Impact of SUSAN Denoising and ComBat Harmonization on Machine Learning Model Performance for Malignant Brain Neoplasms

BACKGROUND AND PURPOSE

Feature variability in radiomics studies due to technical and magnet strength parameters is well known and may be addressed through various pre-processing methods. However, very few studies have evaluated downstream impact of variable pre-processing on model classification performance in a multi-class setting. We sought to evaluate the impact of SUSAN denoising and ComBat harmonization on model classification performance.

MATERIALS AND METHODS

A total of 493 cases (410 internal and 83 external dataset) of glioblastoma (GB), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL) underwent semi-automated 3D-segmentation post baseline image processing (BIP) consisting of resampling, realignment, co-registration, skull stripping and image normalization. Post BIP, two sets were generated, one with and another without SUSAN denoising (SD). Radiomics features were extracted from both datasets and batch corrected to produce four datasets: (a) BIP, (b) BIP with SD, (c) BIP with ComBat and (d) BIP with both SD and ComBat harmonization. Performance was then summarized for models using a combination of six feature selection techniques and six machine learning models across four mask-sequence combinations with features derived from one-three (multi-parametric) MRI sequences.

RESULTS

Most top performing models on the external test set used BIP+SD derived features. Overall, use of SD and ComBat harmonization led to a slight but generally consistent improvement in model performance on the external test set.

CONCLUSIONS

The use of image pre-processing steps such as SD and ComBat harmonization may be more useful in a multiinstitutional setting and improve model generalizability. Models derived from only T1-CE images showed comparable performance to models derived from multiparametric MRI.

Back to the Future: Dynamic Contrast-Enhanced Photon-Counting Detector CT for the Detection of Pituitary Adenoma in Cushing Disease

Historically, MR imaging has been unable to detect a pituitary adenoma in up to one-half of patients with Cushing disease. This issue is problematic because the standard-of-care treatment is surgical resection, and its success is correlated with finding the tumor on imaging. Photon-counting detector CT is a recent advancement that has multiple benefits over conventional energy-integrating detector CT. We present the use of dynamic contrast-enhanced imaging using photon-counting detector CT for the detection of pituitary adenomas in patients with Cushing disease.

High-Grade Astrocytoma with Piloid Features: A Dual Institutional Review of Imaging Findings of a Novel Entity

High-grade astrocytoma with piloid features (HGAP) is a recently identified brain tumor characterized by a distinct DNA methylation profile. Predominantly located in the posterior fossa of adults, HGAP is notably prevalent in individuals with neurofibromatosis type 1. We present an image-centric review of HGAP and explore the association between HGAP and neurofibromatosis type 1. Data were collected from 8 HGAP patients treated at two tertiary care institutions between January 2020 and October 2023. Demographic details, clinical records, management, and tumor molecular profiles were analyzed. Tumor characteristics, including location and imaging features on MR imaging, were reviewed. Clinical or imaging features suggestive of neurofibromatosis 1 or the presence of NF1 gene alteration were documented. The mean age at presentation was 45.5 years (male/female = 5:3). Tumors were midline, localized in the posterior fossa (n = 4), diencephalic/thalamic (n = 2), and spinal cord (n = 2). HGAP lesions were T1 hypointense, T2-hyperintense, mostly without diffusion restriction, predominantly peripheral irregular enhancement with central necrosis (n = 3) followed by mixed heterogeneous enhancement (n = 2). Two NF1 mutation carriers showed signs of neurofibromatosis type 1 before HGAP diagnosis, with one diagnosed during HGAP evaluation, strengthening the HGAP-NF1 link, particularly in patients with posterior fossa masses. All tumors were IDH1 wild-type, often with ATRX, CDKN2A/B, and NF1 gene alteration. Six patients underwent surgical resection followed by adjuvant chemoradiation. Six patients were alive, and two died during the last follow-up. Histone H3 mutations were not detected in our cohort, such as the common H3K27M typically seen in diffuse midline gliomas, linked to aggressive clinical behavior and poor prognosis. HGAP lesions may involve the brain or spine and tend to be midline or paramedian in location. Underlying neurofibromatosis type 1 diagnosis or imaging findings are important diagnostic cues.

A review of histopathologic and radiologic features of non-atherosclerotic pathologies of the extracranial carotid arteries

Diseases of the carotid arteries can be classified into different categories based on their origin. Atherosclerotic carotid disease remains the most encountered arterial wall pathology. However, other less-common non-atherosclerotic diseases can have detrimental clinical consequences if not appropriately recognized. The underlying histological features of each disease process may result in imaging findings that possess features that are obvious of the disease. However, some carotid disease processes may have histological characteristics that manifest as non-specific radiologic findings. The purpose of this manuscript is to review various non-atherosclerotic causes of carotid artery disease as well as their histologic-radiologic characteristics to aid in the appropriate recognition of these less-commonly encountered pathologies.

Preliminary Results from Retrospective Correlation of Circulating Tumor DNA (ct-DNA) with Imaging for HPV-positive Oropharyngeal Squamous Cell Carcinoma

The role of molecular markers is increasingly being recognized for head and neck tumors ranging from benign lesions like paragangliomas to malignancies like squamous cell carcinomas (SCCa). Multiple studies have recently validated blood tests for circulating tumor tissue modified viral- human papillomavirus DNA (HPV ct-DNA) (NavDx, Naveris Laboratories) for posttreatment surveillance of HPV-driven oropharyngeal SCCa. This technology quantifies fragments of circulating DNA that are shed into the blood stream with very high (>95%) positive and negative predictive values and are also highly sensitive in distinguishing tumor HPV-DNA from a non-cancerous source. This study has a cohort of 34 patients with HPV-driven oropharyngeal SCCa, having at least three sequential imaging studies and ct-DNA values. The study showed a strong positive correlation between the imaging findings and ct-DNA level in recurrent HPV positive oropharyngeal SCCa. Findings also include 100% negative predictive value of HPV ct-DNA tests to rule out tumor recurrence. At our institution, we are now routinely performing the ct-DNA assay for surveillance of treated HPV-oropharyngeal SCCa. Correlation between clinical, radiological, and biomarker findings are now part of routine discussions during the multidisciplinary tumor boards.ABBREVIATIONS: ct-DNA=circulating tumor deoxyribonucleic acid; HPV=Human Papilloma virus;OPC=Oropharyngeal SCCa=Squamous cell carcinomas; PCR= Polymerase chain reaction.

High yield clinical applications for photon counting CT in neurovascular imaging

Photon-counting CT uses a novel x-ray detection mechanism that confers many advantages over that used in traditional energy integrating CT. As photon-counting CT becomes more available, it is important to thoroughly understand its benefits and highest yield areas for improvements in diagnosis of various diseases. Based on our early experience, we have identified several areas of neurovascular imaging in which photon-counting CT shows promise. Here, we describe the benefits in diagnosing arterial and venous diseases in the head, neck, and spine. Specifically, we focus on applications in head and neck CT angiography, spinal CT angiography, and CT myelography for detection of CSF-venous fistulas. Each of these applications highlights the technological advantages of PCCT in neurovascular imaging. Further understanding of these applications will not only benefit institutions incorporating PCCT into their practices but will also help guide future directions for implementation of PCCT for diagnosing other pathologies in neuroimaging.

Molecular GBM versus Histopathological GBM: Radiology-Pathology-Genetic Correlation and the New WHO 2021 Definition of Glioblastoma

Given the recent advances in molecular pathogenesis of tumors, with better correlation with tumor behavior and prognosis, major changes were made to the new 2021 WHO (CNS5) classification of CNS tumors, including updated criteria for diagnosis of glioblastoma. Diagnosis of GBM now requires absence of isocitrate dehydrogenase and histone 3 mutations (IDH-wildtype and H3-wildtype) as the basic cornerstone, with elimination of the IDH-mutated category. The requirements for diagnosis were conventionally histopathological, based on the presence of pathognomonic features such as microvascular proliferation and necrosis. However, even if these histological features are absent, many lower grade (WHO grade 2/3) diffuse astrocytic gliomas behave clinically similar to GBM (grade 4). The 2021 WHO classification introduced new molecular criteria that can be used to upgrade the diagnosis of such histologically lower-grade, IDH-wildtype, astrocytomas to GBM. The three molecular criteria include: concurrent gain of whole chromosome 7 and loss of whole chromosome 10 (+7/-10); TERT promoter mutation; epidermal growth factor receptor (EGFR) amplification. Given these changes, it is now strongly recommended to have molecular analysis of WHO grade 2/3 diffuse astrocytic, IDH-wildtype, gliomas in adult patients, as identification of any of the above mutations allows for upgrading the tumor to WHO grade 4 ("molecular GBM") with important prognostic implications. Despite at an early stage, there is active ongoing research on the unique MRI features of molecular GBM. This paper highlights the differences between "molecular" and "histopathological" GBM, with the aim of providing a basic understanding about these changes.ABBREVIATIONS: GBM=Glioblastoma; TERT=telomerase reverse transcriptase; EGFR=epidermal growth factor receptor; MGMT= methylguanine-DNA methyltransferase; NGS= next-generation sequencing; IDH= isocitrate dehydrogenase.

