Diffusion-weighted imaging for determining response to neoadjuvant therapy in pancreatic cancer: a systematic review and meta-analysis.

OBJECTIVES: To determine the role of diffusion-weighted imaging (DWI) for predicting response to neoadjuvant therapy (NAT) in pancreatic cancer. MATERIALS AND METHODS: MEDLINE, EMBASE, and Cochrane Library databases were searched for studies evaluating the performance of apparent diffusion coefficient (ADC) to assess response to NAT. Data extracted included ADC pre- and post-NAT, for predicting response as defined by imaging, histopathology, or clinical reference standards. ADC values were compared with standardized mean differences. Risk of bias was assessed using the Quality Assessment of Diagnostic Studies (QUADAS-2). RESULTS: Of 337 studies, 7 were included in the analysis (161 patients). ADC values reported for the pre- and post-NAT assessments overlapped between responders and non-responders. One study reported inability of ADC increase after NAT for distinguishing responders and non-responders. A correlation with histopathological response was reported for pre- and post-NAT ADC in 4 studies. DWI's diagnostic performance was reported to be high in three studies, with a 91.6-100% sensitivity and 62.5-94.7% specificity. Finally, heterogeneity and high risk of bias were identified across studies, affecting the domains of patient selection, index test, reference standard, and flow and timing. CONCLUSION: DWI might be useful for determining response to NAT in pancreatic cancer. However, there are still too few studies on this matter, which are also heterogeneous and at high risk for bias. Further studies with standardized procedures for data acquisition and accurate reference standards are needed. CLINICAL RELEVANCE STATEMENT: Diffusion-weighted MRI might be useful for assessing response to neoadjuvant therapy in pancreatic cancer. However, further studies with robust data are needed to provide specific recommendations for clinical practice. KEY POINTS: •The role of DWI with ADC measurements for assessing response to neoadjuvant therapy in pancreatic cancer is still unclear. •Pre- and post-neoadjuvant therapy ADC values overlap between responders and non-responders. •DWI has a reported high diagnostic performance for determining response when using histopathological or clinical reference standards; however, studies are still few and at high risk for bias.

Performances and variability of CT radiomics for the prediction of microvascular invasion and survival in patients with HCC: a matter of chance or standardisation?

OBJECTIVES: To measure the performance and variability of a radiomics-based model for the prediction of microvascular invasion (MVI) and survival in patients with resected hepatocellular carcinoma (HCC), simulating its sequential development and application. METHODS: This study included 230 patients with 242 surgically resected HCCs who underwent preoperative CT, of which 73/230 (31.7%) were scanned in external centres. The study cohort was split into training set (158 patients, 165 HCCs) and held-out test set (72 patients, 77 HCCs), stratified by random partitioning, which was repeated 100 times, and by a temporal partitioning to simulate the sequential development and clinical use of the radiomics model. A machine learning model for the prediction of MVI was developed with least absolute shrinkage and selection operator (LASSO). The concordance index (C-index) was used to assess the value to predict the recurrence-free (RFS) and overall survivals (OS). RESULTS: In the 100-repetition random partitioning cohorts, the radiomics model demonstrated a mean AUC of 0.54 (range 0.44-0.68) for the prediction of MVI, mean C-index of 0.59 (range 0.44-0.73) for RFS, and 0.65 (range 0.46-0.86) for OS in the held-out test set. In the temporal partitioning cohort, the radiomics model yielded an AUC of 0.50 for the prediction of MVI, a C-index of 0.61 for RFS, and 0.61 for OS, in the held-out test set. CONCLUSIONS: The radiomics models had a poor performance for the prediction of MVI with a large variability in the model performance depending on the random partitioning. Radiomics models demonstrated good performance in the prediction of patient outcomes. CLINICAL RELEVANCE STATEMENT: Patient selection within the training set strongly influenced the performance of the radiomics models for predicting microvascular invasion; therefore, a random approach to partitioning a retrospective cohort into a training set and a held-out set seems inappropriate. KEY POINTS: • The performance of the radiomics models for the prediction of microvascular invasion and survival widely ranged (AUC range 0.44-0.68) in the randomly partitioned cohorts. • The radiomics model for the prediction of microvascular invasion was unsatisfying when trying to simulate its sequential development and clinical use in a temporal partitioned cohort imaged with a variety of CT scanners. • The performance of the radiomics models for the prediction of survival was good with similar performances in the 100-repetition random partitioning and temporal partitioning cohorts.

