Inversion Recovery Magnetic Resonance Imaging Research Articles

Technical note: A spiral fluid-attenuated inversion recovery magnetic resonance imaging technique for stereotactic radiosurgery treatment planning for trigeminal neuralgia.

Magnetic resonance imaging (MRI) is commonly used in treatment planning for stereotactic radiosurgery (SRS) of trigeminal neuralgia (TN). With current MRI techniques, the delineation of the trigeminal nerve root entry zone (REZ) may be degraded due to poor contrast and artifacts. The purpose of this work is to develop an MRI technique with better delineation of the trigeminal nerve REZ to improve SRS treatment planning for TN. A spiral fluid-attenuated inversion recovery (FLAIR) MRI technique was developed to improve image quality by improving tissue contrast, fluid suppression, artifact reduction, and signal-to-noise ratio (SNR). A concomitant-phase compensation method based on spiral gradient waveforms was implemented to minimize artifacts due to magnetic field change induced by the metal frame used in Gamma Knife treatment planning. The image quality of spiral FLAIR was assessed in four healthy volunteers. The geometric accuracy was quantitatively evaluated by registering spiral FLAIR to computed tomography (CT) images and comparing it with existing MRI techniques. The spiral FLAIR technique demonstrated better delineation of the trigeminal nerve REZ, improved tissue contrast of the brain stem, and minimized flow artifacts, compared to steady-state free precession (SSFP) MRI. Spiral FLAIR also improved fluid suppression, SNR, and artifacts, which contributed to better delineation of the trigeminal nerve REZ compared to conventional Cartesian FLAIR. The measured mean (± standard deviation) distance between spiral FLAIR and CT images is 0.98 ± 0.56mm, comparable to 0.40 ± 0.26mm in 3T T1 spoiled gradient echo (T1-SPGR), 0.59 ± 0.25mm in 3T SSFP, 0.66 ± 0.38mm in 1.5T T1-SPGR, and 0.61 ± 0.25mm in 1.5T Cartesian FLAIR. A spiral FLAIR technique with improved image quality and good geometric accuracy provides a potential alternative for treatment planning in SRS for TN patients.

Automatic quantification of white matter hyperintensities on T2-weighted fluid attenuated inversion recovery magnetic resonance imaging

White matter hyperintensities (WMH) are areas of increased signal visualized on T2-weighted fluid attenuated inversion recovery (FLAIR) brain magnetic resonance imaging (MRI) sequences. They are typically attributed to small vessel cerebrovascular disease in the context of aging. Among older adults, WMH are associated with risk of cognitive decline and dementia, stroke, and various other health outcomes. There has been increasing interest in incorporating quantitative WMH measurement as outcomes in clinical trials, observational research, and clinical settings. Here, we present a novel, fully automated, unsupervised detection algorithm for WMH segmentation and quantification. The algorithm uses a robust preprocessing pipeline, including brain extraction and a sample-specific mask that incorporates spatial information for automatic false positive reduction, and a half Gaussian mixture model (HGMM). The method was evaluated in 24 participants with varying degrees of WMH (4.9–78.6 cm3) from a community-based study of aging and dementia with dice coefficient, sensitivity, specificity, correlation, and bias relative to the ground truth manual segmentation approach performed by two expert raters. Results were compared with those derived from commonly used available WMH segmentation packages, including SPM lesion probability algorithm (LPA), SPM lesion growing algorithm (LGA), and Brain Intensity AbNormality Classification Algorithm (BIANCA). The HGMM algorithm derived WMH values that had a dice score of 0.87, sensitivity of 0.89, and specificity of 0.99 compared to ground truth. White matter hyperintensity volumes derived with HGMM were strongly correlated with ground truth values (r = 0.97, p = 3.9e-16), with no observable bias (−1.1 [−2.6, 0.44], p-value = 0.16). Our novel algorithm uniquely uses a robust preprocessing pipeline and a half-Gaussian mixture model to segment WMH with high agreement with ground truth for large scale studies of brain aging.

Morphometric evaluation of traumatic axonal injury and the correlation with post-traumatic cerebral atrophy and functional outcome.

