2 mapping sequence. ASSESSMENT: Primary endpoint was the analysis of quantitative T2 values and contrast differences/ratios (CD/CR) between PCa and benign tissue. Secondary objectives were the correlation between T2 values, ISUP grade, apparent diffusion coefficient (ADC) value, and PI-RADS, and the evaluation of thresholds for differentiating PCa and clinically significant PCa (csPCa). STATISTICAL TESTS: Mann-Whitney test, Spearman's rank (rs ) correlation, receiver operating curves, Youden's index (J), and AUC were performed. Statistical significance was defined as P < 0.05. RESULTS: Median quantitative T2 values were significantly lower for PCa in PZ (85 msec) and PCa in TZ (75 msec) compared to benign PZ (141 msec) or TZ (97 msec) (P < 0.001). CD/CR between PCa and benign PZ (51.2/1.77), respectively TZ (19.8/1.29), differed significantly (P < 0.001). The best T2 -mapping threshold for PCa/csPCa detection was for TZ 81/86 msec (J = 0.929/1.0), and for PZ 110 msec (J = 0.834/0.905). Quantitative T2 values of PCa did not correlate significantly with the ISUP grade (rs  = 0.186; P = 0.226), ADC value (rs  = 0.138; P = 0.372), or PI-RADS (rs  = 0.132; P = 0.392). DATA CONCLUSION: Quantitative T2 values could differentiate PCa in TZ and PZ and might support standardization of mpMRI of the prostate. Different thresholds seem to apply for PZ and TZ lesions. However, in the present study quantitative T2 values were not able to indicate PCa aggressiveness. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.

Voxel-level Classification of Prostate Cancer on Magnetic Resonance Imaging: Improving Accuracy Using Four-Compartment Restriction Spectrum Imaging.

BACKGROUND: Diffusion magnetic resonance imaging (MRI) is integral to detection of prostate cancer (PCa), but conventional apparent diffusion coefficient (ADC) cannot capture the complexity of prostate tissues and tends to yield noisy images that do not distinctly highlight cancer. A four-compartment restriction spectrum imaging (RSI4 ) model was recently found to optimally characterize pelvic diffusion signals, and the model coefficient for the slowest diffusion compartment, RSI4 -C1 , yielded greatest tumor conspicuity. PURPOSE: To evaluate the slowest diffusion compartment of a four-compartment spectrum imaging model (RSI4 -C1 ) as a quantitative voxel-level classifier of PCa. STUDY TYPE: Retrospective. SUBJECTS: Forty-six men who underwent an extended MRI acquisition protocol for suspected PCa. Twenty-three men had benign prostates, and the other 23 men had PCa. FIELD STRENGTH/SEQUENCE: A 3 T, multishell diffusion-weighted and axial T2-weighted sequences. ASSESSMENT: High-confidence cancer voxels were delineated by expert consensus, using imaging data and biopsy results. The entire prostate was considered benign in patients with no detectable cancer. Diffusion images were used to calculate RSI4 -C1 and conventional ADC. Classifier images were also generated. STATISTICAL TESTS: Voxel-level discrimination of PCa from benign prostate tissue was assessed via receiver operating characteristic (ROC) curves generated by bootstrapping with patient-level case resampling. RSI4 -C1 was compared to conventional ADC for two metrics: area under the ROC curve (AUC) and false-positive rate for a sensitivity of 90% (FPR90 ). Statistical significance was assessed using bootstrap difference with two-sided α = 0.05. RESULTS: RSI4 -C1 outperformed conventional ADC, with greater AUC (mean 0.977 [95% CI: 0.951-0.991] vs. 0.922 [0.878-0.948]) and lower FPR90 (0.032 [0.009-0.082] vs. 0.201 [0.132-0.290]). These improvements were statistically significant (P < 0.05). DATA CONCLUSION: RSI4 -C1 yielded a quantitative, voxel-level classifier of PCa that was superior to conventional ADC. RSI classifier images with a low false-positive rate might improve PCa detection and facilitate clinical applications like targeted biopsy and treatment planning. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Transperineal freehand multiparametric MRI fusion targeted biopsies under local anaesthesia for prostate cancer diagnosis: a multicentre prospective study of 1014 cases.

