Exploring the Acceleration Limits of Deep Learning Variational Network-based Two-dimensional Brain MRI.

Purpose: To explore the limits of deep learning-based brain MRI reconstruction and identify useful acceleration ranges for general-purpose imaging and potential screening. Materials and Methods: In this retrospective study conducted from 2019 through 2021, a model was trained for reconstruction on 5847 brain MR images. Performance was evaluated across a wide range of accelerations (up to 100-fold along a single phase-encoded direction for two-dimensional [2D] sections) on the fastMRI test set collected at New York University, consisting of 558 image volumes. In a sample of 69 volumes, reconstructions were classified by radiologists for identification of two clinical thresholds: (a) general-purpose diagnostic imaging and (b) potential use in a screening protocol. A Monte Carlo procedure was developed to estimate reconstruction error with only undersampled data. The model was evaluated on both in-domain and out-of-domain data. The 95% CIs were calculated using the percentile bootstrap method. Results: Radiologists rated 100% of 69 volumes as having sufficient image quality for general-purpose imaging at up to 4× acceleration and 65 of 69 volumes (94%) as having sufficient image quality for screening at up to 14× acceleration. The Monte Carlo procedure estimated ground truth peak signal-to-noise ratio and mean squared error with coefficients of determination greater than 0.5 at 2× to 20× acceleration levels. Out-of-distribution experiments demonstrated the model's ability to produce images substantially distinct from the training set, even at 100× acceleration. Conclusion: For 2D brain images using deep learning-based reconstruction, maximum acceleration for potential screening was three to four times higher than that for diagnostic general-purpose imaging.Keywords: MRI Reconstruction, High Acceleration, Deep Learning, Screening, Out of Distribution Supplemental material is available for this article. © RSNA, 2022.

A Worm's Tale or Why to Avoid the Raccoon Latrine: A Case of Baylisascaris procyonis Meningoencephalitis.

The raccoon roundworm Baylisascaris procyonis (B. procyonis) may infect humans to cause severe or fatal meningoencephalitis, as well as ocular and visceral larva migrans. Young children are at greater risk for cerebral larva migrans with severe meningoencephalitis, and early empiric therapy may improve outcomes. Familiarity with characteristic brain imaging findings may prompt earlier diagnosis, particularly in the setting of CSF eosinophilia. We report a case of a 19-month-old boy who presented with truncal ataxia and was found to have peripheral and CSF eosinophilia. MRI demonstrated symmetric, confluent T2 hyperintense signal in the cerebral and cerebellar deep white mater, which helped differentiate B. procyonis meningoencephalitis from other infectious and non-infectious causes of eosinophilic meningoencephalitis. Early recognition and treatment of B. procyonis meningoencephalitis are important for improved outcomes, and careful review of neuroimaging can play a critical role in suggesting the diagnosis.

Clinical, Pathological, and Molecular Characteristics of Diffuse Spinal Cord Gliomas.

Diffuse spinal cord gliomas (SCGs) are rare tumors associated with a high morbidity and mortality that affect both pediatric and adult populations. In this retrospective study, we sought to characterize the clinical, pathological, and molecular features of diffuse SCG in 22 patients with histological and molecular analyses. The median age of our cohort was 23.64 years (range 1-82) and the overall median survival was 397 days. K27M mutation was significantly more prevalent in males compared to females. Gross total resection and chemotherapy were associated with improved survival, compared to biopsy and no chemotherapy. While there was no association between tumor grade, K27M status (p = 0.366) or radiation (p = 0.772), and survival, males showed a trend toward shorter survival. K27M mutant tumors showed increased chromosomal instability and a distinct DNA methylation signature.

MRI Radiogenomics of Pediatric Medulloblastoma: A Multicenter Study.