Imaging plays a crucial role in the diagnosis, prognosis and follow-up of traumatic brain injury. Whereas computed tomography plays a pivotal role in the acute setting, magnetic resonance imaging is best suited to detect the true extent of traumatic brain injury, and more specifically diffuse axonal injury. Post-traumatic brain atrophy is a well-known complication of traumatic brain injury. This study investigated the correlation between diffuse axonal injury detected with fluid-attenuated inversion recovery and susceptibility-weighted imaging magnetic resonance imaging, post-traumatic brain atrophy and functional outcome (Glasgow outcome scale - extended). Twenty patients with a closed head injury and diffuse axonal injury detected with fluid-attenuated inversion recovery and susceptibility-weighted imaging were included. The total volumes of the diffuse axonal injury fluid-attenuated inversion recovery lesions were determined for each subject's initial (<14 days) and follow-up magnetic resonance scan (average: day 303 ± 83 standard deviation). The different brain volumes were automatically quantified using a validated and both US Food and Drug Administration-cleared and CE-marked machine learning algorithm (icobrain). The number of susceptibility-weighted imaging lesions and functional outcome scores (Glasgow outcome scale - extended) were retrieved from the Collaborative European NeuroTrauma Effectiveness Research Traumatic Brain Injury dataset. The volumetric fluid-attenuated inversion recovery diffuse axonal injury lesion load showed a significant inverse correlation with functional outcome (Glasgow outcome scale - extended) (r = -0.57; P = 0.0094) and white matter volume change (r = -0.50; P = 0.027). In addition, white matter volume change correlated significantly with the Glasgow outcome scale - extended score (P = 0.0072; r = 0.58). Moreover, there was a strong inverse correlation between longitudinal fluid-attenuated inversion recovery lesion volume change and whole brain volume change (r = -0.63; P = 0.0028). No significant correlation existed between the number of diffuse axonal injury susceptibility-weighted imaging lesions, brain atrophy and functional outcome. Volumetric analysis of diffuse axonal injury on fluid-attenuated inversion recovery imaging and automated brain atrophy calculation are potentially useful tools in the clinical management and follow-up of traumatic brain injury patients with diffuse axonal injury.

Clinical usefulness of labyrinthine three-dimensional fluid-attenuated inversion recovery magnetic resonance images in idiopathic sudden sensorineural hearing loss.

Literature on the use of current magnetic resonance imaging (MRI) for patients with idiopathic sudden sensorineural hearing loss (ISSNHL) is reviewed, emphasizing the role of three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) MRI. The discussion focuses on the diagnostic role of temporal bone MRI using 3D-FLAIR and the relationship between MRI findings, clinical symptoms, and hearing outcome. The currently suggested MRI protocol for SSNHL includes a 3D T2-weighted steady-state free procession sequence or its equivalent, pre and postcontrast T1-weighted, and pre and postcontrast 3D-FLAIR sequences. The 3D-FLAIR image identifies an underlying labyrinthine condition in 24-57% of patients with ISSNHL, contributing to understanding the pathophysiologic mechanisms (e.g., labyrinthitis or labyrinthine hemorrhage). Recent studies demonstrated consistent results that initial hearing loss could be related to the signal change on the 3D-FLAIR image. Various results on 3D-FLAIR image value prediction for the final hearing outcome were shown. 3D-FLAIR MRI application identifies an underlying labyrinthine condition. Abnormal MRI findings correlate with initial hearing loss and accompanying symptoms and hearing outcome. Performing temporal bone MRI with 3D-FLAIR sequence may clarify probable ISSNHL pathophysiology, improve diagnostic accuracy, provide prognostic information to physicians, and possibly guide toward a more specific treatment.

A Crowdsourced Consensus on Supratotal Resection Versus Gross Total Resection for Anatomically Distinct Primary Glioblastoma.

Gross total resection (GTR) of contrast-enhancing tumor is associated with increased survival in primary glioblastoma. Recently, there has been increasing interest in performing supratotal resections (SpTRs) for glioblastoma. To address the published results, which have varied in part due to lack of consensus on the definition and appropriate use of SpTR. A crowdsourcing approach was used to survey 21 neurosurgical oncologists representing 14 health systems nationwide. Participants were presented with 11 definitions of SpTR and asked to rate the appropriateness of each definition. Participants reviewed T1-weighed postcontrast and fluid-attenuated inversion-recovery magnetic resonance imaging for 22 anatomically distinct glioblastomas. Participants were asked to assess the tumor location's eloquence, the perceived equipoise of enrolling patients in a randomized trial comparing gross total to SpTR, and their personal treatment plans. Most neurosurgeons surveyed (n=18, 85.7%) agree that GTR plus resection of some noncontrast enhancement is an appropriate definition for SpTR. Overall, moderate inter-rater agreement existed regarding eloquence, equipoise, and personal treatment plans. The 4 neurosurgeons who had performed >10 SpTRs for glioblastomas in the past year were more likely to recommend it as their treatment plan (P <.005). Cases were divided into 3 anatomically distinct groups based upon perceived eloquence. Anterior temporal and right frontal glioblastomas were considered the best randomization candidates. We established a consensus definition for SpTR of glioblastoma and identified anatomically distinct locations deemed most amenable to SpTR. These results may be used to plan prospective trials investigating the potential clinical utility of SpTR for glioblastoma.