OBJECTIVE: To assess the outcomes of multiparametric magnetic resonance imaging (mpMRI) transperineal targeted fusion biopsy (TPFBx) under local anaesthesia. PATIENTS AND METHODS: We prospectively screened 1327 patients with a positive mpMRI undergoing TPFBx (targeted cores and systematic cores) under local anaesthesia, at two tertiary referral institutions, between September 2016 and May 2019, for inclusion in the present study. Primary outcomes were detection of clinically significant prostate cancer (csPCa) defined as (1) International Society of Urological Pathologists (ISUP) grade >1 or ISUP grade 1 with >50% involvement of prostate cancer (PCa) in a single core or in >2 cores (D1) and (2) ISUP grade >1 PCa (D2). Secondary outcomes were: assessment of peri-procedural pain (numerical rating scale [NRS]) and procedure timings; erectile (International Index of Erectile Function) and urinary (International Prostate Symptom Score) function changes; and complications. We also investigated the value of systematic sampling and concordance with radical prostatectomy (RP). RESULTS: A total of 1014 patients were included, of whom csPCa was diagnosed in 39.4% (n = 400). The procedure was tolerable (NRS pain score 3.1 ± 2.3), with no impact on erectile (P = 0.45) or urinary (P = 0.58) function, and a low rate of complications (Clavien-Dindo grades 1 or 2, n = 8; grade >2, n = 0). No post-biopsy sepsis was recorded. Twenty-two men (95% confidence interval [CI] 17-29) needed to undergo additional systematic biopsy to diagnose one csPCa missed by targeted biopsies (D1). ISUP grade concordance of biopsies with RP was as follows: k = 0.40 (95% CI 0.31-0.49) for targeted cores alone and k = 0.65 (95% CI 0.57-0.72; P < 0.05) overall. CONCLUSIONS: The use of TPFBx under local anaesthesia yielded good csPCa detection and was feasible, quick, well tolerated and safe. Infectious risk was negligible. Addition of systematic to targeted cores may not be needed in all men, although it improves csPCa detection and concordance with RP.

Prostate-specific antigen density during dutasteride treatment for 1 year predicts the presence of prostate cancer in benign prostatic hyperplasia after the first negative biopsy (PREDICT study).

OBJECTIVES: To prospectively evaluate the detection rate of prostate cancer, and to identify the risk factors of prostate cancer detection after a 1-year administration of dutasteride and first negative prostate biopsy. METHODS: Patients with benign prostatic hyperplasia who presented high prostate-specific antigen levels after the first negative prostate biopsy were administered 0.5 mg dutasteride daily for 1 year. They underwent a repeat prostate biopsy after 1 year. The primary end-point was the detection rate of prostate cancer. The secondary end-point was the ability of prostate-specific antigen kinetics to predict prostate cancer detection. Prostate-specific antigen was measured before the initial prostate biopsy and at 6, 9 and 12 months after starting dutasteride. Patients were classified into a prostate cancer and a non-prostate cancer group. RESULTS: Prostate cancer was detected in 15 of 149 participants (10.1%). The total prostate-specific antigen change between the prostate cancer and non-prostate cancer group at 1 year was significantly different (P = 0.002). Although prostate-specific antigen levels at baseline did not significantly differ between study groups (P = 0.102), prostate-specific antigen levels at 6, 9 and 12 months were significantly different (P = 0.002, P = 0.001 and P < 0.001, respectively). The mean reduction rate of prostate-specific antigen density between the prostate cancer and non-prostate cancer group at 1 year was significantly different (-4.25 ± 76.5% vs -38.0 ± 28.7%, P = 0.001). Using a multivariate analysis, a >10% increase of prostate-specific antigen density at 1 year post-dutasteride treatment was the only predictive risk factor for prostate cancer after the first negative prostate biopsy (odds ratio 11.238, 95% confidence interval 3.112-40.577, P < 0.001). CONCLUSION: In the present study cohort, >10% increase in prostate-specific antigen density represented the only significant predictive risk factor for prostate cancer diagnosis in patients with elevated prostate-specific antigen after the first negative prostate biopsy.

