BACKGROUND: Numerous studies have shown that magnetic resonance imaging (MRI)-targeted biopsy approaches are superior to traditional systematic transrectal ultrasound guided biopsy (TRUS-Bx). The optimal number of biopsy cores to be obtained per lesion identified on multiparametric MRI (mpMRI) images, however, remains a matter of debate. The aim of this study was to evaluate the incremental value of additional biopsy cores in an MRI-targeted \"in-bore\"-biopsy (MRI-Bx) setting.
METHODS: Two hundred and forty-five patients, who underwent MRI-Bx between June 2014 and September 2021, were included in this retrospective single-center analysis. All lesions were biopsied with at least five biopsy cores and cumulative detection rates for any cancer (PCa) as well as detection rates of clinically significant cancers (csPCa) were calculated for each sequentially labeled biopsy core. The cumulative per-core detection rates are presented as whole numbers and as proportion of the maximum detection rate reached, when all biopsy cores were considered. CsPCa was defined as Gleason Score (GS) ≥ 7 (3 + 4).
RESULTS: One hundred and thirty-two of 245 Patients (53.9%) were diagnosed with prostate cancer and csPCa was found in 64 (26.1%) patients. The first biopsy core revealed csPCa/ PCa in 76.6% (49/64)/ 81.8% (108/132) of cases. The second, third and fourth core found csPCa/ PCa not detected by previous cores in 10.9% (7/64)/ 8.3% (11/132), 7.8% (5/64)/ 5.3% (7/132) and 3.1% (2/64)/ 3% (4/132) of cases, respectively. Obtaining one or more cores beyond the fourth biopsy core resulted in an increase in detection rate of 1.6% (1/64)/ 1.5% (2/132).
CONCLUSIONS: We found that obtaining five cores per lesion maximized detection rates. If, however, future research should establish a clear link between the incidence of serious complications and the number of biopsy cores obtained, a three-core biopsy might suffice as our results suggest that about 95% of all csPCa are detected by the first three cores.

Prostate cancer is one of the most common and fatal diseases among men, and its early diagnosis can have a significant impact on the treatment process and prevent mortality. Since it does not have apparent clinical symptoms in the early stages, it is difficult to diagnose. In addition, the disagreement of experts in the analysis of magnetic resonance images is also a significant challenge. In recent years, various research has shown that deep learning, especially convolutional neural networks, has appeared successfully in machine vision (especially in medical image analysis). In this research, a deep learning approach was used on multi-parameter magnetic resonance images, and the synergistic effect of clinical and pathological data on the accuracy of the model was investigated. The data were collected from Trita Hospital in Tehran, which included 343 patients (data augmentation and learning transfer methods were used during the process). In the designed model, four different types of images are analyzed with four separate ResNet50 deep convolutional networks, and their extracted features are transferred to a fully connected neural network and combined with clinical and pathological features. In the model without clinical and pathological data, the maximum accuracy reached 88%, but by adding these data, the accuracy increased to 96%, which shows the significant impact of clinical and pathological data on the accuracy of diagnosis.

BACKGROUND: To evaluate the accuracy of PSMA PET/CT in men with mpMRI PI-RADS score 5 negative biopsy histology.
METHODS: From January 2011 to January 2023, 180 men with PI-RADS score 5 underwent systematic plus mpMRI/TRUS biopsy; 25/180 (13.9%) patients had absence of cancer and six months from biopsy were submitted to: digital rectal examination, PSA and PSA density exams, mpMRI and 68GaPSMA PET/CT evaluation (standardized uptake value \"SUVmax\" was reported).
RESULTS: In 24/25 (96%) patients PSA and PSA density significantly decreased, moreover, the PI-RADS score was downgraded resulting < 3; in addition, median SUVmax was 7.5. Only 1/25 (4%) man had an increased PSA value (from 10.5 to 31 ng/ml) with a confirmed PI-RADS score 5, SUVmax of 32 and repeated prostate biopsy demonstrating a Gleason score 9/ISUP Grade Group 5 PCa.
CONCLUSIONS: The strict follow up of men with PI-RADS score 5 and negative histology reduce the risk of missing csPCa especially if PSMA PET/CT evaluation is in agreement with downgrading of mpMRI (PI-RADS score < 3).

