Co-ordinate-based positional embedding that captures resolution to enhance transformer's performance in medical image analysis.

Vision transformers (ViTs) have revolutionized computer vision by employing self-attention instead of convolutional neural networks and demonstrated success due to their ability to capture global dependencies and remove spatial biases of locality. In medical imaging, where input data may differ in size and resolution, existing architectures require resampling or resizing during pre-processing, leading to potential spatial resolution loss and information degradation. This study proposes a co-ordinate-based embedding that encodes the geometry of medical images, capturing physical co-ordinate and resolution information without the need for resampling or resizing. The effectiveness of the proposed embedding is demonstrated through experiments with UNETR and SwinUNETR models for infarct segmentation on MRI dataset with AxTrace and AxADC contrasts. The dataset consists of 1142 training, 133 validation and 143 test subjects. Both models with the addition of co-ordinate based positional embedding achieved substantial improvements in mean Dice score by 6.5% and 7.6%. The proposed embedding showcased a statistically significant advantage p-value< 0.0001 over alternative approaches. In conclusion, the proposed co-ordinate-based pixel-wise positional embedding method offers a promising solution for Transformer-based models in medical image analysis. It effectively leverages physical co-ordinate information to enhance performance without compromising spatial resolution and provides a foundation for future advancements in positional embedding techniques for medical applications.

Artificial Intelligence with Statistical Confidence Scores for Detection of Acute or Subacute Hemorrhage on Noncontrast CT Head Scans.

Purpose: To present a method that automatically detects, subtypes, and locates acute or subacute intracranial hemorrhage (ICH) on noncontrast CT (NCCT) head scans; generates detection confidence scores to identify high-confidence data subsets with higher accuracy; and improves radiology worklist prioritization. Such scores may enable clinicians to better use artificial intelligence (AI) tools. Materials and Methods: This retrospective study included 46 057 studies from seven "internal" centers for development (training, architecture selection, hyperparameter tuning, and operating-point calibration; n = 25 946) and evaluation (n = 2947) and three "external" centers for calibration (n = 400) and evaluation (n = 16 764). Internal centers contributed developmental data, whereas external centers did not. Deep neural networks predicted the presence of ICH and subtypes (intraparenchymal, intraventricular, subarachnoid, subdural, and/or epidural hemorrhage) and segmentations per case. Two ICH confidence scores are discussed: a calibrated classifier entropy score and a Dempster-Shafer score. Evaluation was completed by using receiver operating characteristic curve analysis and report turnaround time (RTAT) modeling on the evaluation set and on confidence score-defined subsets using bootstrapping. Results: The areas under the receiver operating characteristic curve for ICH were 0.97 (0.97, 0.98) and 0.95 (0.94, 0.95) on internal and external center data, respectively. On 80% of the data stratified by calibrated classifier and Dempster-Shafer scores, the system improved the Youden indexes, increasing them from 0.84 to 0.93 (calibrated classifier) and from 0.84 to 0.92 (Dempster-Shafer) for internal centers and increasing them from 0.78 to 0.88 (calibrated classifier) and from 0.78 to 0.89 (Dempster-Shafer) for external centers (P < .001). Models estimated shorter RTAT for AI-prioritized worklists with confidence measures than for AI-prioritized worklists without confidence measures, shortening RTAT by 27% (calibrated classifier) and 27% (Dempster-Shafer) for internal centers and shortening RTAT by 25% (calibrated classifier) and 27% (Dempster-Shafer) for external centers (P < .001). Conclusion: AI that provided statistical confidence measures for ICH detection on NCCT scans reliably detected and subtyped hemorrhages, identified high-confidence predictions, and improved worklist prioritization in simulation.Keywords: CT, Head/Neck, Hemorrhage, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2022.

Evaluating deep learning methods in detecting and segmenting different sizes of brain metastases on 3D post-contrast T1-weighted images.