A Comprehensive Study of Data Augmentation Strategies for Prostate Cancer Detection in Diffusion-Weighted MRI Using Convolutional Neural Networks.

Data augmentation refers to a group of techniques whose goal is to battle limited amount of available data to improve model generalization and push sample distribution toward the true distribution. While different augmentation strategies and their combinations have been investigated for various computer vision tasks in the context of deep learning, a specific work in the domain of medical imaging is rare and to the best of our knowledge, there has been no dedicated work on exploring the effects of various augmentation methods on the performance of deep learning models in prostate cancer detection. In this work, we have statically applied five most frequently used augmentation techniques (random rotation, horizontal flip, vertical flip, random crop, and translation) to prostate diffusion-weighted magnetic resonance imaging training dataset of 217 patients separately and evaluated the effect of each method on the accuracy of prostate cancer detection. The augmentation algorithms were applied independently to each data channel and a shallow as well as a deep convolutional neural network (CNN) was trained on the five augmented sets separately. We used area under receiver operating characteristic (ROC) curve (AUC) to evaluate the performance of the trained CNNs on a separate test set of 95 patients, using a validation set of 102 patients for finetuning. The shallow network outperformed the deep network with the best 2D slice-based AUC of 0.85 obtained by the rotation method.

Combining prostate health index and multiparametric magnetic resonance imaging in the diagnosis of clinically significant prostate cancer in an Asian population.

OBJECTIVE: To evaluate the practicability of combining prostate health index (PHI) and multiparametric magnetic resonance imaging (mpMRI) for the detection of clinically significant prostate cancer (csPC) in an Asian population. PATIENTS AND METHODS: We prospectively enrolled patients who underwent prostate biopsy due to elevated serum prostate-specific antigen (PSA > 4 ng/mL) and/or abnormal digital rectal examination in a tertiary referral center. Before prostate biopsy, the serum samples were tested for PSA, free PSA, and p2PSA to calculate PHI. Besides, mpMRI was performed using a 3-T scanner and reported in the Prostate Imaging Reporting and Data System version 2 (PI-RADS v2). The diagnostic performance of PHI, mpMRI, and combination of both was assessed. RESULT: Among 102 subjects, 39 (38.2%) were diagnosed with PC, including 24 (23.5%) with csPC (Gleason ≥ 7). By the threshold of PI-RADS ≥ 3, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) to predict csPC were 100%, 44.9%, 35.8%, and 100%, respectively. By the threshold of PHI ≥ 30, the sensitivity, specificity, PPV, and NPV to predict csPC were 91.7%, 43.6%, 33.3%, and 94.4%, respectively. The area under the receiver operator characteristic curve of combining PHI and mpMRI was greater than that of PHI alone (0.873 vs. 0.735, p = 0.002) and mpMRI alone (0.873 vs. 0.830, p = 0.035). If biopsy was restricted to patients with PI-RADS 5 as well as PI-RADS 3 or 4 and PHI ≥ 30, 50% of biopsy could be avoided with one csPC patient being missed. CONCLUSION: The combination of PHI and mpMRI had higher accuracy for detection of csPC compared with PHI or mpMRI alone in an Asian population.

Evaluation of Artificial Tactile Sense in Mass Detection in Silicone Phantom for the Diagnosis of Prostate Tumor.