Neuroprotective Effects of GV1001 in Animal Stroke Model and Neural Cells Subject to Oxygen-Glucose Deprivation/Reperfusion Injury.

Background and Purpose Previous studies have revealed the diverse neuroprotective effects of GV1001. In this study, we investigated the effects of GV1001 on focal cerebral ischemia-reperfusion injury (IRI) in rats and oxygen-glucose deprivation/reoxygenation (OGD/R)-induced injury in neural stem cells (NSCs) and cortical neurons. Methods Focal cerebral IRI was induced by transient middle cerebral artery occlusion (MCAO). Brain diffusion-weighted imaging (DWI) was performed 2 hours after occlusion, and a total of 37 rats were treated by reperfusion with GV1001 or saline 2 hours after occlusion. Fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging, immunohistochemistry, and neurobehavioral function analyses were performed. Additionally, OGD/R-injured NSCs and cortical neurons were treated with different GV1001 concentrations. Cell viability, proliferation, migration, and oxidative stress were determined by diverse molecular analyses. Results In the stroke model, GV1001 protected neural cells against IRI. The most effective dose of GV1001 was 60 μM/kg. The infarct volume on FLAIR 48 hours after MCAO compared to lesion volume on DWI showed a significantly smaller ratio in the GV1001-treated group. GV1001-treated rats exhibited better behavioral functions than the saline-treated rats. Treatment with GV1001 increased the viability, proliferation, and migration of the OGD/R-injured NSCs. Free radicals were significantly restored by treatment with GV1001. These neuroprotective effects of GV1001 have also been demonstrated in OGD/R-injured cortical neurons. Conclusions The results suggest that GV1001 has neuroprotective effects against IRI in NSCs, cortical neurons, and the rat brain. These effects are mediated through the induction of cellular proliferation, mitochondrial stabilization, and anti-apoptotic, anti-aging, and antioxidant effects.

Assessment of Artificial Intelligence Automatic Multiple Sclerosis Lesion Delineation Tool for Clinical Use

PurposeTo implement and validate an existing algorithm for automatic delineation of white matter lesions on magnetic resonance imaging (MRI) in patients with multiple sclerosis (MS) on a local single-center dataset.MethodsWe implemented a white matter hyperintensity segmentation model, based on a 2D convolutional neural network, using the conventional T2-weighted fluid attenuated inversion recovery (FLAIR) MRI sequence as input. The model was adapted for delineation of MS lesions by further training on a local dataset of 93 MS patients with a total of 3040 lesions. A quantitative evaluation was performed on ten test patients, in which model-generated masks were compared to manually delineated masks from two expert delineators. A subsequent qualitative evaluation of the implemented model was performed by two expert delineators, in which generated delineation masks on a clinical dataset of 53 patients were rated acceptable (< 10% errors) or unacceptable (> 10% errors) based on the total number of true lesions.ResultsThe quantitative evaluation resulted in an average accuracy score (F1) of 0.71, recall of 0.77 and dice similarity coefficient of 0.62. Our implemented model obtained the highest scores in all three metrics, when compared to three out of the box lesion segmentation models. In the clinical evaluation an average of 94% of our 53 model-generated masks were rated acceptable.ConclusionAfter adaptation to our local dataset, the implemented segmentation model was able to delineate MS lesions with a high clinical value as rated by delineation experts while outperforming popular out of the box applications. This serves as a promising step towards implementation of automatic lesion delineation in our MS clinic.

White Matter Hyperintensity Burden and Collateral Circulation in Large Vessel Occlusion Stroke.

White matter hyperintensity (WMH), a marker of chronic cerebral small vessel disease, might impact the recruitment of leptomeningeal collaterals. We aimed to assess whether the WMH burden is associated with collateral circulation in patients treated by endovascular thrombectomy for anterior circulation acute ischemic stroke. Consecutive acute ischemic stroke due to anterior circulation large vessel occlusion and treated with endovascular thrombectomy from January 2015 to December 2017 were included. WMH volumes (periventricular, deep, and total) were assessed by a semiautomated volumetric analysis on fluid-attenuated inversion recovery-magnetic resonance imaging. Collateral status was graded on baseline catheter angiography using the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology grading system (good when ≥3). We investigated associations of WMH burden with collateral status. A total of 302 patients were included (mean age, 69.1±19.4 years; women, 55.6%). Poor collaterals were observed in 49.3% of patients. Median total WMH volume was 3.76 cm3 (interquartile range, 1.09-11.81 cm3). The regression analyses showed no apparent relationship between WMH burden and the collateral status measured at baseline angiography (adjusted odds ratio, 0.987 [95% CI, 0.971-1.003]; P=0.12). WMH burden exhibits no overt association with collaterals in large vessel occlusive stroke.