7T MRI for Cushing's Disease: A Single Institutional Experience and Literature Review

BACKGROUND AND PURPOSE

Cushing disease is typically caused by a pituitary adenoma that frequently is small and challenging to detect on conventional MRI. High field strength 7T MRI can leverage increased signal-to-noise and contrast-to-noise ratios compared to lower-field strength MRI to help identify small pituitary lesions. We aim to describe our institutional experience with 7T MRI in patients with Cushing disease and perform a review of the literature.

MATERIALS AND METHODS

A retrospective analysis of 7T MRI findings in patients with pathology proven cases of Cushing disease from a single institution, followed by a review of the literature on 7T MRI for Cushing disease.

RESULTS

Our institutional experience identified Cushing adenomas in 10/13 (76.9%) patients on 7T, however only 5/13 (38.5%) lesions were discrete. Overall, the imaging protocols used were heterogeneous in terms of contrast dose as well as type of post-contrast T1-weighted sequences (Dynamic, 2D vs 3D, and type of 3D sequence). From our institutional data, specific post-gadolinium T1-weighted sequences were helpful in identifying a surgical lesion as follows: Dynamic Contrast Enhanced 2/7 (28.6%), 2D FSE 4/8 (50%), 3D SPACE 5/6 (83.3%), and 3D MPRAGE 8/11 (72.7%). The literature review identified Cushing adenomas in 31/33 (93.9%) patients on 7T.

CONCLUSIONS

7T MRI for pituitary lesion localization in Cushing disease is a new technique with imaging protocols that varied widely. Further comparative research is needed to identify the optimal imaging technique as well as to assess the benefit of 7T over lower-field strength MRI.

ABBREVIATIONS

MRI = Magnetic Resonance Imaging, CT = Computed Tomography, 7T = 7 Tesla, DCE = Dynamic Contrast Enhanced.

Newly Recognized Genetic Tumor Syndromes of the CNS in the 5th WHO Classification: Imaging Overview with Genetic Updates

The nervous system is commonly involved in a wide range of genetic tumor-predisposition syndromes. The classification of genetic tumor syndromes has evolved during the past years; however, it has now become clear that these syndromes can be categorized into a relatively small number of major mechanisms, which form the basis of the new 5th edition of the World Health Organization book (beta online version) on genetic tumor syndromes. For the first time, the World Health Organization has also included a separate chapter on genetic tumor syndromes in the latest edition of all the multisystem tumor series, including the 5th edition of CNS tumors. Our understanding of these syndromes has evolved rapidly since the previous edition (4th edition, 2016) with recognition of 8 new syndromes, including the following: Elongator protein complex-medulloblastoma syndrome, BRCA1-associated protein 1 tumor-predisposition syndrome, DICER1 syndrome, familial paraganglioma syndrome, melanoma-astrocytoma syndrome, Carney complex, Fanconi anemia, and familial retinoblastoma. This review provides a description of these new CNS tumor syndromes with a focus on imaging and genetic characteristics.

Survival and neurological outcomes following management of intramedullary spinal metastasis patients: a case series with comprehensive review of the literature

Intramedullary spinal cord metastasis (ISCM), though rare, represents a potentially debilitating manifestation of systemic cancer. With emerging advances in cancer care, ISCMs are increasingly being encountered in clinical practice. Herein, we describe one of the larger retrospective single institutional case series on ISCMs, analyze survival and treatment outcomes, and review the literature. All surgically evaluated ISCMs at our institution between 2005 and 2023 were retrospectively reviewed. Demographics, tumor features, treatment, and clinical outcome characteristics were collected. Neurological function was quantified via the Frankel grade and the McCormick score (MCS). The pre- and post-operative Karnofsky performance scores (KPS) were used to assess functional status. Descriptive statistics, univariate analysis, log-rank test, and the Kaplan-Meier survival analysis were performed. A total of 9 patients were included (median age 67 years (range, 26-71); 6 were male). Thoracic and cervical spinal segments were most affected (4 patients each). Six patients (75%) underwent surgical management (1 biopsy and 5 resections), and 3 cases underwent chemoradiation only. Post-operatively, 2 patients had an improvement in their neurological exam with one patient becoming ambulatory after surgery; three patients maintained their neurological exam, and 1 had a decline. There was no statistically significant difference in the pre- and post-operative MCS and median KPS scores in surgically treated patients. Median OS after ISCM diagnosis was 7 months. Absence of brain metastasis, tumor histology (renal and melanoma), cervical/thoracic location, and post-op KPS ≥ 70 showed a trend toward improved overall survival. The incidence of ISCM is increasing, and earlier diagnosis and treatment are considered key for the preservation of neurological function. When patient characteristics are favorable, surgical resection of ISCM can be considered in patients with rapidly progressive neurological deficits. Surgical treatment was not associated with an improvement in overall survival in patients with ISCMs.

Cauda equina neuroendocrine tumors: A single institutional imaging review of cases over two decades

Cauda Equina Neuroendocrine Tumors (CE-NET), previously referred to as paragangliomas are a rare subset of spinal tumors, with limited data on imaging. Herein, we present a retrospective review of clinical and imaging findings of CE-NETs in ten patients who were evaluated at our institution over the past two decades. All patients had well-defined intradural lesions in the lumbar spine which demonstrated slow growth. A review of imaging findings revealed the presence of an eccentric vascular pedicle along the dorsal aspect of the tumor in 8 of the 10 patients (eccentric vessel sign), a distinctive finding that has not previously been reported with this tumor and may help improve the accuracy of imaging-based diagnosis. In all cases, a gross-total resection was performed, with resolution of symptoms in most of the cases.

Surgical Management of Craniospinal Axis Solitary Fibrous Tumors: A Single-Institution Case Series and Comprehensive Review of the Literature

BACKGROUND AND OBJECTIVES

Meningeal solitary fibrous tumors (SFTs) comprise 0.4% of primary central nervous system neoplasms and carry metastatic potential. Disease course and optimal management are largely unknown, and there is currently no literature rigorously describing neurological outcomes in surgically managed SFTs. We present one of the largest craniospinal SFT series, analyze patient outcomes, and extensively review the associated literature.

METHODS

All surgically managed SFTs at our institution between January 2005 and March 2023 were retrospectively reviewed. Patient demographics, tumor and radiographic features, treatment, and clinical outcomes were collected. Neurological function was quantified using Frankel grade and Neurologic Assessment in Neuro-Oncology scores. Descriptive statistics, multivariate analysis, log-rank test, and Kaplan-Meier survival analysis were performed.

RESULTS

Twenty-one patients satisfied inclusion criteria. Tumor locations included 15 supratentorial, three infratentorial, and three spinal. All patients underwent surgical resection, and 16 (76.2%) underwent radiation. Six (28.6%) patients had tumor recurrence, and three (14.3%) developed metastasis. Younger age and higher postoperative Frankel grade were significantly associated with increased overall survival (OS) ( P = .011, P = .002, respectively). All patients symptomatically improved or stabilized after surgery, and Neurologic Assessment in Neuro-Oncology score ( P = .001) and functional status significantly improved postoperatively (Karnofsky Performance Status: 65.2 ± 25.2 vs 91.4 ± 13.5, P = .001). Sex, adjuvant radiation, and extent of resection were not significantly associated with OS.