We analyzed the kinetic and kinematic variables of artificial tactile and artificial vibrotactile sensing test for mass detection in silicon phantom to determine tactile intensity and speed to obtain the best result in detecting the type and location of the mass. This study has utilized Artificial Tactile Sensing Instrument for Mass Detection (ATSIMD) in cylindrical silicone phantoms. The masses embedded in these samples were inserted in axial and environmental, deep and surface positions. The loading velocity, probe location, and the frequency of the applied force were considered as the independent variables in this study. It was found that for superficial mases the accuracy of detection at low speed 5 mm/sec, although dependent on the probe, but was 50% higher than under other conditions. For deep masses, with increasing mass depth, the accuracy of detection at medium speed of 8 mm/sec was 30% higher than at low speed. Mass detection by ATSIMD used in this study showed maximum efficiency at medium loading velocity. At low and high loading velocities, the dependence of mass detection on the probe location is related to the interaction of the testing method, tissue, and viscoelastic properties of the tissue.

Detection and Localization of Prostate Cancer at 3-T Multiparametric MRI Using PI-RADS Segmentation.

OBJECTIVE. The purpose of this study is to determine the overall and sector-based performance of 3-T multiparametric MRI for prostate cancer (PCa) detection and localization by using Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) scoring and segmentation compared with whole-mount histopathologic analysis. MATERIALS AND METHODS. Multiparametric 3-T MRI examinations of 415 consecutive men were compared with thin-section whole-mount histopathologic analysis. A genitourinary radiologist and pathologist collectively determined concordance. Two radiologists assigned PI-RADSv2 categories and sectoral location to all detected foci by consensus. Tumor detection rates were calculated for clinical and pathologic (Gleason score) variables. Both rigid and adjusted sector-matching models were used to account for fixation-related issues. RESULTS. The 415 patients had 863 PCa foci (52.7% had a Gleason score ≥ 7, 61.9% were ≥ 1 cm, and 90.4% (375/415) of index lesions were ≥ 1 cm) and 16,185 prostate sectors. Multiparametric MRI enabled greater detection of PCa lesions 1 cm or larger (all lesions vs index lesions, 61.6% vs 81.6%), lesions with Gleason score greater than or equal to 7 (all lesions vs index lesions, 71.4% vs 80.9%), and index lesions with both Gleason score greater than or equal to 7 and size 1 cm or larger (83.3%). Higher sensitivity was obtained for adjusted versus rigid tumor localization for all lesions (56.0% vs 28.5%), index lesions (55.4% vs 34.3%), lesions with Gleason score greater than or equal to 7 (55.7% vs 36.0%), and index lesions 1 cm or larger (56.1% vs 35.0%). Multiparametric 3-T MRI had similarly high specificity (96.0-97.9%) for overall and index tumor localization with adjusted and rigid sector-matching approaches. CONCLUSION. Using 3-T multiparametric MRI and PI-RADSv2, we achieved the highest sensitivity (83.3%) for the detection of lesions 1 cm or larger with Gleason score greater than or equal to 7. Sectoral localization of PCa within the prostate was moderate and was better with an adjusted model than with a rigid model.

Prediction Medicine: Biomarkers, Risk Calculators and Magnetic Resonance Imaging as Risk Stratification Tools in Prostate Cancer Diagnosis.

This review discusses the most recent evidence for currently available risk stratification tools in the detection of clinically significant prostate cancer (csPCa), and evaluates diagnostic strategies that combine these tools. Novel blood biomarkers, such as the Prostate Health Index (PHI) and 4Kscore, show similar ability to predict csPCa. Prostate cancer antigen 3 (PCA3) is a urinary biomarker that has inferior prediction of csPCa compared to PHI, but may be combined with other markers like TMPRSS2-ERG to improve its performance. Original risk calculators (RCs) have the advantage of incorporating easy to retrieve clinical variables and being freely accessible as a web tool/mobile application. RCs perform similarly well as most novel biomarkers. New promising risk models including novel (genetic) markers are the SelectMDx and Stockholm-3 model (S3M). Prostate magnetic resonance imaging (MRI) has evolved as an appealing tool in the diagnostic arsenal with even stratifying abilities, including in the initial biopsy setting. Merging biomarkers, RCs and MRI results in higher performances than their use as standalone tests. In the current era of prostate MRI, the way forward seems to be multivariable risk assessment based on blood and clinical parameters, potentially extended with information from urine samples, as a triaging test for the selection of candidates for MRI and biopsy.