BACKGROUND: The follow-up findings of patients who underwent prostate biopsy for prostate image reporting and data system (PIRADS) 4 or 5 multiparametric magnetic resonance imaging (mpMRI) findings and had benign histology were retrospectively reviewed.
METHODS: There were 190 biopsy-naive patients. Patients with at least 12 months of follow-up between 2012 and 2023 were evaluated. All MRIs were interpreted by two very experienced uroradiologists. Of the patients, 125 had either cognitive or software fusion MR-targeted biopsies with 4 + 8/10 cores. The remaining 65 patients had in-bore biopsies with 4-5 cores. Prostate-specific antigen (PSA) levels below 4 ng/mL were defined as PSA regression following biopsy. PIRADS 1-3 lesions on new MRI images were classified as MRI regression.
RESULTS: Median patient age and PSA were 62 (39-82) years and six (0.4-33) ng/mL, respectively, at the initial work-up. During a median follow-up period of 44 months, 37 (19.4%) patients were lost to follow-up. Of the remaining 153 patients, 82 (53.6%) had persistently high PSA. Among them, 72 (87.8%) had repeat mpMRI within 6-24 months which showed regressive findings (PIRADS 1-3) in 53 patients (73.6%) and PIRADS 4-5 index lesion persistence in 19 cases (26.4%). The latter group was recommended to have rebiopsy. Of these 19 patients, 16 underwent MRI-targeted rebiopsy. Prostate cancer was diagnosed in six (37.5%) patients and of these four (25%) were clinically significant (>Grade Group 1). Totally, clinically significant prostate cancer was detected in 4/153 (2.6%) patients followed up.
CONCLUSIONS: Patients should be warned against the relative relaxing effect of a negative biopsy after identification of PIRADS 4-5 index lesion. While PSA decrease was observed in many patients during follow-up, persistent MRI findings were present in nearly a quarter of patients with persistently high PSA. A rebiopsy is warranted in these patients, with significant prostate cancer diagnosed in a quarter of patients with rebiopsy.

The Vesical Imaging-Reporting and Data System (VI-RADS) is a standard magnetic resonance imaging (MRI) and diagnostic method for muscle-invasive bladder cancer that was published in 2018. Several studies have demonstrated that VI-RADS has high diagnostic power and reproducibility. However, reading VI-RADS requires a certain amount of expertise, and radiologists need to be aware of the various pitfalls. MRI of the bladder includes T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced imaging (DCEI). T2WI is excellent for understanding anatomy. DWI and DCEI show high contrast between the tumor and normal anatomical structures and are suitable for staging local tumors. Bladder tumors are classified into five categories according to their size and morphology and their positional relationship to the bladder wall based on the VI-RADS diagnostic criteria. If the T2WI, DWI, and DCEI categories are the same, the category is the VI-RADS category. If the categories do not match, the DWI category is the VI-RADS category. If image quality of DWI is not evaluable, the DCEI category is the final category. In many cases, DWI is dominant, but this does not mean that T2WI and DCEI can be omitted from the reading of the bladder. In this educational review, typical and atypical teaching cases are demonstrated, and how to resolve misdiagnosis and the limitations of VI-RADS are discussed. The most important aspect of VI-RADS reading is to practice multiparametric reading with a solid understanding of the characteristics and role of each sequence and an awareness of the various pitfalls.