Comparison of Phase-Sensitive Inversion Recovery and Conventional Magnetic Resonance Imaging for Detection of Cortical Plaques in MS Patients

Background: Precise detection and classification of intracortical (IC) lesions in multiple sclerosis (MS) patients are very important for understanding their role in disease progression and determining their effects on the clinical presentations of the disease. Objectives: This study aimed to evaluate the efficacy of phase-sensitive inversion recovery (PSIR) in delineation of cortical lesions in MS patients. Patients and Methods: This cross-sectional, single-center study was performed among 38 patients with the mean age of 31 years, who were recruited from December 2018 to August 2020. All MS patients underwent magnetic resonance imaging (MRI), using a 1.5-Tesla scanner. Two expert neuroradiologists interpreted the fluid-attenuated inversion recovery (FLAIR), T2-weighted turbo spin echo (T2W-TSE), and PSIR images. The lesions were classified as purely IC, mixed gray/white matter (WM) [leukocortical (LC)], and juxtacortical (JC). The number of lesions in each region was compared between the FLAIR, T2W-TSE, and PSIR sequences. Results: The number of cortical lesions (IC and LC) was significantly higher in PSIR compared to T2W-TSE and Fluid attenuated inversion recovery (FLAIR) (P &lt; 0.001), while the number of JC lesions was lower; in other words, the mean number of plaques was higher in T2W-TSE and FLAIR as compared to PSIR. Conclusion: The PSIR sequence significantly improved the delineation of cortical lesions and could be useful in monitoring cortical injuries and disease progression in MS patients.

Cortical involvement and leptomeningeal inflammation in myelin oligodendrocyte glycoprotein antibody disease: A three-dimensional fluid-attenuated inversion recovery MRI study

Background: Cortical demyelination and meningeal inflammation have been detected neuropathologically in multiple sclerosis (MS) and recently in myelin oligodendrocyte glycoprotein antibody disease (MOGAD). Objectives: To assess in vivo cortical and leptomeningeal involvement in MOGAD. Methods: We prospectively evaluated 11 MOGAD and 12 relapsing-remitting MS (RRMS) patients combining three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) and 3D-T1-weighted (3D-T1w) sequences at 3-Tesla magnetic resonance imaging (MRI). Leptomeningeal contrast enhancement (LMCE) was assessed on 3D-FLAIR post-gadolinium (3D-FLAIRGd). Cerebral cortical lesions (CCLs) were classified as either intracortical–subpial (IC–SP) or leukocortical (LC). Results: CCLs were present in 8/11 MOGAD and 12/12 RRMS patients, with the number of CCLs being significantly lower in MOGAD (median (interquartile range (IQR)) 3 (0.5–4) vs 12 (4.75–19), p = 0.0032). In MOGAD, IC–SP lesions were slightly more prevalent than LC lesions (2 (0–2.5) vs 1 (0–2), p = 0.6579); whereas in RRMS, IC–SP lesions were less prevalent than LC lesions (3.5 (2.75–5.5) vs 9 (2–12.75), p = 0.27). LMCE was observed in 3/11 MOGAD and 1/12 RRMS patients; MOGAD with LMCE showed an increased median number of CCLs compared with MOGAD without LMCE (8 (4–9) vs 2.5 (0.75–3.25), p = 0.34). No correlation was observed between MOGAD MRI findings and (a) MOGAD duration, (b) serum MOG-immunoglobulin G1 titers, and (c) oligoclonal band presence. Conclusion: We described cortical lesion topography and detected for the first time LMCE using 3D-FLAIRGd sequences in MOGAD patients.

Assessment of Magnetic Resonance Imaging Changes and Functional Outcomes Among Adults With Severe Herpes Simplex Encephalitis