Novel Biomarkers for Prostate Cancer Detection and Prognosis.

Prostate cancer (PCa) remains as one of the most controversial issues in health care because of the dilemmas related to screening using Prostate Specific Antigen (PSA). A high number of false positive biopsies and an elevated rate of overdiagnosis are the main problems associated with PSA. New PCa biomarkers have been recently proposed to increase the predictive value of PSA. The published results showed that PCA3 score, Prostate Health Index and 4Kscore can reduce the number of unnecessary biopsies, outperforming better than PSA and the percentage of free PSA. Furthermore, 4Kscore provides with high accuracy an individual risk for high-grade PCa. High values of PHI are also associated with tumor aggressiveness. In contrast, the relationship of PCA3 score with aggressiveness remains controversial, with studies showing opposite conclusions. Finally, the development of molecular biology has opened the study of genes, among them TMPRSS2:ERG fusion gene and miRNAs, in PCa detection and prognosis.

Lesion volume predicts prostate cancer risk and aggressiveness: validation of its value alone and matched with prostate imaging reporting and data system score.

OBJECTIVE: To demonstrate the association between magnetic resonance imaging (MRI) estimated lesion volume (LV), prostate cancer detection and tumour clinical significance, evaluating this variable alone and matched with Prostate Imaging Reporting and Data System version 2 (PI-RADS v2) score. PATIENTS AND METHODS: We retrospectively analysed 157 consecutive patients, with at least one prior negative systematic prostatic biopsy, who underwent transperineal prostate MRI/ultrasonography fusion-targeted biopsy between January 2014 and February 2016. Suspicious lesions were delineated using a 'region of interest' and the system calculated prostate volume and LV. Patients were divided in groups considering LV (≤0.5, 0.5-1, ≥1 mL) and PI-RADS score (1-5). We considered clinically significant prostate cancer as all cancers with a Gleason score of ≥3 + 4 as suggested by PI-RADS v2. A direct comparison between MRI estimated LV (MRI LV) and histological tumour volume (HTV) was done in 23 patients who underwent radical prostatectomy during the study period. Differences between MRI LV and HTV were assessed using the paired sample t-test. MRI LV and HTV concordance was verified using a Bland-Altman plot. The chi-squared test and logistic and ordinal regression models were used to evaluate difference in frequencies. RESULTS: The MRI LV and PI-RADS score were associated both with prostate cancer detection (both P < 0.001) and with significant prostate cancer detection (P < 0.001 and P = 0.008, respectively). When the two variables were matched, increasing LV increased the risk within each PI-RADS group. Prostate cancer detection was 1.4-times higher for LVs of 0.5-1 mL and 1.8-times higher for LVs of ≥1 mL; significant prostate cancer detection was 2.6-times for LVs of 0.5-1 mL and 4-times for LVs of ≥1 mL. There was a positive correlation between MRI LV and HTV (r = 0.9876, P < 0.001). Finally, Bland-Altman analysis showed that MRI LV was underestimated by 4.2% compared to HTV. Study limitations include its monocentric and retrospective design and the limited cohort. CONCLUSIONS: This study demonstrates that PI-RADS score and the MRI LV, independently and in combination, are associated with prostate cancer detection and with tumour clinical significance.

The addition of a sagittal image fusion improves the prostate cancer detection in a sensor-based MRI /ultrasound fusion guided targeted biopsy.