Effective bowel motion reduction in mouse abdominal MRI using hyoscine butylbromide.

PURPOSE: Bowel motion is a significant source of artifacts in mouse abdominal MRI. Fasting and administration of hyoscine butylbromide (BUSC) have been proposed for bowel motion reduction but with inconsistent results and limited efficacy assessments. Here, we evaluate these regimes for mouse abdominal MRI at high field. METHODS: Thirty-two adult C57BL/6J mice were imaged on a 9.4T scanner with a FLASH sequence, acquired over 90 min with ~19 s temporal resolution. During MRI acquisition, 8 mice were injected with a low-dose and 8 mice with a high-dose bolus of BUSC (0.5 and 5 mg/kg, respectively). Eight mice were food deprived for 4.5-6.5 hours before MRI and another group of eight mice was injected with saline during MRI acquisition. Two expert readers reviewed the images and classified bowel motion, and quantitative voxel-wise analyses were performed for identification of moving regions. After defining the most effective protocol, high-resolution T2 -weighted and diffusion-weighted images were acquired from 4 mice. RESULTS: High-dose BUSC was the most effective protocol for bowel motion reduction, for up to 45 min. Fasting and saline protocols were not effective in suppressing bowel motion. High-resolution abdominal MRI clearly demonstrated improved image quality and ADC quantification with the high-dose BUSC protocol. CONCLUSION: Our data show that BUSC administration is advantageous for abdominal MRI in the mouse. Specifically, it endows significant bowel motion reduction, with relatively short onset timings after injection (~8.5 min) and relatively long duration of the effect (~45 min). These features improve the quality of high-resolution images of the mouse abdomen.

Higher-order diffusion MRI characterization of mesorectal lymph nodes in rectal cancer.

PURPOSE: Mesorectal lymph node staging plays an important role in treatment decision making. Here, we explore the benefit of higher-order diffusion MRI models accounting for non-Gaussian diffusion effects to classify mesorectal lymph nodes both 1) ex vivo at ultrahigh field correlated with histology and 2) in vivo in a clinical scanner upon patient staging. METHODS: The preclinical investigation included 54 mesorectal lymph nodes, which were scanned at 16.4 T with an extensive diffusion MRI acquisition. Eight diffusion models were compared in terms of goodness of fit, lymph node classification ability, and histology correlation. In the clinical part of this study, 10 rectal cancer patients were scanned with diffusion MRI at 1.5 T, and 72 lymph nodes were analyzed with Apparent Diffusion Coefficient (ADC), Intravoxel Incoherent Motion (IVIM), Kurtosis, and IVIM-Kurtosis. RESULTS: Compartment models including restricted and anisotropic diffusion improved the preclinical data fit, as well as the lymph node classification, compared to standard ADC. The comparison with histology revealed only moderate correlations, and the highest values were observed between diffusion anisotropy metrics and cell area fraction. In the clinical study, the diffusivity from IVIM-Kurtosis was the only metric showing significant differences between benign (0.80 ± 0.30 µm2 /ms) and malignant (1.02 ± 0.41 µm2 /ms, P = .03) nodes. IVIM-Kurtosis also yielded the largest area under the receiver operating characteristic curve (0.73) and significantly improved the node differentiation when added to the standard visual analysis by experts based on T2 -weighted imaging. CONCLUSION: Higher-order diffusion MRI models perform better than standard ADC and may be of added value for mesorectal lymph node classification in rectal cancer patients.

The split scar sign as an indicator of sustained complete response after neoadjuvant therapy in rectal cancer.