CONCLUSION

SFT of the central nervous system is a rare entity with a variable clinical course. Surgical resection was associated with improved postoperative functional and neurological status. Higher postoperative neurological function was significantly associated with OS. Further studies are warranted to validate a standardized treatment algorithm and investigate the efficacy of adjuvant radiation in SFT.

Diagnostic Performance of Decubitus Photon-Counting Detector CT Myelography for the Detection of CSF-Venous Fistulas

BACKGROUND AND PURPOSE

CSF-venous fistulas are a common cause of spontaneous intracranial hypotension. Lateral decubitus digital subtraction myelography and CT myelography are the diagnostic imaging standards to identify these fistulas. Photon-counting CT myelography has technological advantages that might improve CSF-venous fistula detection, though no large studies have yet assessed its diagnostic performance. We sought to determine the diagnostic yield of photon-counting detector CT myelography for detection of CSF-venous fistulas in patients with spontaneous intracranial hypotension.

MATERIALS AND METHODS

We retrospectively searched our database for all decubitus photon-counting detector CT myelograms performed at our institution since the introduction of the technique in our practice. Per our institutional workflow, all patients had prior contrast-enhanced brain MR imaging and spine MR imaging showing no extradural CSF. Two neuroradiologists reviewed preprocedural brain MRIs, assessing previously described findings of intracranial hypotension (Bern score). Additionally, 2 different neuroradiologists assessed each myelogram for a definitive or equivocal CSF-venous fistula. The yield of photon-counting detector CT myelography was calculated and stratified by the Bern score using low-, intermediate-, and high-probability tiers.

RESULTS

Fifty-seven consecutive photon-counting detector CT myelograms in 57 patients were included. A single CSF-venous fistula was definitively present in 38/57 patients. After we stratified by the Bern score, a definitive fistula was seen in 56.0%, 73.3%, and 76.5% of patients with low-, intermediate-, and high-probability brain MR imaging, respectively.

CONCLUSIONS

Decubitus photon-counting detector CT myelography has an excellent diagnostic performance for the detection of CSF-venous fistulas. The yield for patients with intermediate- and high-probability Bern scores is at least as high as previously reported yields of decubitus digital subtraction myelography and CT myelography using energy-integrating detector scanners. The yield for patients with low-probability Bern scores appears to be greater compared with other modalities. Due to the retrospective nature of this study, future prospective work will be needed to compare the sensitivity of photon-counting detector CT myelography with other modalities.

Differentiation Between Glioblastoma and Metastatic Disease on Conventional MRI Imaging Using 3D-Convolutional Neural Networks: Model Development and Validation

RATIONALE AND OBJECTIVES

Imaging-based differentiation between glioblastoma (GB) and brain metastases (BM) remains challenging. Our aim was to evaluate the performance of 3D-convolutional neural networks (CNN) to address this binary classification problem.

MATERIALS AND METHODS

T1-CE, T2WI, and FLAIR 3D-segmented masks of 307 patients (157 GB and 150 BM) were generated post resampling, co-registration normalization and semi-automated 3D-segmentation and used for internal model development. Subsequent external validation was performed on 59 cases (27 GB and 32 BM) from another institution. Four different mask-sequence combinations were evaluated using area under the curve (AUC), precision, recall and F1-scores. Diagnostic performance of a neuroradiologist and a general radiologist, both without and with the model output available, was also assessed.

RESULTS

3D-model using the T1-CE tumor mask (TM) showed the highest performance [AUC 0.93 (95% CI 0.858-0.995)] on the external test set, followed closely by the model using T1-CE TM and FLAIR mask of peri-tumoral region (PTR) [AUC of 0.91 (95% CI 0.834-0.986)]. Models using T2WI masks showed robust performance on the internal dataset but lower performance on the external set. Both neuroradiologist and general radiologist showed improved performance with model output provided [AUC increased from 0.89 to 0.968 (p = 0.06) and from 0.78 to 0.965 (p = 0.007) respectively], the latter being statistically significant.

CONCLUSION

3D-CNNs showed robust performance for differentiating GB from BMs, with T1-CE TM, either alone or combined with FLAIR-PTR masks. Availability of model output significantly improved the accuracy of the general radiologist.

Clinical and Magnetic Resonance Imaging Radiomics-Based Survival Prediction in Glioblastoma Using Multiparametric Magnetic Resonance Imaging

INTRODUCTION

Survival prediction in glioblastoma remains challenging, and identification of robust imaging markers could help with this relevant clinical problem. We evaluated multiparametric magnetic resonance imaging-derived radiomics to assess prediction of overall survival (OS) and progression-free survival (PFS).

METHODOLOGY

A retrospective, institutional review board-approved study was performed. There were 93 eligible patients, of which 55 underwent gross tumor resection and chemoradiation (GTR-CR). Overall survival and PFS were assessed in the entire cohort and the GTR-CR cohort using multiple machine learning pipelines. A model based on multiple clinical variables was also developed. Survival prediction was assessed using the radiomics-only, clinical-only, and the radiomics and clinical combined models.

RESULTS

For all patients combined, the clinical feature-derived model outperformed the best radiomics model for both OS (C-index, 0.706 vs 0.597; P < 0.0001) and PFS prediction (C-index, 0.675 vs 0.588; P < 0.001). Within the GTR-CR cohort, the radiomics model showed nonstatistically improved performance over the clinical model for predicting OS (C-index, 0.638 vs 0.588; P = 0.4). However, the radiomics model outperformed the clinical feature model for predicting PFS in GTR-CR cohort (C-index, 0.641 vs 0.550; P = 0.004). Combined clinical and radiomics model did not yield superior prediction when compared with the best model in each case.

CONCLUSIONS

When considering all patients, regardless of therapy, the radiomics-derived prediction of OS and PFS is inferior to that from a model derived from clinical features alone. However, in patients with GTR-CR, radiomics-only model outperforms clinical feature-derived model for predicting PFS.

Imaging Guide to Inner Ear Malformations: An Illustrative Review

Inner ear malformation (IEM) with associated sensoryneural hearing loss (SNHL) is a major cause of childhood disability. Computed tomography (CT) and magnetic resonance imaging (MRI) imaging play important and often complementary roles in diagnosing underlying structural abnormalities and surgical planning allows for direct visualization of the cochlear nerve and is the preferred imaging modality prior to cochlear implantation. CT is helpful to assess osseous anatomy and when evaluating children with mixed hearing loss or syndromic associations. When reviewing these cases, it is important for the radiologist to be familiar with the key imaging features. In this article, we will present the imaging findings associated with different inner ear malformations associated with congenital sensorineural hearing loss.

Update from the 5th Edition of the WHO Classification of Nasal, Paranasal, and Skull Base Tumors: Imaging Overview with Histopathologic and Genetic Correlation

Sinonasal and skull base tumors are a heterogeneous group of neoplasms with considerable histologic variation and overlapping imaging features. In 2022, the World Health Organization updated the head and neck tumor classification, further emphasizing the importance of molecular data and genetic alterations in sinonasal neoplasms. The changes include the addition of new entities and discussion of emerging entities, as well as changes to the taxonomy and characterization of tumors. The new classification focuses on entities that develop in these sites either exclusively (eg, olfactory neuroblastoma) or most frequently. Another change includes reduction in the number of categories by creating separate category-specific chapters for soft-tissue, hematolymphoid, and neuroectodermal lesions. In this review, we briefly discuss the various categories in the new classification with a more detailed description of the 2 new entities (SWItch/Sucrose Non-Fermentable complex-deficient sinonasal carcinomas and human papillomavirus-related multiphenotypic sinonasal carcinoma). We also highlight the emerging entities including IDH-mutant sinonasal malignancies and DEK-AFF2 carcinoma, presently classified as sinonasal undifferentiated carcinoma and nonkeratinizing squamous cell carcinoma, respectively.