BACKGROUND: To explore the diagnostic benefit of an additional image fusion of the sagittal plane in addition to the standard axial image fusion, using a sensor-based MRI/US fusion platform. METHODS: During July 2013 and September 2015, 251 patients with at least one suspicious lesion on mpMRI (rated by PI-RADS) were included into the analysis. All patients underwent MRI/US targeted biopsy (TB) in combination with a 10 core systematic prostate biopsy (SB). All biopsies were performed on a sensor-based fusion system. Group A included 162 men who received TB by an axial MRI/US image fusion. Group B comprised 89 men in whom the TB was performed with an additional sagittal image fusion. RESULTS: The median age in group A was 67 years (IQR 61-72) and in group B 68 years (IQR 60-71). The median PSA level in group A was 8.10 ng/ml (IQR 6.05-14) and in group B 8.59 ng/ml (IQR 5.65-12.32). In group A the proportion of patients with a suspicious digital rectal examination (DRE) (14 vs. 29%, p = 0.007) and the proportion of primary biopsies (33 vs 46%, p = 0.046) were significantly lower. The rate of PI-RADS 3 lesions were overrepresented in group A compared to group B (19 vs. 9%; p = 0.044). Classified according to PI-RADS 3, 4 and 5, the detection rates of TB were 42, 48, 75% in group A and 25, 74, 90% in group B. The rate of PCa with a Gleason score ≥7 missed by TB was 33% (18 cases) in group A and 9% (5 cases) in group B; p-value 0.072. An explorative multivariate binary logistic regression analysis revealed that PI-RADS, a suspicious DRE and performing an additional sagittal image fusion were significant predictors for PCa detection in TB. 9 PCa were only detected by TB with sagittal fusion (sTB) and sTB identified 10 additional clinically significant PCa (Gleason ≥7). CONCLUSION: Performing an additional sagittal image fusion besides the standard axial fusion appears to improve the accuracy of the sensor-based MRI/US fusion platform.

Multiparametric magnetic resonance imaging: Current role in prostate cancer management.

Digital rectal examination, serum prostate-specific antigen screening and transrectal ultrasound-guided biopsy are conventionally used as screening, diagnostic and surveillance tools for prostate cancer. However, they have limited sensitivity and specificity. In recent years, the role of multiparametric magnetic resonance imaging has steadily grown, and is now part of the standard clinical management in many institutions. In multiparametric magnetic resonance imaging, the morphological assessment of T2-weighted imaging is correlated with diffusion-weighted imaging, dynamic contrast-enhanced imaging perfusion and/or magnetic resonance spectroscopic imaging. Multiparametric magnetic resonance imaging is currently regarded as the most sensitive and specific imaging technique for the evaluation of prostate cancer, including detection, staging, localization and aggressiveness evaluation. This article presents an overview of multiparametric magnetic resonance imaging, and discusses the current role of multiparametric magnetic resonance imaging in the different fields of prostate cancer management.

Prostate Cancer Detection and Prognosis: From Prostate Specific Antigen (PSA) to Exosomal Biomarkers.

Prostate specific antigen (PSA) remains the most used biomarker in the management of early prostate cancer (PCa), in spite of the problems related to false positive results and overdiagnosis. New biomarkers have been proposed in recent years with the aim of increasing specificity and distinguishing aggressive from non-aggressive PCa. The emerging role of the prostate health index and the 4Kscore is reviewed in this article. Both are blood-based tests related to the aggressiveness of the tumor, which provide the risk of suffering PCa and avoiding negative biopsies. Furthermore, the use of urine has emerged as a non-invasive way to identify new biomarkers in recent years, including the PCA3 and TMPRSS2:ERG fusion gene. Available results about the PCA3 score showed its usefulness to decide the repetition of biopsy in patients with a previous negative result, although its relationship with the aggressiveness of the tumor is controversial. More recently, aberrant microRNA expression in PCa has been reported by different authors. Preliminary results suggest the utility of circulating and urinary microRNAs in the detection and prognosis of PCa. Although several of these new biomarkers have been recommended by different guidelines, large prospective and comparative studies are necessary to establish their value in PCa detection and prognosis.

Diagnostic value of biparametric magnetic resonance imaging (MRI) as an adjunct to prostate-specific antigen (PSA)-based detection of prostate cancer in men without prior biopsies.