OBJECTIVES: To measure the diagnostic performance of a new radiologic pattern on restaging magnetic resonance (MR) high-resolution T2-weighted imaging (T2-WI)-the split scar sign-for the identification of sustained complete response (SCR) after neoadjuvant therapy in rectal cancer. METHODS: Institutional review board approval was obtained for this retrospective study and the informed consent requirement was waived. Fifty-eight consecutive patients with rectal cancer who underwent neoadjuvant therapy were enrolled. Two radiologists blindly and independently reviewed restaging pelvic MR imaging and recorded the presence/absence of the split scar sign (mrSSS). On a second round, they also assessed the relative proportion of intermediate signal intensity on T2-WI (mrT2) and of high signal intensity on high b-value diffusion-weighted imaging (mrDWI). Endoscopic response grading records were retrieved. Qui-square test was employed in search for associations between SCR, defined as pathologic complete response or long-term recurrence-free clinical follow-up, and mrSSS, mrT2, mrDWI and endoscopy. Interobserver agreement for imaging parameters was estimated using Cohen's kappa (k). RESULTS: mrSSS was significantly associated with SCR, with specificity = 0.97/0.97, sensitivity = 0.52/0.64, PPV = 0.93/0.94, NPV = 0.73/0.78, and AuROC = 0.78/0.83, for observers 1/2, respectively. mrDWI was significantly associated with SCR for observer 2, with specificity = 0.76, sensitivity = 0.60, PPV = 0.65, NPV = 0.71, and AuROC = 0.69. mrT2 and endoscopy were not discriminative. Interobserver agreement was substantial for mrSSS (k = 0.69), moderate for mrDWI (k = 0.46), and poor for mrT2 (k = 0.17). CONCLUSION: The split scar sign is a simple morphologic pattern visible on restaging T2-WI which, although not sensitive, is very specific for the identification of sustained complete responders after neoadjuvant therapy in rectal cancer. KEY POINTS: • The split scar sign is a morphologic pattern visible on high-resolution T2-weighted MR imaging in rectal cancer patients after neoadjuvant therapy. It therefore does not require any changes to standard protocol. • At first restaging pelvic MR imaging (mean: 9.1 weeks after the end of radiotherapy), the split scar sign identified patients who sustained a complete response with very high specificity (0.97) and positive predictive value (0.93-0.94). • The split scar sign has the potential to improve patient selection for "watch-and-wait" after neoadjuvant therapy in rectal cancer.

Pancreatic Adenocarcinoma Staging in the Era of Preoperative Chemotherapy and Radiation Therapy.

Pancreatic ductal adenocarcinoma (PDA) remains among the most challenging malignancies to treat. At diagnosis, the tumor often already extends beyond the confines of the pancreas, spreading to an extent such that primary surgery with curative intent is very rarely feasible. Considerable momentum is now being given to a treatment strategy involving neoadjuvant chemotherapy or chemotherapy and radiation therapy in patients with nonmetastatic PDA. The main advantage of this strategy is better selection of patients likely to benefit from curative-intent surgery through the achievement of negative resection margins. Patients with rapidly progressive disease are identified and are spared ineffective surgery with its attendant morbidity. Neoadjuvant therapy can downstage tumors classified as locally advanced at initial imaging studies to resectable tumors. However, the imaging study evaluation of the response to neoadjuvant therapy is extremely complex. Thus, the diagnostic performance of imaging studies is not sufficient to ensure the accurate selection of patients in whom negative-margin resection is likely to be achieved. More specifically, standard criteria for predicting vascular invasion, based on the amount of tumor-vessel contact, are not valid after neoadjuvant therapy. ©RSNA, 2018.

Added value of diffusion-weighted magnetic resonance imaging for the detection of pancreatic fluid collection infection.