Purpose: We investigate the impact of various deep-learning-based methods for detecting and segmenting metastases with different lesion volume sizes on 3D brain MR images. Approach: A 2.5D U-Net and a 3D U-Net were selected. We also evaluated weak learner fusion of the prediction features generated by the 2.5D and the 3D networks. A 3D fully convolutional one-stage (FCOS) detector was selected as a representative of bounding-box regression-based detection methods. A total of 422 3D post-contrast T1-weighted scans from patients with brain metastases were used. Performances were analyzed based on lesion volume, total metastatic volume per patient, and number of lesions per patient. Results: The performance of detection of the 2.5D and 3D U-Net methods had recall of > 0.83 and precision of > 0.44 for lesion volume > 0.3 cm 3 but deteriorated as metastasis size decreased below 0.3 cm 3 to 0.58 to 0.74 in recall and 0.16 to 0.25 in precision. Compared the two U-Nets for detection capability, high precision was achieved by the 2.5D network, but high recall was achieved by the 3D network for all lesion sizes. The weak learner fusion achieved a balanced performance between the 2.5D and 3D U-Nets; particularly, it increased precision to 0.83 for lesion volumes of 0.1 to 0.3 cm 3 but decreased recall to 0.59. The 3D FCOS detector did not outperform the U-Net methods in detecting either the small or large metastases presumably because of the limited data size. Conclusions: Our study provides the performances of four deep learning methods in relationship to lesion size, total metastasis volume, and number of lesions per patient, providing insight into further development of the deep learning networks.

Machine learning automatically detects COVID-19 using chest CTs in a large multicenter cohort.

OBJECTIVES: To investigate machine learning classifiers and interpretable models using chest CT for detection of COVID-19 and differentiation from other pneumonias, interstitial lung disease (ILD) and normal CTs. METHODS: Our retrospective multi-institutional study obtained 2446 chest CTs from 16 institutions (including 1161 COVID-19 patients). Training/validation/testing cohorts included 1011/50/100 COVID-19, 388/16/33 ILD, 189/16/33 other pneumonias, and 559/17/34 normal (no pathologies) CTs. A metric-based approach for the classification of COVID-19 used interpretable features, relying on logistic regression and random forests. A deep learning-based classifier differentiated COVID-19 via 3D features extracted directly from CT attenuation and probability distribution of airspace opacities. RESULTS: Most discriminative features of COVID-19 are the percentage of airspace opacity and peripheral and basal predominant opacities, concordant with the typical characterization of COVID-19 in the literature. Unsupervised hierarchical clustering compares feature distribution across COVID-19 and control cohorts. The metrics-based classifier achieved AUC = 0.83, sensitivity = 0.74, and specificity = 0.79 versus respectively 0.93, 0.90, and 0.83 for the DL-based classifier. Most of ambiguity comes from non-COVID-19 pneumonia with manifestations that overlap with COVID-19, as well as mild COVID-19 cases. Non-COVID-19 classification performance is 91% for ILD, 64% for other pneumonias, and 94% for no pathologies, which demonstrates the robustness of our method against different compositions of control groups. CONCLUSIONS: Our new method accurately discriminates COVID-19 from other types of pneumonia, ILD, and CTs with no pathologies, using quantitative imaging features derived from chest CT, while balancing interpretability of results and classification performance and, therefore, may be useful to facilitate diagnosis of COVID-19. KEY POINTS: • Unsupervised clustering reveals the key tomographic features including percent airspace opacity and peripheral and basal opacities most typical of COVID-19 relative to control groups. • COVID-19-positive CTs were compared with COVID-19-negative chest CTs (including a balanced distribution of non-COVID-19 pneumonia, ILD, and no pathologies). Classification accuracies for COVID-19, pneumonia, ILD, and CT scans with no pathologies are respectively 90%, 64%, 91%, and 94%. • Our deep learning (DL)-based classification method demonstrates an AUC of 0.93 (sensitivity 90%, specificity 83%). Machine learning methods applied to quantitative chest CT metrics can therefore improve diagnostic accuracy in suspected COVID-19, particularly in resource-constrained environments.

Automated detection of critical findings in multi-parametric brain MRI using a system of 3D neural networks.