Current guidelines recommend brain magnetic resonance imaging (MRI) for clinical management of patients with severe herpes simplex encephalitis (HSE). However, the prognostic value of brain imaging has not been demonstrated in this setting. To investigate the association between early brain MRI data and functional outcomes of patients with HSE at 90 days after intensive care unit (ICU) admission. This multicenter cohort study was conducted in 34 ICUs in France from 2007 to 2019 and recruited all patients who received a clinical diagnosis of encephalitis and exhibited cerebrospinal fluid positivity for herpes simplex virus DNA in the polymerase chain reaction analysis. Data analysis was performed from January to April 2020. All patients underwent a standard brain MRI during the first 30 days after ICU admission. MRI acquisitions were analyzed by radiologists blinded to patients' outcomes, using a predefined score. Multivariable logistic regression and supervised hierarchical classifiers methods were used to identify factors associated with poor outcome at 90 days, defined by a score of 3 to 6 (indicating moderate-to-severe disability or death) on the Modified Rankin Scale. Overall, 138 patients (median [interquartile range {IQR}] age, 62.6 [54.0-72.0] years; 75 men [54.3%]) with an admission median (IQR) Glasgow Coma Scale score of 9 (6-12) were studied. The median (IQR) delay between ICU admission and MRI was 1 (1-7) days. At 90 days, 95 patients (68.8%) had a poor outcome, including 16 deaths (11.6%). The presence of fluid-attenuated inversion recovery MRI signal abnormalities in more than 3 brain lobes (odds ratio [OR], 25.71; 95% CI, 1.21-554.42), age older than 60 years (OR, 7.62; 95% CI, 2.02-28.91), and the presence of diffusion-weighted MRI signal abnormalities in the left thalamus (OR, 6.90; 95% CI, 1.12-43.00) were independently associated with poor outcome. Machine learning models identified bilateral diffusion abnormalities as an additional factor associated with poor outcome (34 of 39 patients [87.2%] with bilateral abnormalities had poor outcomes) and confirmed the functional burden of left thalamic lesions, particularly in older patients (all 11 patients aged >60 years had left thalamic lesions). These findings suggest that in adult patients with HSE requiring ICU admission, extensive MRI changes in the brain are independently associated with poor functional outcome at 90 days. Thalamic diffusion signal changes were frequently observed and were associated with poor prognosis, mainly in older patients.

MRI contributes to accurate and early diagnosis of non-radiographic HLA-B27 negative axial spondyloarthritis

BackgroundNonradiographic axial spondyloarthropathies (nr-axSpA) are diagnosed by the absence of radiographic sacroiliitis and the presence of bone marrow edema (BME) on magnetic resonance imaging (MRI). According to the classification criteria of the international Assessment of Spondyloarthritis Society (ASAS), structural changes to sacroiliac joints (SIJs) on MRI cannot be used as criteria in the absence of BME. However, less than half the Asian patients with clinically active axSpA show BME. The incidence of human leukocyte antigen (HLA)-B27 is low in Asian populations, which makes it more difficult to identify nr-axSpA. We used MRI to evaluate the structural damage to SIJs in patients with nr-axSpA with and without BME with the aim of identifying the best methodology for accurate diagnosis, especially in populations with less common BME and HLA-B27.MethodsOne hundred three patients with inflammatory back pain were included in this prospective study. No patient’s radiograph met the definition of positive modified New York criteria. BME and structural damage to SIJ including sclerosis and erosion were assessed independently on coronal and axial short-tau inversion recovery and T1-weighted spin echo MRI scans by two well-trained musculoskeletal radiologists using the Spondyloarthritis Research Consortium of Canada (SPARCC) score. Demographics of patients were collected. Disease characteristics and structural damage were analyzed in patients with and without BME on SIJ MRI. Receiver operating characteristic (ROC) curve analysis was used to assess the diagnostic performance of structural damage.ResultsAll individuals in the cohort had at least one abnormal finding on SIJ MRI, including BME or structural damage; 36 of 103 patients had BME. We identified a significant positive correlation between SPARCC scores and severe erosion assessed by focal joint space widening (fJSW) (p = 0.001) in these 36 patients. Fifty-eight of the 103 enrolled patients fulfilled the ASAS criteria for nr-axSpA in the either absence or presence of BME. Of these 58 patients, 57 and 19 had erosions or fJSW, respectively, and the presence of BME was significantly correlated with fJSW (phi score of 0.319 and p = 0.015). We demonstrated a significant positive correlation between fJSW and either the presence or the severity of BME in patients with nr-axSpA who met the ASAS definition. There was a positive correlation between BME and fJSW across the whole study cohort (phi score of 0.389; p < 0.001). The area under the ROC curve (AUC) for fJSW on SIJ MRI was 0.736, p < 0.001. In both HLA-B27-positive and -negative groups, BME was more common in the presence of fJSW (phi scores of 0.370 and 0.377, p = 0.018 and 0.003, respectively) and SPARCC scores were higher in patients with fJSW (p < 0.001 and p = 0.005). We also identified a positive correlation between fJSW and BME in patients with nr-axSpA and normal serum levels of C-reactive protein (phi score of 0.362 and p = 0.001).ConclusionStructural damage detected on SIJ MRI, sclerosis, erosions and fJSW may be present in patients without detectable inflammation on SIJ MRI. However, fJSW is significantly correlated with the severity of inflammation seen on SIJ MRI, which contributes to the accurate diagnosis of nr-axSpA, and it could be used as an alternative diagnostic test for nr-axSpA in the general population, especially for those who do not carry the HLA-B27 gene, Asian patients without BME, or patients with normal serum inflammatory biomarkers.