Background Radiogenomics of pediatric medulloblastoma (MB) offers an opportunity for MB risk stratification, which may aid therapeutic decision making, family counseling, and selection of patient groups suitable for targeted genetic analysis. Purpose To develop machine learning strategies that identify the four clinically significant MB molecular subgroups. Materials and Methods In this retrospective study, consecutive pediatric patients with newly diagnosed MB at MRI at 12 international pediatric sites between July 1997 and May 2020 were identified. There were 1800 features extracted from T2- and contrast-enhanced T1-weighted preoperative MRI scans. A two-stage sequential classifier was designed-one that first identifies non-wingless (WNT) and non-sonic hedgehog (SHH) MB and then differentiates therapeutically relevant WNT from SHH. Further, a classifier that distinguishes high-risk group 3 from group 4 MB was developed. An independent, binary subgroup analysis was conducted to uncover radiomics features unique to infantile versus childhood SHH subgroups. The best-performing models from six candidate classifiers were selected, and performance was measured on holdout test sets. CIs were obtained by bootstrapping the test sets for 2000 random samples. Model accuracy score was compared with the no-information rate using the Wald test. Results The study cohort comprised 263 patients (mean age ± SD at diagnosis, 87 months ± 60; 166 boys). A two-stage classifier outperformed a single-stage multiclass classifier. The combined, sequential classifier achieved a microaveraged F1 score of 88% and a binary F1 score of 95% specifically for WNT. A group 3 versus group 4 classifier achieved an area under the receiver operating characteristic curve of 98%. Of the Image Biomarker Standardization Initiative features, texture and first-order intensity features were most contributory across the molecular subgroups. Conclusion An MRI-based machine learning decision path allowed identification of the four clinically relevant molecular pediatric medulloblastoma subgroups. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Chaudhary and Bapuraj in this issue.

Fetal corpus callosum abnormalities: Ultrasound and magnetic resonance imaging role.

The corpus callosum (CC) is the major interhemispheric commissure and its abnormalities include agenesis, hypoplasia, and hyperplasia. The CC anomalies are typically related to other central nervous system (CNS) or extra-CNS malformations. The antenatal diagnosis of complete CC agenesis is easy after mid-trimester by ultrasound (US) even in the axial plane. The non-visualization of cavum septum pellucidum and colpocephaly are critical signs in the axial view. More subtle findings (i.e., hypoplasia and partial agenesis) might also be recognized antenatally. In this review, the focus was given on the prenatal diagnosis of CC abnormalities in US and magnetic resonance imaging.

Case 299: Sellar Spine.

History Part one of this case appeared 4 months previously and may contain larger images. A 32-year-old woman presented to an ophthalmologist for bilateral blurry vision. She underwent MRI of the brain and orbits, which showed a focal abnormality within the pituitary gland. The patient was referred to an endocrinologist for further evaluation. Review of systems and physical examination by the endocrinologist revealed no symptoms or signs of endocrine dysfunction. Anterior pituitary hormone levels, including growth hormone, prolactin, thyroid stimulating hormone, follicular-stimulating hormone, luteinizing hormone, and adrenocorticotropic hormone, were normal. Dynamic contrast-enhanced MRI of the sella and pituitary gland and subsequent CT of the anterior skull base were performed.

Radiomic signatures of posterior fossa ependymoma: Molecular subgroups and risk profiles.