With the rapid growth and increasing use of brain MRI, there is an interest in automated image classification to aid human interpretation and improve workflow. We aimed to train a deep convolutional neural network and assess its performance in identifying abnormal brain MRIs and critical intracranial findings including acute infarction, acute hemorrhage and mass effect. A total of 13,215 clinical brain MRI studies were categorized to training (74%), validation (9%), internal testing (8%) and external testing (8%) datasets. Up to eight contrasts were included from each brain MRI and each image volume was reformatted to common resolution to accommodate for differences between scanners. Following reviewing the radiology reports, three neuroradiologists assigned each study to abnormal vs normal, and identified three critical findings including acute infarction, acute hemorrhage, and mass effect. A deep convolutional neural network was constructed by a combination of localization feature extraction (LFE) modules and global classifiers to identify the presence of 4 variables in brain MRIs including abnormal, acute infarction, acute hemorrhage and mass effect. Training, validation and testing sets were randomly defined on a patient basis. Training was performed on 9845 studies using balanced sampling to address class imbalance. Receiver operating characteristic (ROC) analysis was performed. The ROC analysis of our models for 1050 studies within our internal test data showed AUC/sensitivity/specificity of 0.91/83%/86% for normal versus abnormal brain MRI, 0.95/92%/88% for acute infarction, 0.90/89%/81% for acute hemorrhage, and 0.93/93%/85% for mass effect. For 1072 studies within our external test data, it showed AUC/sensitivity/specificity of 0.88/80%/80% for normal versus abnormal brain MRI, 0.97/90%/97% for acute infarction, 0.83/72%/88% for acute hemorrhage, and 0.87/79%/81% for mass effect. Our proposed deep convolutional network can accurately identify abnormal and critical intracranial findings on individual brain MRIs, while addressing the fact that some MR contrasts might not be available in individual studies.

Prediction of Patient Management in COVID-19 Using Deep Learning-Based Fully Automated Extraction of Cardiothoracic CT Metrics and Laboratory Findings.

OBJECTIVE: To extract pulmonary and cardiovascular metrics from chest CTs of patients with coronavirus disease 2019 (COVID-19) using a fully automated deep learning-based approach and assess their potential to predict patient management. MATERIALS AND METHODS: All initial chest CTs of patients who tested positive for severe acute respiratory syndrome coronavirus 2 at our emergency department between March 25 and April 25, 2020, were identified (n = 120). Three patient management groups were defined: group 1 (outpatient), group 2 (general ward), and group 3 (intensive care unit [ICU]). Multiple pulmonary and cardiovascular metrics were extracted from the chest CT images using deep learning. Additionally, six laboratory findings indicating inflammation and cellular damage were considered. Differences in CT metrics, laboratory findings, and demographics between the patient management groups were assessed. The potential of these parameters to predict patients' needs for intensive care (yes/no) was analyzed using logistic regression and receiver operating characteristic curves. Internal and external validity were assessed using 109 independent chest CT scans. RESULTS: While demographic parameters alone (sex and age) were not sufficient to predict ICU management status, both CT metrics alone (including both pulmonary and cardiovascular metrics; area under the curve [AUC] = 0.88; 95% confidence interval [CI] = 0.79-0.97) and laboratory findings alone (C-reactive protein, lactate dehydrogenase, white blood cell count, and albumin; AUC = 0.86; 95% CI = 0.77-0.94) were good classifiers. Excellent performance was achieved by a combination of demographic parameters, CT metrics, and laboratory findings (AUC = 0.91; 95% CI = 0.85-0.98). Application of a model that combined both pulmonary CT metrics and demographic parameters on a dataset from another hospital indicated its external validity (AUC = 0.77; 95% CI = 0.66-0.88). CONCLUSION: Chest CT of patients with COVID-19 contains valuable information that can be accessed using automated image analysis. These metrics are useful for the prediction of patient management.

Automated Quantification of CT Patterns Associated with COVID-19 from Chest CT.

PURPOSE: To present a method that automatically segments and quantifies abnormal CT patterns commonly present in coronavirus disease 2019 (COVID-19), namely ground glass opacities and consolidations. MATERIALS AND METHODS: In this retrospective study, the proposed method takes as input a non-contrasted chest CT and segments the lesions, lungs, and lobes in three dimensions, based on a dataset of 9749 chest CT volumes. The method outputs two combined measures of the severity of lung and lobe involvement, quantifying both the extent of COVID-19 abnormalities and presence of high opacities, based on deep learning and deep reinforcement learning. The first measure of (PO, PHO) is global, while the second of (LSS, LHOS) is lobewise. Evaluation of the algorithm is reported on CTs of 200 participants (100 COVID-19 confirmed patients and 100 healthy controls) from institutions from Canada, Europe and the United States collected between 2002-Present (April, 2020). Ground truth is established by manual annotations of lesions, lungs, and lobes. Correlation and regression analyses were performed to compare the prediction to the ground truth. RESULTS: Pearson correlation coefficient between method prediction and ground truth for COVID-19 cases was calculated as 0.92 for PO (P < .001), 0.97 for PHO(P < .001), 0.91 for LSS (P < .001), 0.90 for LHOS (P < .001). 98 of 100 healthy controls had a predicted PO of less than 1%, 2 had between 1-2%. Automated processing time to compute the severity scores was 10 seconds per case compared to 30 minutes required for manual annotations. CONCLUSION: A new method segments regions of CT abnormalities associated with COVID-19 and computes (PO, PHO), as well as (LSS, LHOS) severity scores.