White Matter Hyperintensities Are Associated With Severity of Essential Tremor in the Elderly.

Background: Essential tremor (ET) occurs with steeply increasing prevalence in the elderly, and apart from disease duration, age is independently associated with an increase of tremor amplitude and a decrease of frequency. White matter hyperintensities (WMHs) are a common finding in the elderly, and their role in the pathophysiology of ET is unknown. The aims of this study were to examine whether ET patients differ in their total or region-specific WMH volumes from healthy controls and to determine the impact of WMH on tremor characteristics.Methods: A total of 47 elderly ET patients with a mean age of 72 years and 39 age-matched healthy controls underwent a thorough clinical assessment and 3T MRI. Total WMH volumes were derived from T2-weighted fluid-attenuated inversion recovery (FLAIR) MR images. Additionally, region of interest-based WMH volumes for the Johns Hopkins University (JHU) white matter tracts and labels were calculated, and WMHs were assessed semiquantitatively using the Fazekas scale.Results: Essential tremor patients and healthy controls did not differ in their total or tract-specific WMH volumes or Fazekas scores. However, WMH volume was significantly positively correlated with tremor severity on the TETRAS scale, and there was a significant negative correlation with the mean accelerometric tremor frequency. In a multiple linear regression model including disease duration, age, and age-adjusted total WMH volume, only the WMH volume significantly predicted tremor severity, while age and disease duration were not significant.Conclusion: We found evidence for a direct association between WMH volume and tremor severity. If confirmed by larger studies, our findings could explain the well-known relation between age and tremor severity.

Effect of white matter hyperintensity on dopamine transporter availability of striatum measured by F-18 FP-CIT PET.

We aimed to evaluate the white matter hyperintensity (WMH) effect on dopamine transporter availability (DAT) of striatum. A total of 48 patients who showed visually normal F-18 FP-CIT uptake were included in this study. Each FP-CIT image were pre-processed using SPM12. Co-registration and spatial normalization of FP-CIT image conducted using T1-weighted magnetic resonance imaging (MRI). And then smoothing of normalized FP-CIT image was performed. Intensity normalization was performed using cerebellum as a reference region. With pre-defined volume of interest template, the specific binding ratio (SBR) of both side of caudate nucleus and putamen was calculated. Fluid attenuated inversion recovery MRI scans were used to evaluate WMH number and volume. SBRs of left and right caudate nucleus were correlated with age (r = -0.615; p < 0.0001; n = 48, r = -0.607; p < 0.0001; n = 48, respectively), high density lipoprotein cholesterol (r = 0.296; p = 0.041; n = 48, r = 0.29; p = 0.0455; n = 48, respectively), and WMH number (r = -0.459; p = 0.001; n = 48, r = -0.481; p = 0.0005; n = 48, respectively) and volume (r = -0.407; p = 0.0041; n = 48, r = -0.428; p = 0.0024; n = 48, respectively). DAT availability of patients who showed visually normal F-18 FP-CIT uptake was correlated with number and volume of WMH.

Predicting 1p/19q chromosomal deletion of brain tumors using machine learning

Advances in molecular and genetic technologies have enabled the study of mutation and molecular changes in gliomas. The 1p/19q coding state of gliomas is important in predicting pathogenesis-based pharmacological treatments and determining innovative immunotherapeutic strategies. In this study, T1-weighted and T2-weighted fluid-attenuated inversion recovery magnetic resonance imaging (MRI) images of 121 low-grade glioma patients with biopsy-proven 1p/19q coding status and no deletion (n = 40) or co-deletion (n = 81) were used. First, regions of interests were segmented with the grow-cut algorithm. Later, 851 radiomic features including three-dimensional wavelet preprocessed and non-preprocessed ones were extracted from six different matrices such as first order, shape and texture. The extracted features were preprocessed with the synthetic minority over-sampling technique algorithm. Next, the 1p/19q decoding states of gliomas were classified using machine-learning algorithms. The best classification in the classification of glioma grades (grade II and grade III) according to 1p/19q coding status was obtained by using the logistic regression algorithm, with 93.94% accuracy and 94.74% area under the curve values. In conclusion, it was determined that non-invasive estimation of 1p/19q status from MRI images enables the selection of effective treatment strategies with early diagnosis using machine-learning algorithms without the need for surgical biopsy.

Semi-Supervised Learning in Medical MRI Segmentation: Brain Tissue with White Matter Hyperintensity Segmentation Using FLAIR MRI.