OBJECTIVES: To investigate the added value of diffusion-weighted (DW) magnetic resonance (MR) imaging in the detection of infection in pancreatic fluid collections (PFC). METHODS: Forty-patients with PFC requiring endoscopic-transmural drainage underwent conventional-MR and DW-MR imaging (b = 1000 s/mm2) before endoscopy. MR images were divided into two sets (set1, conventional-MR; set2, conventional-MR, DW-MR and ADC maps) and randomized. Two independent readers performed qualitative and quantitative (apparent diffusion coefficient, ADC) image analysis. Bacteriological analysis of PFC content was the gold standard. Non-parametric tests were used for comparisons. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) and accuracy were calculated for the two sets for both readers. Receiver operating characteristic curves (ROC) were drawn to assess quantitative DW-MR imaging diagnostic performance. RESULTS: For both readers, sensitivity, specificity, NPV, PPV and accuracy for infected PFCs were higher for set2 (P > .05). ADC were lower in infected versus non-infected PFCs (P ≤ .031). Minimum ADC cut-off: 1,090×10-3 mm2/s for reader 1 and 1,012×10-3 mm2/s for reader 2 (sensitivity and specificity 67 % and 96 % for both readers). CONCLUSION: Qualitative information provided by DW-MR may help to assess PFCs infection. Infected PFCs show significantly lower ADCs compared to non-infected ones. KEY POINTS: • DW improves MR diagnostic accuracy to detect infection of PFC • Infected PFCs show lower ADC compared to non-infected ones (P < .031) • DW-MR images are easy to interpret especially for non-experienced radiologist.

Liver apparent diffusion coefficient repeatability with individually predetermined optimal cardiac timing and artifact elimination by signal filtering.

PURPOSE: To prospectively assess liver ADC (apparent diffusion coefficient) repeatability from cardiac-triggered diffusion-weighted images obtained with an individually predetermined optimal cardiac time window minimizing cardiac-related effects and to evaluate a signal filtering method aimed at artifact elimination. MATERIALS AND METHODS: After Institutional Review Board approval and written informed consent, eight healthy volunteers underwent four repetitions of respiratory-triggered diffusion-weighted sequences (3T, b: 0,150,500 s/mm(2) ) without (RTnoCT, 51 sec) and with individually optimized cardiac triggering (RTCT, 306 sec). The optimal cardiac delay was individually predetermined using a 5-second breath-hold sequence. Monoexponential liver ADC and left-to-right-liver ADC ratio were computed from region of interest (ROI) signal measurements (two independent readers). A filtering method, excluding signal intensities lower than the mean intensity at fixed b-value, provided ADC recalculation. Limits-of-agreement (LOAs) from 95% confidence intervals for differences across the four repetitions provided the variability range. RESULTS: For Reader 1 (Reader 2), left-to-right-liver ADC ratios were significantly higher in RTnoCT 1.51 (1.52) than in RTCT 1.12 (1.15), P = 0.012 (P = 0.017). Respectively for RTnoCT and RTCT: left liver LOAs were ±835 (±775), ± 315 (±369) 10(-6) mm(2) /s; right liver LOAs were ±392 (±445), ± 172 (±140) 10(-6) mm(2) /s: LOAs were larger in the left than in the right lobe (both P < 0.001). After filtering, left liver ADC LOAs narrowed to ±650 (±367) 10(-6) mm(2) /s, P = 0.17 (P < 0.001); ± 152 (±208) 10(-6) mm(2) /s (both P < 0.002) and left-to-right-liver ADC ratio decreased to 1.28 (1.20), P = 0.017 (P = 0.012); 1.09 (1.08), P = 0.106 (P = 0.105). CONCLUSION: Compared to noncardiac-triggered acquisitions, individually optimized cardiac-triggered acquisitions improved ADC repeatability in both liver lobes and reduced ADC differences between left and right liver. Left liver ADC repeatability was further improved after signal filtering.

Diffusion-weighted imaging outside the brain: Consensus statement from an ISMRM-sponsored workshop.