Amyloid-related Imaging Abnormalities in Alzheimer Disease Treated with Anti-Amyloid-β Therapy

Alzheimer disease (AD) is the most common form of dementia worldwide. Treatment of AD has mainly been focused on symptomatic treatment until recently with the advent and approval of monoclonal antibody (MAB) immunotherapy. U.S. Food and Drug Administration-approved drugs such as aducanumab, as well as upcoming newer-generation drugs, have provided an exciting new therapy focused on reducing the amyloid plaque burden in AD. Although this new frontier has shown benefits for patients, it is not without complications, which are mainly neurologic. Increased use of MABs led to the discovery of amyloid-related imaging abnormalities (ARIA). ARIA has been further classified into two categories, ARIA-E and ARIA-H, representing edema and/or effusion and hemorrhage, respectively. ARIA is thought to be caused by increased vascular permeability following an inflammatory response, leading to the extravasation of blood products and proteinaceous fluid. Patients with ARIA may present with headaches, but they are usually asymptomatic and ARIA is only diagnosable at MRI; it is essential for the radiologist to recognize and monitor ARIA. Increased incidence and investigation into this concern have led to the creation of grading scales and monitoring guidelines to diagnose and guide treatment using MABs. Cerebral amyloid angiopathy has an identical pathogenesis to that of ARIA and is its closest differential diagnosis, with imaging findings being the same for both entities and only a history of MAB administration allowing differentiation. The authors discuss the use of MABs for treating AD, expand on ARIA and its consequences, and describe how to identify and grade ARIA to guide treatment properly. ©RSNA, 2023 Quiz questions for this article are available through the Online Learning Center See the invited commentary by Yu in this issue.

Surgical management of symptomatic vertebral hemangiomas: a single institution experience and literature review

Vertebral hemangiomas (VHs), formed from a vascular proliferation in bone marrow spaces limited by bone trabeculae, are the most common benign tumors of the spine. While most VHs remain clinically quiescent and often only require surveillance, rarely they may cause symptoms. They may exhibit active behaviors, including rapid proliferation, extending beyond the vertebral body, and invading the paravertebral and/or epidural space with possible compression of the spinal cord and/or nerve roots ("aggressive" VHs). An extensive list of treatment modalities is currently available, but the role of techniques such as embolization, radiotherapy, and vertebroplasty as adjuvants to surgery has not yet been elucidated. There exists a need to succinctly summarize the treatments and associated outcomes to guide VH treatment plans. In this review article, a single institution's experience in the management of symptomatic VHs is summarized along with a review of the available literature on their clinical presentation and management options, followed by a proposal of a management algorithm.

Radiological abnormalities in progressive multifocal leukoencephalopathy: Identifying typical and atypical imaging patterns for early diagnosis and differential considerations

Progressive multifocal leukoencephalopathy (PML) is a rare viral central nervous system (CNS) demyelinating disease primarily associated with a compromised immune system. PML is seen mainly in individuals with human immunodeficiency virus, lymphoproliferative disease, and multiple sclerosis. Patients on immunomodulators, chemotherapy, and solid organ or bone marrow transplants are predisposed to PML. Recognition of various PML-associated typical and atypical imaging abnormalities is critical for early diagnosis and differentiating it from other conditions, especially in high-risk populations. Early PML recognition should expedite efforts at immune-system restoration, allowing for a favorable outcome. This review aims to provide a practical overview of radiological abnormalities in PML patients and address differential considerations.

Unpacking the CNS Manifestations of Epstein-Barr Virus: An Imaging Perspective

Epstein-Barr virus is a ubiquitous herpesvirus that may cause both infective (encephalitis, meningitis, and so forth) and postinfection inflammatory (such as Guillain-Barré syndrome, acute disseminated encephalomyelitis) manifestations in the CNS. Diagnosis of Epstein-Barr virus-related CNS pathologies is often complicated due to a nonspecific clinical presentation and overlap with other infectious and noninfectious causes, both clinically and on imaging. The Epstein-Barr virus is also implicated in several lymphoproliferative disorders in both immunocompromised and immunocompetent hosts. MR imaging is preferred for evaluating the extent of involvement and monitoring therapy response, given its high sensitivity and specificity, though imaging findings may be nonspecific. Herein, we review the imaging spectrum of Epstein-Barr virus-associated CNS disorders.

AI-based classification of three common malignant tumors in neuro-oncology: A multi-institutional comparison of machine learning and deep learning methods

PURPOSE

To determine if machine learning (ML) or deep learning (DL) pipelines perform better in AI-based three-class classification of glioblastoma (GBM), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL).

METHODOLOGY

Retrospective analysis included 502 cases for training (208 GBM, 67 PCNSL and 227 IMD), with external validation on 86 cases (27:27:32). Multiparametric MRI images (T1W, T2W, FLAIR, DWI and T1-CE) were co-registered, resampled, denoised and intensity normalized, followed by semiautomatic 3D segmentation of the enhancing tumor (ET) and peritumoral region (PTR). Model performance was assessed using several ML pipelines and 3D-convolutional neural networks (3D-CNN) using sequence specific masks, as well as combination of masks. All pipelines were trained and evaluated with 5-fold nested cross-validation on internal data followed by external validation using multi-class AUC.

RESULTS

Two ML models achieved similar performance on test set, one using T2-ET and T2-PTR masks (AUC: 0.885, 95% CI: [0.816, 0.935] and another using T1-CE-ET and FLAIR-PTR mask (AUC: 0.878, CI: [0.804, 0.930]). The best performing DL models achieved an AUC of 0.854, (CI [0.774, 0.914]) on external data using T1-CE-ET and T2-PTR masks, followed by model derived from T1-CE-ET, ADC-ET and FLAIR-PTR masks (AUC: 0.851, CI [0.772, 0.909]).

CONCLUSION

Both ML and DL derived pipelines achieved similar performance. T1-CE mask was used in three of the top four overall models. Additionally, all four models had some mask derived from PTR, either T2WI or FLAIR.

3D aneurysm wall enhancement is associated with symptomatic presentation

BACKGROUND

Aneurysm wall enhancement (AWE) is a potential surrogate biomarker for aneurysm instability. Previous studies have assessed AWE using 2D multiplanar methods, most of which were conducted qualitatively.

OBJECTIVE

To use a new quantitative tool to analyze a large cohort of saccular aneurysms with 3D-AWE maps METHODS: Saccular aneurysms were imaged prospectively with 3T high resolution MRI. 3D-AWE maps of symptomatic (defined as ruptured or presentation with sentinel headache/cranial nerve neuropathy) and asymptomatic aneurysms were created by extending orthogonal probes from the aneurysm lumen into the wall. Three metrics were used to characterize enhancement: 3D circumferential AWE (3D-CAWE), aneurysm-specific contrast uptake (SAWE), and focal AWE (FAWE). Aneurysms with a circumferential AWE higher than the corpus callosum (3D-CAWE ≥1) were classified as 3D-CAWE+. Symptomatic presentation was analyzed with univariate and multivariate logistic models. Aneurysm size, size ratio, aspect ratio, irregular morphology, and PHASES and ELAPSS scores were compared with the new AWE metrics. Bleb and microhemorrhage analyses were also performed.

RESULTS

Ninety-three aneurysms were analyzed. 3D-CAWE, SAWE, and FAWE were associated with symptomatic status (OR=1.34, 1.25, and 1.08, respectively). A multivariate model including aneurysm size, 3D-CAWE+, age, female gender, and FAWE detected symptomatic status with 80% specificity and 90% sensitivity (area under the curve=0.914, =0.967). FAWE was also associated with irregular morphology and high-risk location (p=0.043 and p=0.001, respectively). In general, blebs enhanced 56% more than the aneurysm body. Areas of microhemorrhage co-localized with areas of increased SAWE (p=0.047).

CONCLUSIONS

3D-AWE mapping provides a new set of metrics that could potentially improve the identification of symptomatic aneurysms.

MR Imaging of Carotid Artery Atherosclerosis: Updated Evidence on High-Risk Plaque Features and Emerging Trends

MR imaging is well-established as the criterion standard for carotid artery atherosclerosis imaging. The capability of MR imaging to differentiate numerous plaque components has been demonstrated, including those features that are associated with a high risk of sudden changes, thrombosis, or embolization. The field of carotid plaque MR imaging is constantly evolving, with continued insight into the imaging appearance and implications of various vulnerable plaque characteristics. This article will review the most up-to-date knowledge of these high-risk plaque features on MR imaging and will delve into 2 major emerging topics: the role of vulnerable plaques in cryptogenic strokes and the potential use of MR imaging to modify carotid endarterectomy treatment guidelines.