Automated Quantification of CT Patterns Associated with COVID-19 from Chest CT.

PURPOSE: To present a method that automatically segments and quantifies abnormal CT patterns commonly present in coronavirus disease 2019 (COVID-19), namely ground glass opacities and consolidations. MATERIALS AND METHODS: In this retrospective study, the proposed method takes as input a non-contrasted chest CT and segments the lesions, lungs, and lobes in three dimensions, based on a dataset of 9749 chest CT volumes. The method outputs two combined measures of the severity of lung and lobe involvement, quantifying both the extent of COVID-19 abnormalities and presence of high opacities, based on deep learning and deep reinforcement learning. The first measure of (PO, PHO) is global, while the second of (LSS, LHOS) is lobe-wise. Evaluation of the algorithm is reported on CTs of 200 participants (100 COVID-19 confirmed patients and 100 healthy controls) from institutions from Canada, Europe and the United States collected between 2002-Present (April 2020). Ground truth is established by manual annotations of lesions, lungs, and lobes. Correlation and regression analyses were performed to compare the prediction to the ground truth. RESULTS: Pearson correlation coefficient between method prediction and ground truth for COVID-19 cases was calculated as 0.92 for PO (P < .001), 0.97 for PHO (P < .001), 0.91 for LSS (P < .001), 0.90 for LHOS (P < .001). 98 of 100 healthy controls had a predicted PO of less than 1%, 2 had between 1-2%. Automated processing time to compute the severity scores was 10 seconds per case compared to 30 minutes required for manual annotations. CONCLUSION: A new method segments regions of CT abnormalities associated with COVID-19 and computes (PO, PHO), as well as (LSS, LHOS) severity scores.

Can cerebellar and brainstem apparent diffusion coefficient (ADC) values predict neuromotor outcome in term neonates with hypoxic-ischemic encephalopathy (HIE) treated with hypothermia?

BACKGROUND AND PURPOSE: To determine the apparent diffusion coefficient (ADC) in specific infratentorial brain structures during the first week of life and its relation with neuromotor outcome for Hypoxic-ischemic encephalopathy (HIE) in term neonates with and without whole-body hypothermia (TH). MATERIALS AND METHODS: We retrospectively evaluated 45 MRI studies performed in the first week of life of term neonates born between 2010 and 2013 at Boston Children's Hospital. Selected cases were classified into three groups: 1) HIE neonates who underwent TH, 2) HIE normothermics (TN), and 3) controls. The neuromotor outcome was categorized as normal, abnormal and death. The ADCmean was calculated for six infratentorial brain regions. RESULTS: A total of 45 infants were included: 28 HIE TH treated, 8 HIE TN, and 9 controls. The mean gestational age was 39 weeks; 57.8% were male; 11.1% were non-survivors. The median age at MRI was 3 days (interquartile range, 1-4 days). A statistically significant relationship was shown between motor outcome or death and the ADCmean in the vermis (P = 0.002), cerebellar left hemisphere (P = 0.002), midbrain (P = 0.009), pons (P = 0.014) and medulla (P = 0.005). In patients treated with TH, the ADC mean remained significantly lower than that in the controls only in the hemispheres (P = 0.01). In comparison with abnormal motor outcome, ADCmean was lowest in the left hemisphere (P = 0.003), vermis (P = 0.003), pons (P = 0.0036) and medulla (P = 0.008) in case of death. CONCLUSION: ADCmean values during the first week of life in the left hemisphere, vermis, pons and medulla are related to motor outcome or death in infants with HIE either with or without hypothermic therapy. Therefore, this objective tool can be assessed prospectively to determine if it can be used to establish prognosis in the first week of life, particularly in severe cases of HIE.