The significant advances in magnetic resonance imaging (MRI) hardware and software, sequence design, and postprocessing methods have made diffusion-weighted imaging (DWI) an important part of body MRI protocols and have fueled extensive research on quantitative diffusion outside the brain, particularly in the oncologic setting. In this review, we summarize the most up-to-date information on DWI acquisition and clinical applications outside the brain, as discussed in an ISMRM-sponsored symposium held in April 2015. We first introduce recent advances in acquisition, processing, and quality control; then review scientific evidence in major organ systems; and finally describe future directions. J. Magn. Reson. Imaging 2016;44:521-540.

Endotherapy for paraduodenal pancreatitis: a large retrospective case series.

BACKGROUND AND STUDY AIMS: Paraduodenal pancreatitis is histologically well defined but its epidemiology, natural history, and connection with chronic pancreatitis are not completely understood. The aim of this study was to review the endoscopic and medical management of paraduodenal pancreatitis. PATIENTS AND METHODS: Medical records of all patients with paraduodenal pancreatitis diagnosed by magnetic resonance cholangiopancreatography (MRCP) or endoscopic ultrasonography (EUS) between 1995 and 2010 were retrospectively reviewed. Clinical features, imaging procedures, and treatments were investigated. The primary end point was the rate of clinical success, and the secondary end points were the radiological or endoscopic improvement, complication rate, and overall survival rate. RESULTS: A total of 51 patients were included in the study (88.2 % alcohol abuse; median age 49 years [range 37 - 70]; 50 men). The most frequent symptoms at presentation were pain (n = 50; 98.0 %) and weight loss (n = 36; 70.6 %). Chronic pancreatitis was present in 36 patients (70.6 %), and 45 patients (88.2 %) had cysts. Other findings included stricture of the pancreatic duct (n = 37; 72.5 %), common bile duct (n = 29; 56.9 %), and duodenum (n = 24; 47.1 %). A total of 39 patients underwent initial endoscopic treatment: cystenterostomy (n = 20), pancreatic and/or biliary duct drainage (n = 19), and/or duodenal dilation (n = 6). For the patients with available follow-up (n = 41), 24 patients required repeat endoscopy and 9 patients required surgery after the initial endoscopic management. After a median follow-up of 54 months (range 6 - 156 months), complete clinical success was achieved in 70.7 % of patients, and the overall survival rate was 94.1 %. CONCLUSIONS: This is the largest series concerning the management of paraduodenal pancreatitis using endotherapy as the first-line intervention. Although repeat endoscopic procedures were required in half of the patients, no severe complication was observed and surgical treatment was ultimately needed in less than 25 % of the patients.

Normal hepatic parenchyma visibility and ADC quantification on diffusion-weighted MRI at 3 T: influence of age, gender, and iron content.

OBJECTIVES: To investigate how normal liver parenchyma visibility on 3 T diffusion-weighted images (DWI) and apparent diffusion coefficient (ADC) quantification are influenced by age, gender, and iron content. METHODS: Between February 2011 and April 2013, 86 patients (52 women) with normal livers who underwent respiratory-triggered abdominal 3 T DWI (b = 0, 150, 600, 1,000 s/mm(2)) were retrospectively included. Normal liver and spleen parenchyma visibility was scored independently by two readers. Correlations between visibility scores or ADC with age, gender, T2*, or recent serum ferritin (SF) were investigated. RESULTS: Liver visibility scores in b = 1,000 s/mm(2) images correlated with the age (Spearman R = -0.56 in women, -0.45 in men), T2* (R = 0.75) and SF (R = -0.64) and were significantly higher in women (P < 0.01). SF and T2* were within normal values (T2*: 13 - 31 ms, SF: 14 - 230 µg/L). Liver ADC correlated with visibility scores (R = 0.69) and T2* (R = 0.64) and was age- and gender-dependent. ADC ROI standard deviation negatively correlated with visibility scores (R = -0.65) and T2* (R = -0.62). The spleen visibility did not depend on age or gender. CONCLUSIONS: Normal liver parenchyma visibility in DWI is age- and gender-dependent, according to the iron content. Visibility scores and iron content significantly affect ADC quantification in the normal liver. KEY POINTS: Normal DWI liver visibility is gender-dependent and superior in women. In women, normal DWI liver visibility is superior before age 50 years. Normal DWI liver visibility negatively correlates with normal range iron content markers. Liver ADC quantification depends on liver iron content even within normal range. Normal liver T2* is age- and gender-dependent.