Advances in Acute Ischemic Stroke Treatment: Current Status and Future Directions

The management of acute ischemic stroke has undergone a paradigm shift in the past decade. This has been spearheaded by the emergence of endovascular thrombectomy, along with advances in medical therapy, imaging, and other facets of stroke care. Herein, we present an updated review of the various stroke trials that have impacted and continue to transform stroke management. It is critical for the radiologist to stay abreast of the ongoing developments to provide meaningful input and remain a useful part of the stroke team.

Newly Recognized CNS Tumors in the 2021 World Health Organization Classification: Imaging Overview with Histopathologic and Genetic Correlation

In 2021, the World Health Organization released an updated classification of CNS tumors. This update reflects the growing understanding of the importance of genetic alterations related to tumor pathogenesis, prognosis, and potential targeted treatments and introduces 22 newly recognized tumor types. Herein, we review these 22 newly recognized entities and emphasize their imaging appearance with correlation to histologic and genetic features.

Suspected Autosomal Recessive Polycystic Kidney Disease but Cerebellar Vermis Hypoplasia, Oligophrenia Ataxia, Coloboma, and Hepatic Fibrosis (COACH) Syndrome in Retrospect, A Delayed Diagnosis Aided by Genotyping and Reverse Phenotyping: A Case Report and A Review of the Literature

The clinical features of cerebellar vermis hypoplasia, oligophrenia, ataxia, coloboma, and hepatic fibrosis (COACH) characterize the rare autosomal recessive multisystem disorder called COACH syndrome. COACH syndrome belongs to the spectrum of Joubert syndrome and related disorders (JSRDs) and liver involvement distinguishes COACH syndrome from the rest of the JSRD spectrum. Developmental delay and oculomotor apraxia occur early but with time, these can improve and may not be readily apparent or no longer need active medical management. Congenital hepatic fibrosis and renal disease, on the other hand, may develop late, and the temporal incongruity in organ system involvement may delay the recognition of COACH syndrome. We present a case of a young adult presenting late to a Renal Genetics Clinic for evaluation of renal cystic disease with congenital hepatic fibrosis, clinically suspected to have autosomal recessive polycystic kidney disease. Following genetic testing, a reevaluation of his medical records from infancy, together with reverse phenotyping and genetic phasing, led to a diagnosis of COACH syndrome.

Withdrawal of bevacizumab is associated with rebound growth of vestibular schwannomas in neurofibromatosis type 2-related schwannomatosis patients

Background

Neurofibromatosis type 2 (NF2)-related schwannomatosis is an autosomal dominant tumor-predisposition syndrome characterized by bilateral vestibular schwannomas (VS). In patients with VS associated with NF2, vascular endothelial growth factor A inhibitor, bevacizumab, is a systemic treatment option. The aim of this study is to retrospectively evaluate NF2 patient responses to bevacizumab on VS growth and symptom progression.

Methods

This is a retrospective analysis of patients seen at the Mayo Clinic Rochester Multidisciplinary NF2 Clinic.

Results

Out of 76 patients with NF2 evaluated between 2020 and 2022, we identified 19 that received treatment with bevacizumab. Thirteen of these patients discontinued bevacizumab after median treatment duration of 12.2 months. The remaining 6 patients are currently receiving bevacizumab treatment for a median duration of 9.4 months as of March, 2023. Fifteen patients had evaluable brain MRI data, which demonstrated partial responses in 5 patients, stable disease in 8, and progression in 2. Within 6 months of bevacizumab discontinuation, 5 patients had rebound growth of their VS greater than 20% from their previous tumor volume, while 3 did not. Three patients with rebound growth went on to have surgery or irradiation for VS management.

Conclusions

Our single-institution experience confirms prior studies that bevacizumab can control progression of VS and symptoms associated with VS growth. However, we note that there is the potential for rapid VS growth following bevacizumab discontinuation, for which we propose heightened surveillance imaging and symptom monitoring for at least 6 months upon stopping anti-VEGF therapy.

Image level detection of large vessel occlusion on 4D-CTA perfusion data using deep learning in acute stroke

OBJECTIVES

Automated image-level detection of large vessel occlusions (LVO) could expedite patient triage for mechanical thrombectomy. A few studies have previously attempted LVO detection using artificial intelligence (AI) on CT angiography (CTA) images. To our knowledge this is the first study to detect LVO existence and location on raw 4D-CTA/ CT perfusion (CTP) images using neural network (NN) models.

MATERIALS AND METHODS

Retrospective study using data from a level-I stroke center was performed. A total of 306 (187 with LVO, and 119 without) patients were evaluated. Image pre-processing included co-registration, normalization and skull stripping. Five consecutive time-points for each patient were selected to provide variable contrast density in data. Additional data augmentation included rotation and horizonal image flipping. Our model architecture consisted of two neural networks, first for classification (based on hemispheric asymmetry), followed by second model for exact site of LVO detection. Only cases deemed positive by the classification model were routed to the detection model, thereby reducing false positives and improving specificity. The results were compared with expert annotated LVO detection.

RESULTS

Using a 80:20 split for training and validation, the combination of both classification and detection model achieved a sensitivity of 86.5%, a specificity of 89.5%, and an accuracy of 87.5%. A 5-fold cross-validation using the entire data achieved a mean sensitivity of 82.7%, a specificity of 89.8%, and an accuracy of 85.5% and a mean AUC of 0.89 (95% CI: 0.85-0.93).

CONCLUSION

Our findings suggest that accurate image-level LVO detection is feasible on CTP raw images.

Spinal Nerve Sheath Tumors: Factors Associated with Postoperative Residual and Recurrent Tumors: A Single-Center Experience

BACKGROUND

Spinal schwannomas (SSs) are usually benign tumors with a good prognosis when treated by surgical excision. However, complete resection can be complicated by factors such as the tumor location and configuration. In the present study, we sought to identify the factors associated with incomplete surgical resection (residual) and the factors associated with tumor recurrence.

METHODS

We performed a retrospective review of 113 cases of SSs treated surgically from 2008 to 2021.

RESULTS

Of the 113 SSs, 102 were benign and 2 were malignant nerve sheath tumors. Of the 102 benign SSs, gross total resection (GTR) was performed for 87, with 8 displaying residual and 7, recurrent tumor. We found a significantly higher ratio of cervical and sacral tumors (P = 0.008 and P = 0.004, respectively), dumbbell and foraminal configurations (P < 0.0001 and P = 0.0006, respectively), and larger tumor volumes (P = 0.003) in the residual and recurrent cohorts compared with the GTR cohort. A second operation was performed for 2 patients in the residual and 4 patients in the recurrent cohorts. The total complication rate was 6%.

CONCLUSIONS

We found that most benign SSs will be amenable to GTR (85% of cases), with an excellent prognosis. The patients with residual or recurrent tumor were more likely to have had a cervical or sacral location, a dumbbell or foraminal configuration, and a larger tumor volume. Except for 1 new SS and 1 recurrent tumor that had necessitated a lateral approach, the remainder had been treated using a posterior approach. At surgery, ultrasonography of the canal is advisable to ensure that the intra- and extraspinal components of dumbbell lesions have both been entirely removed.

3D Enhancement Color Maps in the Characterization of Intracranial Atherosclerotic Plaques

BACKGROUND AND PURPOSE

High-resolution MR imaging allows the identification of culprit symptomatic plaques after the administration of gadolinium. Current high-resolution MR imaging methods are limited by 2D multiplanar views and manual sampling of ROIs. We analyzed a new 3D method to objectively quantify gadolinium plaque enhancement.

MATERIALS AND METHODS

Patients with stroke due to intracranial atherosclerotic disease underwent 7T high-resolution MR imaging. 3D segmentations of the plaque and its parent vessel were generated. Signal intensity probes were automatically extended from the lumen into the plaque and the vessel wall to generate 3D enhancement color maps. Plaque gadolinium (Gd) uptake was quantified from 3D color maps as gadolinium uptake = (µPlaque T1 + Gd -µPlaque T1/SDPlaque T1). Additional metrics of enhancement such as enhancement ratio, variance, and plaque-versus-parent vessel enhancement were also calculated. Conventional 2D measures of enhancement were collected for comparison.