BACKGROUND: The risk profile for posterior fossa ependymoma (EP) depends on surgical and molecular status [Group A (PFA) versus Group B (PFB)]. While subtotal tumor resection is known to confer worse prognosis, MRI-based EP risk-profiling is unexplored. We aimed to apply machine learning strategies to link MRI-based biomarkers of high-risk EP and also to distinguish PFA from PFB. METHODS: We extracted 1800 quantitative features from presurgical T2-weighted (T2-MRI) and gadolinium-enhanced T1-weighted (T1-MRI) imaging of 157 EP patients. We implemented nested cross-validation to identify features for risk score calculations and apply a Cox model for survival analysis. We conducted additional feature selection for PFA versus PFB and examined performance across three candidate classifiers. RESULTS: For all EP patients with GTR, we identified four T2-MRI-based features and stratified patients into high- and low-risk groups, with 5-year overall survival rates of 62% and 100%, respectively (P < .0001). Among presumed PFA patients with GTR, four T1-MRI and five T2-MRI features predicted divergence of high- and low-risk groups, with 5-year overall survival rates of 62.7% and 96.7%, respectively (P = .002). T1-MRI-based features showed the best performance distinguishing PFA from PFB with an AUC of 0.86. CONCLUSIONS: We present machine learning strategies to identify MRI phenotypes that distinguish PFA from PFB, as well as high- and low-risk PFA. We also describe quantitative image predictors of aggressive EP tumors that might assist risk-profiling after surgery. Future studies could examine translating radiomics as an adjunct to EP risk assessment when considering therapy strategies or trial candidacy.

Neuroimaging Features of Intracranial Hypertension in Pediatric Patients With New-Onset Idiopathic Seizures, a Comparison With Patients with Confirmed Diagnosis of Idiopathic Intracranial Hypertension: A Preliminary Study.

CONCLUSION: A cutoff value of 6.0 mm for optic nerve sheath dilation may be used as a screening imaging marker to suspect elevated opening pressure with specificity of 88% in pediatric patients with new-onset idiopathic seizures.

Case 299.

History A 32-year-old woman presented to an ophthalmologist for bilateral blurry vision. She underwent MRI of the brain and orbits, which showed a focal abnormality within the pituitary gland. The patient was referred to an endocrinologist for further evaluation. Review of systems and physical examination by the endocrinologist revealed no symptoms or signs of endocrine dysfunction. Anterior pituitary hormone levels, including growth hormone, prolactin, thyroid stimulating hormone, follicular stimulating hormone, luteinizing hormone, and adrenocorticotropic hormone, were normal. Dynamic contrast-enhanced MRI of the sella and pituitary gland (Figs 1-3) and subsequent CT of the anterior skull base (Figs 4, 5) were performed.

Machine Assist for Pediatric Posterior Fossa Tumor Diagnosis: A Multinational Study.

BACKGROUND: Clinicians and machine classifiers reliably diagnose pilocytic astrocytoma (PA) on magnetic resonance imaging (MRI) but less accurately distinguish medulloblastoma (MB) from ependymoma (EP). One strategy is to first rule out the most identifiable diagnosis. OBJECTIVE: To hypothesize a sequential machine-learning classifier could improve diagnostic performance by mimicking a clinician's strategy of excluding PA before distinguishing MB from EP. METHODS: We extracted 1800 total Image Biomarker Standardization Initiative (IBSI)-based features from T2- and gadolinium-enhanced T1-weighted images in a multinational cohort of 274 MB, 156 PA, and 97 EP. We designed a 2-step sequential classifier - first ruling out PA, and next distinguishing MB from EP. For each step, we selected the best performing model from 6-candidate classifier using a reduced feature set, and measured performance on a holdout test set with the microaveraged F1 score. RESULTS: Optimal diagnostic performance was achieved using 2 decision steps, each with its own distinct imaging features and classifier method. A 3-way logistic regression classifier first distinguished PA from non-PA, with T2 uniformity and T1 contrast as the most relevant IBSI features (F1 score 0.8809). A 2-way neural net classifier next distinguished MB from EP, with T2 sphericity and T1 flatness as most relevant (F1 score 0.9189). The combined, sequential classifier was with F1 score 0.9179. CONCLUSION: An MRI-based sequential machine-learning classifiers offer high-performance prediction of pediatric posterior fossa tumors across a large, multinational cohort. Optimization of this model with demographic, clinical, imaging, and molecular predictors could provide significant advantages for family counseling and surgical planning.

MRI-based radiomics for prognosis of pediatric diffuse intrinsic pontine glioma: an international study.