Imaging of pancreatic ductal adenocarcinoma - An update for all stages of patient management.

Background: Pancreatic ductal adenocarcinoma (PDAC) is a common and lethal cancer. From diagnosis to disease staging, response to neoadjuvant therapy assessment and patient surveillance after resection, imaging plays a central role, guiding the multidisciplinary team in decision-planning. Review aims and findings: This review discusses the most up-to-date imaging recommendations, typical and atypical findings, and issues related to each step of patient management. Example cases for each relevant condition are presented, and a structured report for disease staging is suggested. Conclusion: Despite current issues in PDAC imaging at different stages of patient management, the radiologist is essential in the multidisciplinary team, as the conveyor of relevant imaging findings crucial for patient care.

Development of End-to-End AI-Based MRI Image Analysis System for Predicting IDH Mutation Status of Patients with Gliomas: Multicentric Validation.

Radiogenomics has shown potential to predict genomic phenotypes from medical images. The development of models using standard-of-care pre-operative MRI images, as opposed to advanced MRI images, enables a broader reach of such models. In this work, a radiogenomics model for IDH mutation status prediction from standard-of-care MRIs in patients with glioma was developed and validated using multicentric data. A cohort of 142 (wild-type: 32.4%) patients with glioma retrieved from the TCIA/TCGA was used to train a logistic regression model to predict the IDH mutation status. The model was evaluated using retrospective data collected in two distinct hospitals, comprising 36 (wild-type: 63.9%) and 53 (wild-type: 75.5%) patients. Model development utilized ROC analysis. Model discrimination and calibration were used for validation. The model yielded an AUC of 0.741 vs. 0.716 vs. 0.938, a sensitivity of 0.784 vs. 0.739 vs. 0.875, and a specificity of 0.657 vs. 0.692 vs. 1.000 on the training, test cohort 1, and test cohort 2, respectively. The assessment of model fairness suggested an unbiased model for age and sex, and calibration tests showed a p < 0.05. These results indicate that the developed model allows the prediction of the IDH mutation status in gliomas using standard-of-care MRI images and does not appear to hold sex and age biases.

Deep Learning Features Can Improve Radiomics-Based Prostate Cancer Aggressiveness Prediction.

PURPOSE: Emerging evidence suggests that the use of artificial intelligence can assist in the timely detection and optimization of therapeutic approach in patients with prostate cancer. The conventional perspective on radiomics encompassing segmentation and the extraction of radiomic features considers it as an independent and sequential process. However, it is not necessary to adhere to this viewpoint. In this study, we show that besides generating masks from which radiomic features can be extracted, prostate segmentation and reconstruction models provide valuable information in their feature space, which can improve the quality of radiomic signatures models for disease aggressiveness classification. MATERIALS AND METHODS: We perform 2,244 experiments with deep learning features extracted from 13 different models trained using different anatomic zones and characterize how modeling decisions, such as deep feature aggregation and dimensionality reduction, affect performance. RESULTS: While models using deep features from full gland and radiomic features consistently lead to improved disease aggressiveness prediction performance, others are detrimental. Our results suggest that the use of deep features can be beneficial, but an appropriate and comprehensive assessment is necessary to ensure that their inclusion does not harm predictive performance. CONCLUSION: The study findings reveal that incorporating deep features derived from autoencoder models trained to reconstruct the full prostate gland (both zonal models show worse performance than radiomics only models), combined with radiomic features, often lead to a statistically significant increase in model performance for disease aggressiveness classification. Additionally, the results also demonstrate that the choice of feature selection is key to achieving good performance, with principal component analysis (PCA) and PCA + relief being the best approaches and that there is no clear difference between the three proposed latent representation extraction techniques.