High-angular resolution diffusion imaging tractography of cerebellar pathways from newborns to young adults.

INTRODUCTION: Many neurologic and psychiatric disorders are thought to be due to, or result in, developmental errors in neuronal cerebellar connectivity. In this connectivity analysis, we studied the developmental time-course of cerebellar peduncle pathways in pediatric and young adult subjects. METHODS: A cohort of 80 subjects, newborns to young adults, was studied on a 3T MR system with 30 diffusion-weighted measurements with high-angular resolution diffusion imaging (HARDI) tractography. RESULTS: Qualitative and quantitative results were analyzed for age-based variation. In subjects of all ages, the superior cerebellar peduncle pathway (SCP) and two distinct subpathways of the middle cerebellar peduncle (MCP), as described in previous ex vivo studies, were identified in vivo with this technique: pathways between the rostral pons and inferior-lateral cerebellum (MCP cog), associated predominantly with higher cognitive function, and pathways between the caudal pons and superior-medial cerebellum (MCP mot), associated predominantly with motor function. DISCUSSION: Our findings showed that the inferior cerebellar peduncle pathway (ICP), involved primarily in proprioception and balance appears to have a later onset followed by more rapid development than that exhibited in other tracts. We hope that this study may provide an initial point of reference for future studies of normal and pathologic development of cerebellar connectivity.

Single-Breath Washout Tests to Assess Small Airway Disease in COPD.

BACKGROUND: Current functional assessments do not allow a reliable assessment of small airways, which are a major site of disease in COPD. Single-breath washout (SBW) tests are feasible and reproducible methods for evaluating small airway disease. Their relevance in COPD remains unknown. METHODS: We performed a cross-sectional study in 65 patients with moderate to severe COPD. Phase III slope of nitrogen (SIIIN2) and double tracer gas (SIIIDTG) SBW tests were used as a measure of ventilation inhomogeneity. The association of both markers with established physiological and clinical features of COPD was assessed. RESULTS: Ventilation inhomogeneity as measured by SIIIN2 and SIIIDTG was increased in patients with COPD compared with healthy subjects (P < .001 and P < .001, respectively). SIIIN2 was associated with FEV1 predicted, residual volume (RV)/total lung capacity (TLC) and diffusing capacity of the lung for carbon monoxide (Dlco) (all P < .001). Furthermore, SIIIN2 was related to dyspnea, exercise-induced desaturation, and exercise capacity (P = .001, P < .001, and P = .047, respectively). SIIIDTG was associated with TLC, Dlco, and cough (P < .001, P = .001, and P = .009, respectively). In multivariate regression models, we demonstrated that these associations are largely independent of FEV1 and mostly stronger than associations with FEV1. In contrast, FEV1 was superior in predicting emphysema severity. CONCLUSIONS: SIIIN2 and SIIIDTG, two fast and clinically applicable measures of small airway disease, reflect different physiological and clinical aspects of COPD, largely independent of spirometry. TRIAL REGISTRY: ISRCTN99586989, Ethics committee Beider Basel (approval number 295/07).

Elevated MTSS1 expression associated with metastasis and poor prognosis of residual hepatitis B-related hepatocellular carcinoma.

BACKGROUND: Hepatectomy generally offers the best chance of long-term survival for patients with hepatocellular carcinoma (HCC). Many studies have shown that hepatectomy accelerates tumor metastasis, but the mechanism remains unclear. METHODS: An orthotopic nude mice model with palliative HCC hepatectomy was performed in this study. Metastasis-related genes in tumor following resection were screened; HCC invasion, metastasis, and some molecular alterations were examined in vivo and in vitro. Clinical significance of key gene mRNA expression was also analyzed. RESULTS: Metastasis suppressor 1 (MTSS1) located in the central position of gene function net of residual HCC. MTSS1 was up-regulated in residual tumor after palliative resection. In hepatitis B-related HCC patients undergone palliative hepatectomy, those with higher MTSS1 mRNA expression accompanied by activation of matrix metalloproteinase 2 (MMP2) in residual HCC, had earlier residual HCC detection after hepatectomy and poorer survival when compared to those with lower MTSS1. In different cell lines, the levels of MTSS1 mRNA increased in parallel with metastatic potential. MTSS1 down regulation via siRNA decreased MMP2 activity, reduced invasive potentials of HCC by 28.9 % in vitro, and averted the deteriorated lung metastatic extent in vivo. CONCLUSIONS: The poor prognosis of hepatitis B-related HCC patients following palliative hepatectomy associates with elevated MTSS1 mRNA expression; therefore, MTSS1 may provide a new research field for HCC diagnosis and treatment.