BACKGROUND: Diffuse intrinsic pontine gliomas (DIPGs) are lethal pediatric brain tumors. Presently, MRI is the mainstay of disease diagnosis and surveillance. We identify clinically significant computational features from MRI and create a prognostic machine learning model. METHODS: We isolated tumor volumes of T1-post-contrast (T1) and T2-weighted (T2) MRIs from 177 treatment-naïve DIPG patients from an international cohort for model training and testing. The Quantitative Image Feature Pipeline and PyRadiomics was used for feature extraction. Ten-fold cross-validation of least absolute shrinkage and selection operator Cox regression selected optimal features to predict overall survival in the training dataset and tested in the independent testing dataset. We analyzed model performance using clinical variables (age at diagnosis and sex) only, radiomics only, and radiomics plus clinical variables. RESULTS: All selected features were intensity and texture-based on the wavelet-filtered images (3 T1 gray-level co-occurrence matrix (GLCM) texture features, T2 GLCM texture feature, and T2 first-order mean). This multivariable Cox model demonstrated a concordance of 0.68 (95% CI: 0.61-0.74) in the training dataset, significantly outperforming the clinical-only model (C = 0.57 [95% CI: 0.49-0.64]). Adding clinical features to radiomics slightly improved performance (C = 0.70 [95% CI: 0.64-0.77]). The combined radiomics and clinical model was validated in the independent testing dataset (C = 0.59 [95% CI: 0.51-0.67], Noether's test P = .02). CONCLUSIONS: In this international study, we demonstrate the use of radiomic signatures to create a machine learning model for DIPG prognostication. Standardized, quantitative approaches that objectively measure DIPG changes, including computational MRI evaluation, could offer new approaches to assessing tumor phenotype and serve a future role for optimizing clinical trial eligibility and tumor surveillance.

Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction.

Accelerating MRI scans is one of the principal outstanding problems in the MRI research community. Towards this goal, we hosted the second fastMRI competition targeted towards reconstructing MR images with subsampled k-space data. We provided participants with data from 7,299 clinical brain scans (de-identified via a HIPAA-compliant procedure by NYU Langone Health), holding back the fully-sampled data from 894 of these scans for challenge evaluation purposes. In contrast to the 2019 challenge, we focused our radiologist evaluations on pathological assessment in brain images. We also debuted a new Transfer track that required participants to submit models evaluated on MRI scanners from outside the training set. We received 19 submissions from eight different groups. Results showed one team scoring best in both SSIM scores and qualitative radiologist evaluations. We also performed analysis on alternative metrics to mitigate the effects of background noise and collected feedback from the participants to inform future challenges. Lastly, we identify common failure modes across the submissions, highlighting areas of need for future research in the MRI reconstruction community.

Genotype-phenotype correlation at codon 1740 of SETD2.

The SET domain containing 2, histone lysine methyltransferase encoded by SETD2 is a dual-function methyltransferase for histones and microtubules and plays an important role for transcriptional regulation, genomic stability, and cytoskeletal functions. Specifically, SETD2 is associated with trimethylation of histone H3 at lysine 36 (H3K36me3) and methylation of α-tubulin at lysine 40. Heterozygous loss of function and missense variants have previously been described with Luscan-Lumish syndrome (LLS), which is characterized by overgrowth, neurodevelopmental features, and absence of overt congenital anomalies. We have identified 15 individuals with de novo variants in codon 1740 of SETD2 whose features differ from those with LLS. Group 1 consists of 12 individuals with heterozygous variant c.5218C>T p.(Arg1740Trp) and Group 2 consists of 3 individuals with heterozygous variant c.5219G>A p.(Arg1740Gln). The phenotype of Group 1 includes microcephaly, profound intellectual disability, congenital anomalies affecting several organ systems, and similar facial features. Individuals in Group 2 had moderate to severe intellectual disability, low normal head circumference, and absence of additional major congenital anomalies. While LLS is likely due to loss of function of SETD2, the clinical features seen in individuals with variants affecting codon 1740 are more severe suggesting an alternative mechanism, such as gain of function, effects on epigenetic regulation, or posttranslational modification of the cytoskeleton. Our report is a prime example of different mutations in the same gene causing diverging phenotypes and the features observed in Group 1 suggest a new clinically recognizable syndrome uniquely associated with the heterozygous variant c.5218C>T p.(Arg1740Trp) in SETD2.