White-matter hyperintensity (WMH) is a primary biomarker for small-vessel cerebrovascular disease, Alzheimer’s disease (AD), and others. The association of WMH with brain structural changes has also recently been reported. Although fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) provide valuable information about WMH, FLAIR does not provide other normal tissue information. The multi-modal analysis of FLAIR and T1-weighted (T1w) MRI is thus desirable for WMH-related brain aging studies. In clinical settings, however, FLAIR is often the only available modality. In this study, we thus propose a semi-supervised learning method for full brain segmentation using FLAIR. The results of our proposed method were compared with the reference labels, which were obtained by FreeSurfer segmentation on T1w MRI. The relative volume difference between the two sets of results shows that our proposed method has high reliability. We further evaluated our proposed WMH segmentation by comparing the Dice similarity coefficients of the reference and the results of our proposed method. We believe our semi-supervised learning method has a great potential for use for other MRI sequences and will encourage others to perform brain tissue segmentation using MRI modalities other than T1w.

Detection of multiple sclerosis lesions in the cervical cord: which of the MAGNIMS 'mandatory' non-gadolinium enhanced sagittal sequences is optimal at 3T?

The magnetic resonance imaging in multiple sclerosis consensus guidelines currently mandate three sagittal non-contrast enhanced sequences of T2-weighted fast spin echo, proton density-weighted fast spin echo and short tau inversion recovery; however, these particular three sequences have not previously been compared at 3T. This study compared T2-weighted fast spin echo, proton density-weighted fast spin echo, short tau inversion recovery as well as the double inversion recovery sequence for the sagittal detection of multiple sclerosis lesions in the cervical spinal cord at 3T. Nineteen multiple sclerosis patients underwent magnetic resonance imaging with 3T sagittal T2-weighted fast spin echo, proton density-weighted fast spin echo, short tau inversion recovery and double inversion recovery between November 2012 and April 2013. Two neuroradiologists independently reviewed the images, and the number of lesions detected on each sequence was recorded. Lesion conspicuity was quantitatively assessed with the lesion-to-cord-contrast ratio and lesion contrast-to-noise ratio. The Wilcoxon signed rank test was performed for statistical analysis. Proton density-weighted fast spin echo and short tau inversion recovery detected 32% more lesions compared to T2-weighted fast spin echo, and 37% more lesions compared to double inversion recovery. The lesion-to-cord-contrast ratio was highest in short tau inversion recovery, while the lesion contrast-to-noise ratio was highest for proton density-weighted fast spin echo. This study provides the necessary evidentiary support at 3T for the magnetic resonance imaging in multiple sclerosis spinal magnetic resonance imaging protocol consensus guidelines. At 3T sagittal proton density-weighted fast spin echo and short tau inversion recovery sequences allowed improved detection of cervical spinal cord multiple sclerosis lesions, compared to T2-weighted fast spin echo and three-dimensional double inversion recovery magnetic resonance imaging. Utilising T2-weighted fast spin echo alone at 3T is insufficient for lesion detection.

MRI-based radiomics signature for the prediction of response of lung cancer brain metastases after whole-brain radiotherapy.

e21096 MRI-based Radiomics signature for the Prediction of Response of Lung Cancer Brain Metastases After Whole-Brain Radiotherapy Background: Local response prediction for brain metastases (BMs) from lung cancer after Whole-Brain Radiotherapy (WBRT) is challenging, as existing criteria are based solely on unidimensional measurements. This study sought to determine whether radiomic features of lung cancer BMs derived from pre-treatment magnetic resonance imaging (MRI) could be used to predict local response following WBRT. Methods: A total of 88 Lung Cancer patients with BMs treated with WBRT were analyzed. After volumes of interest were drawn, 944 radiomic features including first-order, shape, Gray Level Co-occurrence Matrix (GLCM), Gray Level Dependence Matrix (GLDM), Gray Level Run Length Matrix (GLRLM), Gray Level Size Zone Matrix (GLSZM), Neighborhood Gray Tone Difference Matrix (NGTDM), and Laplacian of Gaussian (LoG) features were extracted, using the baseline pre-treatment post-contrast T1 (T1c) and T2 fluid-attenuated inversion recovery (FLAIR) MRI sequences, respectively. Local response status was determined by contrasting the baseline and follow-up MRI according to the RANO-BM criteria. The independent samples t test or Mann-Whitney U test, and then least absolute shrinkage and selection operator (LASSO) were used for dimensionality reduction and feature selection. An adaboost classifier was trained using the selected radiomic features and evaluated using the area under the receiver operating characteristic curve (AUC) in both the training and testing sets. Other discrimination metrics, including classification accuracy, positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity, were also calculated. Results: The optimal radiomics signature was developed based on a multivariable logistic regression with 4, 5, 6 radiomic features on T1c, T2 FLAIR and T1c+T2 FLAIR, respectively. The radiomics model based on T1c features presented the AUC of (0.920 vs. 0.805, respectively) for both the training and testing sets, followed by T2 FLAIR features (0.893 vs. 0.701, respectively), and T1c+T2 FLAIR features (0.971 vs. 0.857, respectively). The classification accuracy of the radiomics model also well predicted the local response of BMs for both the the training and testing sets (T1c: 82.9% vs. 77.8%, T2 FLAIR: 82.9% vs. 77.8%, T1c+T2 FLAIR: 90.0% vs. 77.8%, respectively). Conclusions: Radiomics holds promise for predicting local tumor response following WBRT in patients with lung cancer and brain metastases. A predictive model built on radiomic features from an institutional cohort performed well on cross-validation testing. These results warrant further validation in independent datasets. Such work could prove invaluable for guiding management of individual patients and assessing outcomes of novel interventions.