RESULTS

Thirty-six culprit and 44 nonculprit plaques from 36 patients were analyzed. Culprit plaques had higher gadolinium uptake than nonculprit plaques (P < .001). Gadolinium uptake was the most accurate metric for identifying culprit plaques (OR, 3.9; 95% CI 2.1-8.3). Gadolinium uptake was more sensitive (86% versus 70%) and specific (71% versus 68%) in identifying culprit plaques than conventional 2D measurements. A multivariate model, including gadolinium uptake and plaque burden, identified culprit plaques with an 83% sensitivity and 86% specificity.

CONCLUSIONS

The new 3D color map method of plaque-enhancement analysis is more accurate for identifying culprit plaques than conventional 2D methods. This new method generates a new set of metrics that could potentially be used to assess disease progression.

CT Perfusion Maps Improve Detection of M2-MCA Occlusions in Acute Ischemic Stroke

OBJECTIVES

Middle cerebral artery occlusions, particularly M2 branch occlusions are challenging to identify on CTA. We hypothesized that additional review of the CTP maps will increase large vessel occlusion (LVO) detection accuracy on CTA and reduce interpretation time.

MATERIALS AND METHODS

Two readers (R1 and R2) retrospectively reviewed the CT studies in 99 patients (27 normal, 26 M1-MCA, 46 M2-MCA occlusions) who presented with suspected acute ischemic stroke (AIS). The time of interpretation and final diagnosis were recorded for the CTA images (derived from CTP data), both without and with the CTP maps. The time for analysis for all vascular occlusions was compared using McNemar tests. ROC curve analysis and McNemar tests were performed to assess changes in diagnostic performance with the addition of CTP maps.

RESULTS

With the addition of the CTP maps, both readers showed increased sensitivity (p = 0.01 for R1 and p = 0.04 for R2), and accuracy (p = 0.02 for R1 and p = 0.004 for R2) for M2-MCA occlusions. There was a significant improvement in diagnostic performance for both readers for detection of M2-MCA occlusions (AUC R1 = 0.86 to 0.95, R2 = 0.84 to 0.95; p < 0.05). Both readers showed reduced interpretation time for all cases combined, as well as for normal studies (p < 0.001) when CTP images were reviewed along with CTA. Both readers also showed reduced interpretation time for M2-MCA occlusions, which was significant for one of the readers (p < 0.02).

CONCLUSION

The addition of CTP maps improves accuracy and reduces interpretation time for detecting LVO and M2-MCA occlusions in AIS. Incorporation of CTP in acute stroke imaging protocols may improve detection of more distal occlusions.

Prognostic Factors of Stroke-Like Migraine Attacks after Radiation Therapy (SMART) Syndrome

BACKGROUND AND PURPOSE

Prognostic factors of stroke-like migraine attacks after radiation therapy (SMART) syndrome have not been fully explored. This study aimed to assess clinical and imaging features to predict the clinical outcome of SMART syndrome.

MATERIALS AND METHODS

We retrospectively reviewed the clinical manifestations and imaging findings of 20 patients with SMART syndrome (median age, 48 years; 5 women) from January 2016 to January 2020 at 4 medical centers. Patient demographics and MR imaging features at the time of diagnosis were reviewed. This cohort was divided into 2 groups based on the degree of clinical improvement (completely versus incompletely recovered). The numeric and categoric variables were compared as appropriate.

RESULTS

There were statistically significant differences between the completely recovered group (n = 11; median age, 44 years; 2 women) and the incompletely recovered group (n = 9; median age, 55 years; 3 women) in age, months of follow-up, and the presence of steroid treatment at diagnosis (P = .028, .002, and .01, respectively). Regarding MR imaging features, there were statistically significant differences in the presence of linear subcortical WM susceptibility abnormality, restricted diffusion, and subcortical WM edematous changes in the acute SMART region (3/11 versus 8/9, P = .01; 0/11 versus 4/9, P = .026; and 2/11 versus 7/9, P = .022, respectively). Follow-up MRIs showed persistent susceptibility abnormality (11/11) and subcortical WM edematous changes (9/9), with resolution of restricted diffusion (4/4).

CONCLUSIONS

Age, use of steroid treatment at the diagnosis of SMART syndrome, and MR imaging findings of abnormal susceptibility signal, restricted diffusion, and subcortical WM change in the acute SMART region can be prognostic factors in SMART syndrome.

Multi-band- and in-plane-accelerated diffusion MRI enabled by model-based deep learning in q-space and its extension to learning in the spherical harmonic domain

PURPOSE

To propose a new method for the recovery of combined in-plane- and multi-band (MB)-accelerated diffusion MRI data.

METHODS

Combining MB acceleration with in-plane acceleration is crucial to improve the time efficiency of high (angular and spatial) resolution diffusion scans. However, as the MB factor and in-plane acceleration increase, the reconstruction becomes challenging due to the heavy aliasing. The new reconstruction utilizes an additional q-space prior to constrain the recovery, which is derived from the previously proposed qModeL framework. Specifically, the qModeL prior provides a pre-learned representation of the diffusion signal space to which the measured data belongs. We show that the pre-learned q-space prior along with a model-based iterative reconstruction that accommodate multi-band unaliasing, can efficiently reconstruct the in-plane- and MB-accelerated data. The power of joint reconstruction is maximally utilized by using an incoherent under-sampling pattern in the k-q domain. We tested the proposed method on single- and multi-shell data, with high/low angular resolution, high/low spatial resolution, healthy/abnormal tissues, and 3T/7T field strengths. Furthermore, the learning is extended to the spherical harmonic basis, to provide a rotational invariant learning framework.

RESULTS

The qModeL joint reconstruction is shown to simultaneously unalias and jointly recover DWIs with reasonable accuracy in all the cases studied. The reconstruction error from 18-fold accelerated multi-shell datasets was <3%. The microstructural maps derived from the accelerated acquisitions also exhibit reasonable accuracy for both healthy and abnormal tissues. The deep learning (DL)-enabled reconstructions are comparable to those derived using traditional methods.

CONCLUSION

qModeL enables the joint recovery of combined in-plane- and MB-accelerated dMRI utilizing DL.

Computed Tomography Perfusion-Based Prediction of Core Infarct and Tissue at Risk: Can Artificial Intelligence Help Reduce Radiation Exposure?

BACKGROUND AND PURPOSE

We explored the feasibility of automated, arterial input function independent, vendor neutral prediction of core infarct, and penumbral tissue using complete and partial computed tomographic perfusion data sets through neural networks.

METHODS

Using retrospective computed tomographic perfusion data from 57 patients, split as training/validation (60%/40%), we developed and validated separate 2-dimensional U-net models for cerebral blood flow (CBF) and time to maximum (Tmax) maps calculation to predict core infarct and tissue at risk, respectively. Once trained, the full sets of 28 input images were sequentially reduced to equitemporal 14, 10, and 7 time points. The averaged structural similarity index measure between the model-derived images and ground truth perfusion maps was compared. Volumes for core infarct and Tmax were compared using the Pearson correlation coefficient.

RESULTS

Both CBF and Tmax maps derived using 28 and 14 time points had similar structural similarity index measure (0.80-0.81; P>0.05) when compared with ground truth images. The Pearson correlation for the CBF and Tmax volumes derived from the model using 28-tp with ground truth volumes derived from the RAPID software was 0.69 for CBF and 0.74 for Tmax. The predicted maps were fully concordant in terms of laterality to the commercial perfusion maps. The mean Dice scores were 0.54 for the core infarct and 0.63 for the hypoperfusion maps.

CONCLUSIONS

Artificial intelligence model-derived volumes show good correlation with RAPID-derived volumes for CBF and Tmax. Within the constraints of a small sample size, the perfusion map quality is similar when using 14-tp instead of 28-tp. Our findings provide proof of concept that vendor neutral artificial intelligence models for computed tomographic perfusion processing using complete or partial image data sets appear feasible. The model accuracy could be further optimized using larger data sets.

Single Center Experience with Isolated Spinal Cord Neurosarcoidosis

INTRODUCTION

Isolated spinal cord neurosarcoidosis is extremely rare. The potential implications of long-term immunosuppressant therapy make correct diagnosis imperative. However, there are challenges inherent in isolated spinal cord involvement that require a multidisciplinary approach. Here we present the largest series of definite and possible isolated spinal neurosarcoidosis and discuss our institutional experience in managing this rare but morbid condition.