Analysis of domain shift in whole prostate gland, zonal and lesions segmentation and detection, using multicentric retrospective data.

Despite being one of the most prevalent forms of cancer, prostate cancer (PCa) shows a significantly high survival rate, provided there is timely detection and treatment. Computational methods can help make this detection process considerably faster and more robust. However, some modern machine-learning approaches require accurate segmentation of the prostate gland and the index lesion. Since performing manual segmentations is a very time-consuming task, and highly prone to inter-observer variability, there is a need to develop robust semi-automatic segmentation models. In this work, we leverage the large and highly diverse ProstateNet dataset, which includes 638 whole gland and 461 lesion segmentation masks, from 3 different scanner manufacturers provided by 14 institutions, in addition to other 3 independent public datasets, to train accurate and robust segmentation models for the whole prostate gland, zones and lesions. We show that models trained on large amounts of diverse data are better at generalizing to data from other institutions and obtained with other manufacturers, outperforming models trained on single-institution single-manufacturer datasets in all segmentation tasks. Furthermore, we show that lesion segmentation models trained on ProstateNet can be reliably used as lesion detection models.

Pancreatic tuberculosis diagnosed by EUS: one disease, many faces.

CONTEXT: Pancreatic/para-pancreatic tuberculosis is an extremely rare clinical entity even in endemic regions. It can present as a cystic or solid pancreatic mass mimicking pancreatic malignancy. There are no specific imaging criteria and the clinical symptoms remain vague. Therefore, most cases are diagnosed after surgical exploration for presumed pancreatic neoplasia. CASE REPORT: We report five cases of pancreatic tuberculosis each time with a different clinical presentation, in an occidental country setting where the diagnosis was done by EUS guided FNA (EUS-FNA). CONCLUSION: EUS-FNA is a safe and promising technique for the diagnosis of pancreatic/para-pancreatic tuberculosis, avoiding unnecessary surgery.

Gastric schwannoma: Two case reports and review of the literature.

Gastric schwannomas are rare, slow-growing tumors whose clinical presentation is nonspecific. These are mostly benign, with a low probability of malignant transformation and an excellent prognosis. We present 2 cases of gastric schwannomas with distinct clinical features and imaging patterns, whose therapeutic approach differed. Case 1 is a 73-year-old woman with a voluminous subepithelial lesion in the greater gastric curvature, with predominantly endoluminal growth. Clinically the patient presented with nonspecific abdominal complaints and underwent complete surgical excision. Case 2 is a 69-year-old woman with an exophytic lesion adjacent to the gastric antrum, diagnosed incidentally and managed conservatively, with imaging follow-up, for the last 5 years and stable ever since. This article aims to focus on this rare disease, illustrating its main imaging findings, particularly in magnetic resonance imaging, along with pathological correlation, as well as reviewing the literature, discussing the differential diagnosis, and exploring clinical management and prognosis.

Value of handcrafted and deep radiomic features towards training robust machine learning classifiers for prediction of prostate cancer disease aggressiveness.

There is a growing piece of evidence that artificial intelligence may be helpful in the entire prostate cancer disease continuum. However, building machine learning algorithms robust to inter- and intra-radiologist segmentation variability is still a challenge. With this goal in mind, several model training approaches were compared: removing unstable features according to the intraclass correlation coefficient (ICC); training independently with features extracted from each radiologist's mask; training with the feature average between both radiologists; extracting radiomic features from the intersection or union of masks; and creating a heterogeneous dataset by randomly selecting one of the radiologists' masks for each patient. The classifier trained with this last resampled dataset presented with the lowest generalization error, suggesting that training with heterogeneous data leads to the development of the most robust classifiers. On the contrary, removing features with low ICC resulted in the highest generalization error. The selected radiomics dataset, with the randomly chosen radiologists, was concatenated with deep features extracted from neural networks trained to segment the whole prostate. This new hybrid dataset was then used to train a classifier. The results revealed that, even though the hybrid classifier was less overfitted than the one trained with deep features, it still was unable to outperform the radiomics model.