3D-Fast Gray Matter Acquisition with Phase Sensitive Inversion Recovery Magnetic Resonance Imaging at 3 Tesla: Application for detection of spinal cord lesions in patients with multiple sclerosis

To compare 3D-Fast Gray Matter Acquisition with Phase Sensitive Inversion Recovery (3D-FGAPSIR) with conventional 3D-Short-Tau Inversion Recovery (3D-STIR) and sagittal T1-and T2-weighted MRI dataset at 3 Tesla when detecting MS spinal cord lesions. This prospective single-center study was approved by an institutional review board and enrolled participants from December 2016 to August 2018. Two neuroradiologists blinded to all data, individually analyzed the 3D-FGAPSIR and the conventional datasets separately and in random order. Discrepancies were resolved by consensus by a third neuroradiologist. The primary judgment criterion was the number of MS spinal cord lesions. Secondary judgment criteria included lesion enhancement, lesion delineation, reader-reported confidence and lesion-to-cord-contrast-ratio. A Wilcoxon's test was used to compare the two datasets. 51 participants were included. 3D-FGAPSIR detected significantly more lesions than the conventional dataset (344 versus 171 respectively, p<0.001). Two participants had no detected lesion on the conventional dataset, whereas 3D-FGAPSIR detected at least one lesion. 3/51 participants had a single enhancing lesion detected by both datasets. Lesion delineation and reader-reported confidence were significantly higher with 3D-FGAPSIR: 4.5 (IQR 1) versus 2 (IQR 0.5), p<0.0001 and 4.5 (IQR 1) versus 2.5 (IQR 0.5), p<0.0001. Lesion-to-cord-contrast-ratio was significantly higher using 3D-FGAPSIR as opposed to 3D-STIR and T2: 1.4 (IQR 0,3) versus 0.4 (IQR 0,1) and 0.3 (IQR 0,1)(p = 0.04). Correlations with clinical data and inter- and intra-observer agreements were higher with 3D-FGAPSIR. 3D-FGAPSIR improved overall MS spinal cord lesion detection as compared to conventional set and detected all enhancing lesions.

Intracranial volume segmentation for neurodegenerative populations using multicentre FLAIR MRI

Intracranial volume (ICV) segmentation, also known as brain extraction or skull-stripping, is a critical preprocessing step in analytical pipelines for studying neurodegenerative diseases in magnetic resonance imaging (MRI). While the fluid-attenuated inversion recovery (FLAIR) MRI modality has emerged as an important sequence for analyzing cerebrovascular and neurodegenerative disease, most existing automated ICV segmentation methods have been developed for T1-weighted or multi-modal inputs. Additionally, many methods have been designed using single centre data of healthy subjects and encounter difficulties using images with varying acquisition parameters and neurodegenerative pathology. In this work, we develop and evaluate 2 traditional and 8 deep learning algorithms for ICV segmentation in FLAIR MRI. Training and testing were completed on 175 ​vol (8317 images) from 2 dementia and 1 vascular disease cohort. A human phantom FLAIR MRI dataset from a repeatedly scanned, healthy individual was also utilized for reliability analysis. Images were acquired from 47 imaging centres with varying scanners and parameters. To measure and compare performance, we present a novel framework for evaluating the effectiveness of computer generated segmentations on multicentre datasets. The evaluation framework includes assessments of algorithm accuracy, generalization capabilities, robustness to pathology and spatial location, and volumetric measurement reliability – all important dimensions for establishing proof of effectiveness (a prerequisite to clinical translation). The top performing method was a multiple resolution U-Net (MultiResUNet), which achieved a mean Dice similarity coefficient greater than 98% and was robust across pathology levels and spatial locations. Our results confirm a FLAIR-based ICV analytical pipeline can alone be utilized for large-scale neurodegenerative disease research. The presented evaluation framework can be deployed by other researchers to assess the viability of tools proposed for automated analysis of diverse, clinical MRI datasets.