OBJECTIVE: A CSF-in gradient in cortical and thalamic gray matter (GM) damage has been found in multiple sclerosis (MS). We concomitantly explored the patterns of cortical, thalamic, and caudate microstructural abnormalities at progressive distances from CSF using a multiparametric MRI approach.
METHODS: For this cross-sectional study, from 3T 3D T1-weighted scans, we sampled cortical layers at 25%-50%-75% depths from pial surface and thalamic and caudate bands at 2-3-4 voxels from the ventricular-GM interface. Using linear mixed models, we tested between-group comparisons of magnetization transfer ratio (MTR) and R2* layer-specific z-scores, CSF-in across-layer z-score changes, and their correlations with clinical (disease duration and disability) and structural (focal lesions, brain, and choroid plexus volume) MRI measures.
RESULTS: We enrolled 52 patients with MS (33 relapsing-remitting [RRMS], 19 progressive [PMS], mean age: 46.4 years, median disease duration: 15.1 years, median: EDSS 2.0) and 70 controls (mean age 41.5 ± 12.8). Compared with controls, RRMS showed lower MTR values in the outer and middle cortical layers (false-discovery rate [FDR]-p ≤ 0.025) and lower R2* values in all 3 cortical layers (FDR-p ≤ 0.016). PMS had lower MTR values in the outer and middle cortical (FDR-p ≤ 0.016) and thalamic (FDR-p ≤ 0.048) layers, and in the outer caudate layer (FDR-p = 0.024). They showed lower R2* values in the outer cortical layer (FDR-p = 0.003) and in the outer thalamic layer (FDR-p = 0.046) and higher R2* values in all 3 caudate layers (FDR-p ≤ 0.031). Both RRMS and PMS had a gradient of damage, with lower values closer to the CSF, for cortical (FDR-p ≤ 0.002) and thalamic (FDR-p ≤ 0.042) MTR. PMS showed a gradient of damage for cortical R2* (FDR-p = 0.005), thalamic R2* (FDR-p = 0.004), and caudate MTR (FDR-p ≤ 0.013). Lower MTR and R2* of outer cortical, thalamic, and caudate layers and steeper gradient of damage toward the CSF were significantly associated with older age, higher T2-hyperintense white matter lesion volume, higher thalamic lesion volume, and lower brain volume (β ≥ 0.08, all FDR-p ≤ 0.040). Lower MTR of outer caudate layer was associated with more severe disability (β = -0.26, FDR-p = 0.040). No correlations with choroid plexus volume were found.
CONCLUSIONS: CSF-in damage gradients are heterogeneous among different GM regions and through MS course, possibly reflecting different dynamics of demyelination and iron loss/accumulation.

UNASSIGNED: This study aimed to verify the feasibility and short-term prognosis of prostatectomy without biopsy.
UNASSIGNED: Patients with a rising PSA level ranging from 4 to 30 ng/mL were scheduled for multiparametric (mp) MRI and 18F-labeled prostate-specific membrane antigen (PSMA) positron emission tomography (PET). Forty-seven patients (cT2N0M0) with Prostate Imaging Reporting and Data System ≥ 4 and molecular imaging PSMA score ≥ 2 were enrolled. All candidates underwent robot-assisted laparoscopic radical prostatectomy without biopsy. Prostate cancer detection rate, index tumors localization correspondence rate, positive surgical margin, complications, postoperative hospital stay, and PSA level in a 6-week postoperative follow-up visit were collected.
UNASSIGNED: All the patients with positive mpMRI and PSMA PET were diagnosed with clinically significant prostate cancer. A total of 80 lesions were verified as cancer by pathology, of which 63 cancer lesions were clinically significant prostate cancer. Fifty-one lesions were simultaneously found by mpMRI and PSMA PET. A total of 23 lesions were invisible on either image, and all lesions were ≤ International Society of Urological Pathology 2 or ≤ 15 mm. Forty-five (95.7%) index tumors found by mpMRI combined with PSMA PET were consistent with pathology. Nine patients reported positive surgical margin.
UNASSIGNED: Biopsy-free prostatectomy is safe and feasible for patients with evaluation strictly by mpMRI combined with 18F-PSMA PET/CT.

In addition to morphology and tissue perfusion, diffusion-weighted imaging (DWI) is the third pillar of multiparametric diagnostics in oncology. Due to the strong correlation between the apparent diffusion coefficient (ADC) and cell count in hepatocellular carcinoma (HCC), it can be used as a surrogate marker for tumor cell quantity. Therefore, ADC effectively reflects the effects of cytoreductive treatment, such as transarterial chemoembolization (TACE) and systemic chemotherapy and becomes an important clinical marker for treatment response. The DWI should remain an integral part of a magnetic resonance imaging (MRI) protocol in primary HCC diagnostics and treatment monitoring but is of secondary clinical importance compared to contrast-enhanced MRI perfusion sequences and the use of liver-specific contrast agents. For the future, standardization of DWI sequences for better comparability of various study protocols would be desirable.
UNASSIGNED: Neben Morphologie und Gewebeperfusion ist die Diffusionsbildgebung („diffusion-weighted imaging“, DWI) die dritte Säule der multiparametrischen Diagnostik in der Onkologie. Aufgrund der starken Korrelation zwischen apparentem Diffusionskoeffizienten (ADC) und Zellzahl beim hepatozellulären Karzinom (HCC) kann dieser als Surrogatmarker für die Tumorzellmenge dienen. Daher spiegelt der ADC effektiv die Auswirkungen zytoreduzierender Behandlungen wie transarterielle Chemoembolisation und systemische Chemotherapie wider und wird zu einem wichtigen klinischen Marker des Therapieansprechens. Die DWI sollte in HCC-Primärdiagnostik und Therapieverlaufskontrolle ein fester Bestandteil eines MRT(Magnetresonanztomographie)-Protokolls bleiben, ist jedoch in der klinischen Bedeutung nachrangig gegenüber kontrastmittelgestützten MRT-Perfusionssequenzen und dem Einsatz leberspezifischer Kontrastmittel. Für die Zukunft wäre eine Standardisierung von DWI-Sequenzen zur besseren Vergleichbarkeit verschiedener Studienprotokolle wünschenswert.