OBJECTIVES: To determine the diagnostic yield of analysing biparametric (T2- and diffusion-weighted) magnetic resonance imaging (B-MRI) for prostate cancer detection compared with standard digital rectal examination (DRE) and prostate-specific antigen (PSA)-based screening. PATIENTS AND METHODS: Review of patients who were enrolled in a trial to undergo multiparametric-prostate (MP)-MRI and MR/ultrasound fusion-guided prostate biopsy at our institution identified 143 men who underwent MP-MRI in addition to standard DRE and PSA-based prostate cancer screening before any prostate biopsy. Patient demographics, DRE staging, PSA level, PSA density (PSAD), and B-MRI findings were assessed for association with prostate cancer detection on biopsy. RESULTS: Men with detected prostate cancer tended to be older, with a higher PSA level, higher PSAD, and more screen-positive lesions (SPL) on B-MRI. B-MRI performed well for the detection of prostate cancer with an area under the curve (AUC) of 0.80 (compared with 0.66 and 0.74 for PSA level and PSAD, respectively). We derived combined PSA and MRI-based formulas for detection of prostate cancer with optimised thresholds. (i) for PSA and B-MRI: PSA level + 6 x (the number of SPL) > 14 and (ii) for PSAD and B-MRI: 14 × (PSAD) + (the number of SPL) >4.25. AUC for equations 1 and 2 were 0.83 and 0.87 and overall accuracy of prostate cancer detection was 79% in both models. CONCLUSIONS: The number of lesions positive on B-MRI outperforms PSA alone in detection of prostate cancer. Furthermore, this imaging criteria coupled as an adjunct with PSA level and PSAD, provides even more accuracy in detecting clinically significant prostate cancer.

Part II: Effect of different evaluation methods to the application of a computer-aided prostate MRI detection/diagnosis (CADe/CADx) device on reader performance.

INTRODUCTION: The construction and results of a multiple-reader multiple-case prostate MRI study are described and reported to illustrate recommendations for how to standardize artificial intelligence (AI) prostate studies per the review constituting Part I1. METHODS: Our previously reported approach was applied to review and report an IRB approved, HIPAA compliant multiple-reader multiple-case clinical study of 150 bi-parametric prostate MRI studies across 9 readers, measuring physician performance both with and without the use of the recently FDA cleared CADe/CADx software ProstatID. RESULTS: Unassisted reader AUC values ranged from 0.418 - 0.759, with AI assisted AUC values ranging from 0.507 - 0.787. This represented a statistically significant AUC improvement of 0.045 (α = 0.05). A free-response ROC (FROC) analysis similarly demonstrated a statistically significant increase in θ from 0.405 to 0.453 (α = 0.05). The standalone performance of ProstatID performed across all prostate tissues demonstrated an AUC of 0.929, while the standalone lesion level performance of ProstatID at all biopsied locations achieved an AUC of 0.710. CONCLUSION: This study applies and illustrates suggested reporting and standardization methods for prostate AI studies that will make it easier to understand, evaluate and compare between AI studies. Providing radiologists with the ProstatID CADe/CADx software significantly increased diagnostic performance as assessed by both ROC and free-response ROC metrics. Such algorithms have the potential to improve radiologist performance in the detection and localization of clinically significant prostate cancer.

Current Approach to Complications and Difficulties during Transrectal Ultrasound-Guided Prostate Biopsies.

Prostate cancer is one of the most common male malignancies worldwide. It affects middle-aged men (45-60 years) and is the leading cause of cancer-related mortality in Western countries. The TRUS (trans rectal ultrasound)-guided prostate biopsy has been a standard procedure in prostate cancer detection for more than thirty years, and it is recommended in male patients with an abnormal PSA (prostate-specific antigens) or abnormalities found during digital rectal examinations. During this procedure, urologists might encounter difficulties which may cause subsequent complications. This manuscript aims to present both the complications and the technical difficulties that may occur during TRUS-guided prostate biopsy, along with resolutions and solutions found in the specialized literature. The conclusions of this manuscript will note that the TRUS-guided prostate biopsy remains a solid, cost-efficient, and safe procedure with which to diagnose prostate cancer. The complications are usually self-limiting and do not require additional medical assistance. The difficulties posed by the procedure can be safely overcome if there are no other available alternatives. Open communication with the patients improves both pre- and post-procedure compliance.