METHODS

A retrospective review was performed to identify all neurosarcoidosis cases starting from 2002 to 2020 at our institution. Patients were screened for cases of isolated spinal neurosarcoidosis. A descriptive analysis was performed for each case.

RESULTS

A total of 64 cases of neurosarcoidosis were identified. The spine was involved in 26 (40.6%) patients. Only 4 (6.3%) cases had isolated spinal cord involvement. A full medical and imaging workup was performed in determining isolated spinal cord involvement. Three patients subsequently underwent surgical biopsy, and 1 did not undergo biopsy because of patient preference. One of the patients who underwent biopsy had an initial nondiagnostic biopsy and had a repeat biopsy. Corticosteroids were employed in all cases with additional immunosuppressive agents for maintenance therapy and refractory cases. All showed radiographic improvement and were clinically stable to improved.

CONCLUSION

Isolated spinal cord involvement of neurosarcoidosis is rare and can present challenges in diagnosis. A biopsy can be performed when necessary. However, a biopsy of the spinal cord carries inherent risks and may not always be possible or result in a nondiagnostic sample. In the setting of high clinical suspicion, maximal medical therapy is still employed.

Vascular Involvement in Neurosarcoidosis: Early Experiences From Intracranial Vessel Wall Imaging

BACKGROUND AND OBJECTIVES

Cerebrovascular manifestations in neurosarcoidosis (NS) were previously considered rare but are being increasingly recognized. We report our preliminary experience in patients with NS who underwent high-resolution vessel wall imaging (VWI).

METHODS

A total of 13 consecutive patients with NS underwent VWI. Images were analyzed by 2 neuroradiologists in consensus. The assessment included segment-wise evaluation of larger- and medium-sized vessels (internal carotid artery, M1-M3 middle cerebral artery; A1-A3 anterior cerebral artery; V4 segments of vertebral arteries; basilar artery; and P1-P3 posterior cerebral artery), lenticulostriate perforator vessels, and medullary and deep cerebral veins. Cortical veins were not assessed due to flow-related artifacts. Brain biopsy findings were available in 6 cases and were also reviewed.

RESULTS

Mean patient age was 54.9 years (33-71 years) with an M:F of 8:5. Mean duration between initial diagnosis and VWI study was 18 months. Overall, 9/13 (69%) patients had vascular abnormalities. Circumferential large vessel enhancement was seen in 3/13 (23%) patients, whereas perforator vessel involvement was seen in 6/13 (46%) patients. Medullary and deep vein involvement was also seen in 6/13 patients. In addition, 7/13 (54%) patients had microhemorrhages in susceptibility-weighted imaging, and 4/13 (31%) had chronic infarcts. On biopsy, 5/6 cases showed perivascular granulomas with vessel wall involvement in all 5 cases.

DISCUSSION

Our preliminary findings suggest that involvement of intracranial vascular structures may be a common finding in patients with NS and should be routinely looked for. These findings appear concordant with previously reported autopsy literature and need to be validated on a larger scale.

Outcomes and complications of surgical treatment of anterior osteophytes causing dysphagia: Single center experience

STUDY DESIGN

Retrospective case series.

OBJECTIVE

To better understand the functional swallow outcomes, cervical balance, and surgical complications, we examined patients with anterior osteophytes and dysphagia who were treated operatively.

SUMMARY OF BACKGROUND DATA

Anterior osteophytes from diffuse idiopathic skeletal hyperostosis (DISH) or degenerative etiology of the cervical spine can cause dysphagia from mechanical compression of the esophagus. Osteophytectomy is generally accepted as a safe surgical treatment, but the risk of instability is unclear. The potential for associated complications must be considered.

METHODS

Patients who had anterior osteophytes and dysphagia from 2005 to 2020 were reviewed retrospectively. Demographics, radiographic parameters, functional swallow outcome, and complications were examined.

RESULTS

There were 15 patients identified treated surgically. Increased osteophyte height positively correlated with severity of dysphagia with Pearson coefficient of 0.53 (p = 0.042). Functional Outcome Swallowing Scale (FOSS) scores improved after surgical treatment from median of 2 to 0 (p = 0.002). C2-7 SVA did increase by 8 mm (p = 0.007) but was generally well tolerated. There was a 27% complication rate including a case of C5 lateral mass fracture with central cord syndrome after a fall 4 days following osteophytectomy. There was one patient who was preoperatively dependent on gastrostomy tube who required a tracheostomy and had continued reliance on the gastrostomy tube.

CONCLUSION

Surgical treatment of anterior osteophytes causing dysphagia with osteophytectomy can lead to overall improved FOSS scores for most patients. However, a high preoperative FOSS score may be a prognostic indicator of poor postoperative functional swallow outcome. It is important to consider the potential for instability when osteophytectomy is performed at 3 or more spinal segments.

Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models

Side experiments are performed on radiomics models to improve their reproducibility. We measure the impact of myocardial masks, radiomic side experiments and data augmentation for information transfer (DAFIT) approach to differentiate patients with and without pulmonary hypertension (PH) using cardiac MRI (CMRI) derived radiomics. Feature extraction was performed from the left ventricle (LV) and right ventricle (RV) myocardial masks using CMRI in 82 patients (42 PH and 40 controls). Various side study experiments were evaluated: Original data without and with intraclass correlation (ICC) feature-filtering and DAFIT approach (without and with ICC feature-filtering). Multiple machine learning and feature selection strategies were evaluated. Primary analysis included all PH patients with subgroup analysis including PH patients with preserved LVEF (≥ 50%). For both primary and subgroup analysis, DAFIT approach without feature-filtering was the highest performer (AUC 0.957-0.958). ICC approaches showed poor performance compared to DAFIT approach. The performance of combined LV and RV masks was superior to individual masks alone. There was variation in top performing models across all approaches (AUC 0.862-0.958). DAFIT approach with features from combined LV and RV masks provide superior performance with poor performance of feature filtering approaches. Model performance varies based upon the feature selection and model combination.

High-resolution vessel wall imaging after mechanical thrombectomy

OBJECTIVES

High-resolution magnetic resonance imaging has the potential of characterising arterial wall changes after endovascular mechanical thrombectomy. The purpose of this study is to evaluate high-resolution magnetic resonance imaging features of large intracranial arteries following mechanical thrombectomy.

METHODS

Patients who presented with acute ischaemic stroke due to large vessel occlusion and underwent mechanical thrombectomy were prospectively recruited. Subjects underwent high-resolution magnetic resonance imaging within 24 hours of the procedure. Magnetic resonance imaging sequences included whole brain T1 pre and post-contrast black-blood imaging, three-dimensional T2, contrast-enhanced magnetic resonance angiography and susceptibility-weighted imaging. Arterial wall enhancement was objectively assessed after normalisation with the pituitary stalk. The contrast ratio of target vessels was compared with non-affected reference vessels.

RESULTS

Twenty patients with 22 target vessels and 20 reference vessels were included in the study. Sixteen patients were treated with stentriever with or without aspiration, and four with contact aspiration only. Significantly higher arterial wall enhancement was identified on the target vessel when compared to the reference vessel (U = 22.5, P < 0.01). The stentriever group had an 82% increase in the contrast ratio of the target vessel (x̄ = 0.75 ± 0.21) when compared to the reference vessel (x̄ = 0.41 ± 0.13), whereas the contact aspiration group had a 64% increase of the contrast ratio difference between target (x̄ = 0.62 ± 0.07) and reference vessels (x̄ = 0.38 ± 0.12). Approximately 65% of patients in the stentriever group had a positive parenchymal susceptibility-weighted imaging versus 25% in the contact aspiration group. There was no statistically significant correlation between susceptibility-weighted imaging volume and the percentage increase in the contrast ratio (r_s = 0.098, P = 0.748).

CONCLUSIONS

This prospective pilot study used the objective quantification of arterial wall enhancement in determining arterial changes after mechanical thrombectomy. Preliminary data suggest that the use of stentrievers is associated with a higher enhancement as compared to reperfusion catheters.