Use of read-across and computer-based predictive analysis for the safety assessment of PEG cocamines.

In the European Union animal testing has been eliminated for cosmetic ingredients while the US Cosmetic Ingredient Review Expert Panel may request data from animal studies. The use of read-across and predictive toxicology provides a path for filling data gaps without additional animal testing. The PEG cocamines are tertiary amines with an alkyl group derived from coconut fatty acids and two PEG chains of varying length. Toxicology data gaps for the PEG cocamines can be addressed by read-across based on structure-activity relationship using the framework described by Wu et al. (2010) for identifying suitable structural analogs. Data for structural analogs supports the conclusion that the PEG cocamines are non-genotoxic and not expected to exhibit systemic or developmental/reproductive toxicity with use in cosmetics. Due to lack of reliable dermal sensitization data for suitable analogs, this endpoint was addressed using predictive software (TIMES SS) as a first step (Laboratory of Mathematical Chemistry). The prediction for PEG cocamines was the same as that for PEGs, which have been concluded to not present a significant concern for dermal sensitization. This evaluation for PEG cocamines demonstrates the utility of read-across and predictive toxicology tools to assess the safety of cosmetic ingredients.

Magnetic Resonance Fiber Tracking in a Neonate with Hemimegalencephaly.

A magnetic resonance diffusion fiber tracking study in neonate diagnosed with left hemisphere hemimegalencephaly is presented. Despite diffuse morphologic deformities identified in conventional imaging, all major pathways were identifiable bilaterally with minor aberrations in vicinity of morphologic lesions.

Prenatal magnetic resonance imaging detection of temporal lobes and hippocampal anomalies in hypochondroplasia.

Hypochondroplasia (HCH) is a genetic skeletal dysplasia, inherited in an autosomal dominant fashion. About 50-70% of HCH patients have a mutation in FGFR3 gene and in the majority of cases it is a de novo mutation. Recent magnetic resonance imaging studies on relative large cohorts of HCH patients have showed a central nervous system involvement with a high incidence of characteristic temporal lobe and hippocampal abnormalities. To the best of our knowledge, this report shows the first magnetic resonance imaging prenatal detection of characteristic brain anomalies in a case of HCH, molecularly confirmed through postnatal FGFR3 analysis.

Autoimmune pancreatitis: multidetector-row computed tomography (MDCT) and magnetic resonance (MR) findings in the Italian experience.

Multidetector-row computed tomography (MDCT) and magnetic resonance (MR) imaging are currently the most frequently performed imaging modalities for the study of pancreatic disease. In cases of suspected autoimmune pancreatitis (AIP), a dynamic quadriphasic (precontrast, contrast-enhanced pancreatic, venous and late phases) study is recommended in both techniques. In the diffuse form of autoimmune pancreatitis (DAIP), the pancreatic parenchyma shows diffuse enlargement and appears, during the MDCT and MR contrast-enhanced pancreatic phase, diffusely hypodense and hypointense, respectively, compared to the spleen because of lymphoplasmacytic infiltration and pancreatic fibrosis. During the venous phase of MDCT and MR imaging, the parenchyma appears hyperdense and hyperintense, respectively, in comparison to the pancreatic phase. In the delayed phase of both imaging modalities, it shows retention of contrast media. A "capsule-like rim" may be recognised as a peripancreatic MDCT hyperdense and MR hypointense halo in the T2-weighted images, compared to the parenchyma. DAIP must be differentiated from non-necrotizing acute pancreatitis (NNAP) and lymphoma since both diseases show diffuse enlargement of the pancreatic parenchyma. The differential diagnosis is clinically difficult, and dynamic contrast-enhanced MDCT has an important role. In the focal form of autoimmune pancreatitis (FAIP), the parenchyma shows segmental enlargement involving the head, the body-tail or the tail, with the same contrast pattern as the diffuse form on both modalities. FAIP needs to be differentiated from pancreatic adenocarcinoma to avoid unnecessary surgical procedures, since both diseases have similar clinical and imaging presentation. The differential diagnosis is clinically difficult, and dynamic contrast-enhanced MDCT and MR imaging both have an important role. MR cholangiopancreatography helps in the differential diagnosis. Furthermore, MDCT and MR imaging can identify the extrapancreatic manifestations of AIP, most commonly biliary, renal and retroperitoneal. Finally, in all cases of uncertain diagnosis, MDCT and/or MR follow-up after short-term treatment (2-3 weeks) with high-dose steroids can identify a significant reduction in size of the pancreatic parenchyma and, in FAIP, normalisation of the calibre of the upstream main pancreatic duct.