Deep Learning-based Unsupervised Domain Adaptation via a Unified Model for Prostate Lesion Detection Using Multisite Biparametric MRI Datasets.

Purpose To determine whether the unsupervised domain adaptation (UDA) method with generated images improves the performance of a supervised learning (SL) model for prostate cancer (PCa) detection using multisite biparametric (bp) MRI datasets. Materials and Methods This retrospective study included data from 5150 patients (14 191 samples) collected across nine different imaging centers. A novel UDA method using a unified generative model was developed for PCa detection using multisite bpMRI datasets. This method translates diffusion-weighted imaging (DWI) acquisitions, including apparent diffusion coefficient (ADC) and individual diffusion-weighted (DW) images acquired using various b values, to align with the style of images acquired using b values recommended by Prostate Imaging Reporting and Data System (PI-RADS) guidelines. The generated ADC and DW images replace the original images for PCa detection. An independent set of 1692 test cases (2393 samples) was used for evaluation. The area under the receiver operating characteristic curve (AUC) was used as the primary metric, and statistical analysis was performed via bootstrapping. Results For all test cases, the AUC values for baseline SL and UDA methods were 0.73 and 0.79 (P < .001), respectively, for PCa lesions with PI-RADS score of 3 or greater and 0.77 and 0.80 (P < .001) for lesions with PI-RADS scores of 4 or greater. In the 361 test cases under the most unfavorable image acquisition setting, the AUC values for baseline SL and UDA were 0.49 and 0.76 (P < .001) for lesions with PI-RADS scores of 3 or greater and 0.50 and 0.77 (P < .001) for lesions with PI-RADS scores of 4 or greater. Conclusion UDA with generated images improved the performance of SL methods in PCa lesion detection across multisite datasets with various b values, especially for images acquired with significant deviations from the PI-RADS-recommended DWI protocol (eg, with an extremely high b value). Keywords: Prostate Cancer Detection, Multisite, Unsupervised Domain Adaptation, Diffusion-weighted Imaging, b Value Supplemental material is available for this article. © RSNA, 2024.

Targeted Prostate Biopsy: How, When, and Why? A Systematic Review.

OBJECTIVE: Prostate cancer, the second most diagnosed cancer among men, requires precise diagnostic techniques to ensure effective treatment. This review explores the technological advancements, optimal application conditions, and benefits of targeted prostate biopsies facilitated by multiparametric magnetic resonance imaging (mpMRI). METHODS: A systematic literature review was conducted to compare traditional 12-core systematic biopsies guided by transrectal ultrasound with targeted biopsy techniques using mpMRI. We searched electronic databases including PubMed, Scopus, and Web of Science from January 2015 to December 2024 using keywords such as "targeted prostate biopsy", "fusion prostate biopsy", "cognitive prostate biopsy", "MRI-guided biopsy", and "transrectal ultrasound prostate biopsy". Studies comparing various biopsy methods were included, and data extraction focused on study characteristics, patient demographics, biopsy techniques, diagnostic outcomes, and complications. CONCLUSION: mpMRI-guided targeted biopsies enhance the detection of clinically significant prostate cancer while reducing unnecessary biopsies and the detection of insignificant cancers. These targeted approaches preserve or improve diagnostic accuracy and patient outcomes, minimizing the risks associated with overdiagnosis and overtreatment. By utilizing mpMRI, targeted biopsies allow for precise targeting of suspicious regions within the prostate, providing a cost-effective method that reduces the number of biopsies performed. This review highlights the importance of integrating advanced imaging techniques into prostate cancer diagnosis to improve patient outcomes and quality of life.