Prediction of glioma is crucial to provide a precise treatment plan to optimize the prognosis of children with glioma. However, studies on the grading of pediatric gliomas using radiomics are limited. Meanwhile, existing methods are mainly based on only radiomics features, ignoring intuitive information about tumor morphology on traditional imaging features. This study aims to utilize multiparametric magnetic resonance imaging (MRI) to identify high-grade and low-grade gliomas in children and establish a classification model based on radiomics features and clinical features. A total of 85 children with gliomas underwent tumor resection, and part of the tumor tissue was examined pathologically. Patients were categorized into high-grade and low-grade groups according to World Health Organization guidelines. Preoperative multiparametric MRI data, including contrast-enhanced T1-weighted imaging, T2-weighted imaging, T2-weighted fluid-attenuated inversion recovery, diffusion-weighted images, and apparent diffusion coefficient sequences, were obtained and labeled by two radiologists. The images were preprocessed, and radiomics features were extracted for each MRI sequence. Feature selection methods were used to select radiomics features, and statistically significant clinical features were identified using t-tests. The selected radiomics features and conventional MRI features were used to train the AutoGluon models. The improved model, based on radiomics features and conventional MRI features, achieved a balanced classification accuracy of 66.59%. The cross-validated areas under the receiver operating characteristic curve for the classifier of AutoGluon frame were 0.8071 on the test dataset. The results indicate that the performance of AutoGluon models can be improved by incorporating conventional MRI features, highlighting the importance of the experience of radiologists in accurately grading pediatric gliomas. This method can help predict the grade of pediatric glioma before pathological examination and assist in determining the appropriate treatment plan, including radiotherapy, chemotherapy, drugs, and gene surgery.

UNASSIGNED: To develop a deep learning model based on contrast-enhanced magnetic resonance imaging (MRI) data to predict post-surgical overall survival (OS) in patients with hepatocellular carcinoma (HCC).
UNASSIGNED: This bi-center retrospective study included 564 surgically resected patients with HCC and divided them into training (326), testing (143), and external validation (95) cohorts. This study used a three-dimensional convolutional neural network (3D-CNN) ResNet to learn features from the pretreatment MR images (T1WIpre, late arterial phase, and portal venous phase) and got the deep learning score (DL score). Three cox regression models were established separately using the DL score (3D-CNN model), clinical features (clinical model), and a combination of above (combined model). The concordance index (C-index) was used to evaluate model performance.
UNASSIGNED: We trained a 3D-CNN model to get DL score from samples. The C-index of the 3D-CNN model in predicting 5-year OS for the training, testing, and external validation cohorts were 0.746, 0.714, and 0.698, respectively, and were higher than those of the clinical model, which were 0.675, 0.674, and 0.631, respectively (P = 0.009, P = 0.204, and P = 0.092, respectively). The C-index of the combined model for testing and external validation cohorts was 0.750 and 0.723, respectively, significantly higher than the clinical model (P = 0.017, P = 0.016) and the 3D-CNN model (P = 0.029, P = 0.036).
UNASSIGNED: The combined model integrating the DL score and clinical factors showed a higher predictive value than the clinical and 3D-CNN models and may be more useful in guiding clinical treatment decisions to improve the prognosis of patients with HCC.