Magnetic resonance neurography of peroneal palsy in crossed-legs habit: case report.

"Drop foot" palsy attributed to the prolonged and repetitive maintenance of the crossed-leg posture has been occasionally reported. We report, to the best of our knowledge, the first case of magnetic resonance imaging evidence of peroneal nerve abnormalities related to right drop-foot palsy in a tall healthy subject with habit of prolonged daily leg crossing.

Autoimmune pancreatitis and non-necrotizing acute pancreatitis: computed tomography pattern.

OBJECTIVES: To retrospectively differentiate diffuse autoimmune pancreatitis from non-necrotizing acute pancreatitis at clinical onset with multi detector row computed tomography. METHODS: 36 Patients suffering from diffuse autoimmune pancreatitis (14) or non-necrotizing acute pancreatitis (22) were enrolled. Qualitative analysis included stranding, retroperitoneal fluid film, capsule-like rim enhancement and pleural effusion. In quantitative analysis pancreatic density was measured in all phases. The vascularization behaviour was assessed using the relative enhancement rate across all phases. RESULTS: Pancreatic density resulted lower in non-necrotizing acute pancreatitis compared to diffuse autoimmune pancreatitis patients in pre-contrast phase and higher in pancreatic phase. Relative enhancement rate evaluation confirmed different vascularization behaviours of the two diseases. Only non-necrotizing acute pancreatitis Patients presented peripancreatic stranding and fluid in the retromesenteric interfascial plane. CONCLUSIONS: Multi detector row computed tomography is a useful technique for differentiating diffuse autoimmune pancreatitis from non-necrotizing acute pancreatitis at clinical onset. Peripancreatic stranding and retroperitoneal fluid film, characteristic of non-necrotizing acute pancreatitis, and late-phase peripheral rim enhancement, characteristic of diffuse autoimmune pancreatitis, provide qualitative clues to the differentiation. A quantitative study of contrast enhancement patterns, considering the relative enhancement rate, can assist in the differential diagnoses of two diseases.

Early formative stage of human focal cortical gyration anomalies: fetal MRI.

OBJECTIVE: Limited information is available about the development of focal cortical gyration anomalies in the human brain. Using prenatal MRI, we characterized focal cortical gyration anomalies at an early formative stage and sought clues about the mechanisms of their development. MATERIALS AND METHODS: From a large prenatal MRI database, 30 cases (gestational age, ≤ 24 weeks) with reported focal distortion of the cortical rim profile were selected. Eight cases were matched with histologic examinations; another seven had prenatal MRI, MRI autopsy, or postnatal MRI follow-up; and 15 had no follow-up but did present analogous abnormal cortical features. Focal cortical gyration anomalies were detectable when the brain was still smooth (i.e., physiological lissencephaly). RESULTS: Four patterns of cortical plate anomaly were identified: wartlike (11 cases), abnormal invaginating sulcus (11 cases), sawtooth (six cases), and single or multiple bumps (two cases). A thinned or blurred subplate and intermediate zone in the focal cortical gyration anomaly site was detected in 80% of cases. All but two cases had other intracranial anomalies. Seven cases were classified as hypoxic-ischemic, five as genetic, and three as infective. In 15 cases, the cause could not be established. In five fetuses with further intrauterine or postnatal MRI, focal cortical gyration anomalies increased in complexity, fulfilling postnatal imaging criteria of polymicrogyria. CONCLUSION: Focal cortical gyration anomalies can be detected at the early sulcation process stage. The process leading to abnormal gyration may evolve faster than physiologic ones and seems to be related to alterations of parenchymal layering occurring before 24 weeks' gestation. Most focal cortical gyration anomalies evolve toward what is currently considered polymicrogyria.