Machine learning based differentiation of glioblastoma from brain metastasis using MRI derived radiomics

Few studies have addressed radiomics based differentiation of Glioblastoma (GBM) and intracranial metastatic disease (IMD). However, the effect of different tumor masks, comparison of single versus multiparametric MRI (mp-MRI) or select combination of sequences remains undefined. We cross-compared multiple radiomics based machine learning (ML) models using mp-MRI to determine optimized configurations. Our retrospective study included 60 GBM and 60 IMD patients. Forty-five combinations of ML models and feature reduction strategies were assessed for features extracted from whole tumor and edema masks using mp-MRI [T1W, T2W, T1-contrast enhanced (T1-CE), ADC, FLAIR], individual MRI sequences and combined T1-CE and FLAIR sequences. Model performance was assessed using receiver operating characteristic curve. For mp-MRI, the best model was LASSO model fit using full feature set (AUC 0.953). FLAIR was the best individual sequence (LASSO-full feature set, AUC 0.951). For combined T1-CE/FLAIR sequence, adaBoost-full feature set was the best performer (AUC 0.951). No significant difference was seen between top models across all scenarios, including models using FLAIR only, mp-MRI and combined T1-CE/FLAIR sequence. Top features were extracted from both the whole tumor and edema masks. Shape sphericity is an important discriminating feature.

Multiparametric magnetic resonance imaging and positron emission tomography findings in neurodegenerative diseases: Current status and future directions

Neurodegenerative diseases (NDDs) are characterized by progressive neuronal loss, leading to dementia and movement disorders. NDDs broadly include Alzheimer's disease, frontotemporal lobar degeneration, parkinsonian syndromes, and prion diseases. There is an ever-increasing prevalence of mild cognitive impairment and dementia, with an accompanying immense economic impact, prompting efforts aimed at early identification and effective interventions. Neuroimaging is an essential tool for the early diagnosis of NDDs in both clinical and research settings. Structural, functional, and metabolic imaging modalities, including magnetic resonance imaging (MRI) and positron emission tomography (PET), are widely available. They show encouraging results for diagnosis, monitoring, and treatment response evaluation. The current review focuses on the complementary role of various imaging modalities in relation to NDDs, the qualitative and quantitative utility of newer MRI techniques, novel radiopharmaceuticals, and integrated PET/MRI in the setting of NDDs.

Glioblastoma and primary central nervous system lymphoma: differentiation using MRI derived first-order texture analysis - a machine learning study

OBJECTIVES

To evaluate the diagnostic performance of multiple machine learning classifier models derived from first-order histogram texture parameters extracted from T1-weighted contrast-enhanced images in differentiating glioblastoma and primary central nervous system lymphoma.

METHODS

Retrospective study with 97 glioblastoma and 46 primary central nervous system lymphoma patients. Thirty-six different combinations of classifier models and feature selection techniques were evaluated. Five-fold nested cross-validation was performed. Model performance was assessed for whole tumour and largest single slice using receiver operating characteristic curve.

RESULTS

The cross-validated model performance was relatively similar for the top performing models for both whole tumour and largest single slice (area under the curve 0.909-0.924). However, there was a considerable difference between the worst performing model (logistic regression with full feature set, area under the curve 0.737) and the highest performing model for whole tumour (least absolute shrinkage and selection operator model with correlation filter, area under the curve 0.924). For single slice, the multilayer perceptron model with correlation filter had the highest performance (area under the curve 0.914). No significant difference was seen between the diagnostic performance of the top performing model for both whole tumour and largest single slice.

CONCLUSIONS

T1 contrast-enhanced derived first-order texture analysis can differentiate between glioblastoma and primary central nervous system lymphoma with good diagnostic performance. The machine learning performance can vary significantly depending on the model and feature selection methods. Largest single slice and whole tumour analysis show comparable diagnostic performance.

Atypical presentation of rotational vertebral artery insufficiency: illustrative case

BACKGROUND

Rotational vertebral artery insufficiency (RVAI), also known as bow hunter’s syndrome, is an uncommon cause of vertebrobasilar insufficiency that leads to signs of posterior circulation ischemia during head rotation. RVAI can be subdivided on the basis of the anatomical location of vertebral artery compression into atlantoaxial RVAI (pathology at C1-C2) or subaxial RVAI (pathology below C2). Typically, RVAI is only seen with contralateral vertebral artery pathologies, such as atherosclerosis, hypoplasia, or morphological atypia.

OBSERVATIONS

The authors present a unique case of atlantoaxial RVAI due to rotational instability, causing marked subluxation of the C1-C2 facet joints. This case is unique in both the mechanism of compression and the lack of contralateral vertebral artery pathology. The patient was successfully treated with posterior C1-C2 instrumentation and fusion.

LESSONS

When evaluating patients for RVAI, neurosurgeons should be aware of the variety of pathological causes, including rotational instability from facet joint subluxation. Due to the heterogeneous nature of the pathologies causing RVAI, care must be taken to decide if conservative management or surgical correction is the right course of action. Because of this heterogeneous nature, there is no set guideline for the treatment or management of RVAI.

Survival prediction in glioblastoma on post-contrast magnetic resonance imaging using filtration based first-order texture analysis: Comparison of multiple machine learning models

OBJECTIVE

Magnetic resonance texture analysis (MRTA) is a relatively new technique that can be a valuable addition to clinical and imaging parameters in predicting prognosis. In the present study, we investigated the efficacy of MRTA for glioblastoma survival using T1 contrast-enhanced (CE) images for texture analysis.

METHODS

We evaluated the diagnostic performance of multiple machine learning models based on first-order histogram statistical parameters derived from T1-weighted CE images in the survival stratification of glioblastoma multiforme (GBM). Retrospective evaluation of 85 patients with GBM was performed. Thirty-six first-order texture parameters at six spatial scale filters (SSF) were extracted on the T1 CE axial images for the whole tumor using commercially available research software. Several machine learning classification models (in four broad categories: linear, penalized linear, non-linear, and ensemble classifiers) were evaluated to assess the survival prediction performance using optimal features. Principal component analysis was used prior to fitting the linear classifiers in order to reduce the dimensionality of the feature inputs. Fivefold cross-validation was used to partition the data iteratively into training and testing sets. The area under the receiver operating characteristic curve (AUC) was used to assess the diagnostic performance.

RESULTS

The neural network model was the highest performing model with the highest observed AUC (0.811) and cross-validated AUC (0.71). The most important variable was the age at diagnosis, with mean and mean of positive pixels (MPP) for SSF = 0 being the second and third most important, followed by skewness for SSF = 0 and SSF = 4.

CONCLUSIONS

First-order texture features, when combined with age at presentation, show good accuracy in predicting GBM survival.

MRI characteristics and resectability in spinal cord glioma

OBJECTIVE

The histopathology of intramedullary spinal cord tumors (IMSCT) can be suspected from the MRI features and characteristics. Ultimately, the confirmation of diagnosis requires surgery. This retrospective study addresses MRI features including homogeneity of enhancement, margination, and associated syrinx in intramedullary astrocytomas (IMA) and ependymomas (IME) that assist in diagnosis and predict resectability of these tumors.

METHODS

Single-center retrospective analysis of IMA and IME cases since 2005 extracted from the departmental registry/electronic medical records post IRB approval (IRB 201,710,760). We compared imaging findings (enhancement, margination, homogeneity, and associated syrinxes) between tumor types and examined patient outcomes.

RESULTS

There were 18 IME and 21 IMA. On preoperative MRI, IME was favored to have homogenous enhancement (OR 1.8, p = 0.0001), well-marginated (p < 0.0001, OR 0.019 [95 % CI 0.002-0.184]), and associated syrinx (p = 0.015, OR 0.192 [95 % CI 0.049-0.760]). Total excision, subtotal excision, and biopsy were performed in 12, 5, and 1 patients in the IME cohort, respectively. In the IMA group, tumors were heterogeneous and poorly marginated in 20 of the 21 patients. Total excision, subtotal excision, and biopsy were undertaken in 2, 13, and 6 patients, respectively. The success of excision was predicted by MRI, with a significant difference in the extent of resection between IME and IMA (X2 = 14.123, p = 0.001). In terms of outcome, ordinal regression analysis showed that well-margined tumors and those with homogeneous enhancement were associated with a better postoperative McCormick score. Extent of resection had statistically significant survival (p = 0.026) and recurrence-free survival (p = 0.008) benefits.

CONCLUSION

The imaging characteristics of IME and IMA have meaningful clinical significance. Homogeneity, margination, and associated syrinxes in IME can predict resectability and complexity of surgery.