Exploring DNA Methylation for Prognosis and Analyzing the Tumor Microenvironment in Pleomorphic Xanthoastrocytoma.

Pleomorphic xanthoastrocytoma (PXA) is a rare type of brain tumor that affects children and young adults. Molecular prognostic markers of PXAs remain poorly established. Similar to gangliogliomas, PXAs show prominent immune cell infiltrate, but its composition also remains unknown. In this study, we correlated DNA methylation and BRAF status with clinical outcome and explored the tumor microenvironment. We performed DNA methylation in 21 tumor samples from 18 subjects with a histological diagnosis of PXA. MethylCIBERSORT was used to deconvolute the PXA microenvironment by analyzing the associated immune cell-types. Median age at diagnosis was 16 years (range 7-32). At median follow-up of 30 months, 3-year and 5-year overall survival was 73% and 71%, respectively. Overall survival ranged from 1 to 139 months. Eleven out of 18 subjects (61%) showed disease progression. Progression-free survival ranged from 1 to 89 months. Trisomy 7 and CDKN2A/B (p16) homozygous deletion did not show any association with overall survival (p = 0.67 and p = 0.74, respectively). Decreased overall survival was observed for subjects with tumors lacking the BRAF V600E mutation (p = 0.02). PXAs had significantly increased CD8 T-cell epigenetic signatures compared with previously profiled gangliogliomas (p = 0.0019). The characterization of immune cell-types in PXAs may have implications for future development of immunotherapy.

COVID-19-associated Diffuse Leukoencephalopathy and Microhemorrhages.

Diffuse leukoencephalopathy and juxtacortical and/or callosal microhemorrhages were brain imaging features in critically ill patients with coronavirus disease 2019. Coronavirus disease 2019 (COVID-19) has been reported in association with a variety of brain imaging findings such as ischemic infarct, hemorrhage, and acute hemorrhagic necrotizing encephalopathy. Herein, the authors report brain imaging features in 11 critically ill patients with COVID-19 with persistently diminished mental status who underwent MRI between April 5 and April 25, 2020. These imaging features include (a) confluent T2 hyperintensity and mild restricted diffusion in bilateral supratentorial deep and subcortical white matter (in 10 of 11 patients) and (b) multiple punctate microhemorrhages in juxtacortical and callosal white matter (in seven of 11 patients). The authors also discuss potential pathogeneses.

MR Imaging of the Penis and Scrotum.

Traditionally, due to its low cost, ready availability, and proved diagnostic accuracy, ultrasonography (US) has been the primary imaging modality for the evaluation of scrotal and, to a lesser extent, penile disease. However, US is limited by its relatively small useful field of view, operator dependence, and inability to provide much information on tissue characterization. Magnetic resonance (MR) imaging, with its excellent soft-tissue contrast and good spatial resolution, is increasingly being used as both a problem-solving tool in patients who have already undergone US and as a primary modality for the evaluation of suspected disease. Specifically, MR imaging can aid in differentiating between benign and malignant lesions seen at US, help define the extent of inflammatory processes or traumatic injuries, and play a vital role in locoregional staging of tumors. Consequently, it is becoming more important for radiologists to be familiar with the wide range of penile and scrotal disease entities and their MR imaging appearances. The authors review the basic anatomy of the penis and scrotum as seen at MR imaging and provide a basic protocol for penile and scrotal imaging, with emphasis on the advantages of MR imaging. Pathologic processes are organized into traumatic (including penile fracture and contusion), infectious or inflammatory (including Fournier gangrene and scrotal abscess), and neoplastic (including both benign and malignant scrotal and penile tumors) processes.