AJNR. American journal of neuroradiology最新文献

Background and purpose: Identifying patients with MS at higher risk of clinical progression is essential to inform clinical management. We aimed to build prognostic models by using machine learning (ML) algorithms predicting long-term clinical outcomes based on a systematic mapping of volumetric, radiomic, and macrostructural disconnection features from routine brain MRI scans of patients with MS.

Materials and methods: In this longitudinal monocentric study, 3T structural MRI scans of patients with MS were retrospectively analyzed. Based on a 10-year clinical follow-up (average duration = 9.4 ± 1.1 years), patients were classified according to confirmed disability progression (CDP) and cognitive impairment (CI) as assessed through the Expanded Disability Status Scale and the Brief International Cognitive Assessment of Multiple Sclerosis battery, respectively. Three-dimensional T1-weighted and FLAIR images were automatically segmented to obtain volumes, disconnection scores (estimated based on lesion masks and normative tractography data), and radiomic features from 116 GM regions defined according to the automated anatomic labeling atlas. Three ML algorithms (Extra Trees, Logistic Regression, and Support Vector Machine) were used to build models predicting long-term CDP and CI based on MRI-derived features. Feature selection was performed on the training set with a multistep process, and models were validated with a holdout approach, randomly splitting the patients into training (75%) and test (25%) sets.

Results: We studied 177 patients with MS (men/women = 51/126; mean ± standard deviation age: 35.2 ± 8.7 years). Long-term CDP and CI were observed in 71 and 55 patients, respectively. Regarding the CDP class prediction analysis, the feature selection identified 13-, 12-, and 10-feature subsets obtaining an accuracy on the test set of 0.71, 0.69, and 0.67 for the Extra Trees, Logistic Regression, and Support Vector Machine classifiers, respectively. Similarly, for the CI prediction, subsets of 16, 17, and 19 features were selected, with 0.69, 0.64, and 0.62 accuracy values on the test set, respectively. There were no significant differences in accuracy between ML models for CDP (P = .65) or CI (P = .31).

Conclusions: Building on quantitative features derived from conventional MRI scans, we obtained long-term prognostic models, potentially informing patients' stratification and clinical decision-making.

Background and purpose: CT perfusion is widely used to assess infarct core and penumbra in acute stroke, but scan durations vary and may be affected by patient-specific delays in contrast arrival. Our purpose was assess the impact of radiologic and clinical variables on brain CTP curves in patients with acute ischemic stroke.

Materials and methods: We included 295 patients who underwent CTP for acute ischemic stroke in our institution (January 2020 to March 2024). Two radiologists evaluated arterial input function and reference vessel curves to assess bolus arrival delay and time to equilibrium; discrepancies were resolved by consensus. Additionally, they evaluated the unenhanced brain CTs acquired before CTP for the presence of microangiopathy (van Swieten scale) and intracranial arterial wall calcifications (yes/no). CTA was evaluated for the site of occlusion. Age, sex, arterial blood pressure, heart rate, presence of arrhythmias, and NIHSS were retrieved from an institutional database. A univariate analysis was performed to establish significant variables; variables with a P value < .1 in the univariate analysis were subsequently included in a multivariate logistic regression model to adjust for potential confounding factors.

Results: Logistic regression identified cardiac arrhythmias and increasing age as independent predictors of nondiagnostic perfusion CT examinations (P < .001). Other factors, including arterial calcifications, white matter lesions, NIHSS score, and large vessel occlusion, were not significantly associated with nondiagnostic outcomes. Logistic regression analysis revealed that the arterial time-to-peak value was significantly associated with the presence of cardiac arrhythmias (P < .0001), with higher time-to-peak values observed among patients with arrhythmias (24.0 seconds; interquartile range [IQR]: 20.2-27.1 seconds) compared with those without (18.6 seconds; IQR: 15.5-21.7 seconds). Similarly, the venous time-to-peak was found to be longer in patients with cardiac arrhythmias (median 30.2 seconds; IQR: 26.4-32.0 seconds) compared with those without (25.6 seconds; IQR: 22.5-28.7 seconds), P < .0001.

Conclusions: Our study showed that patients with cardiac arrhythmias need longer CTP acquisition times to avoid perfusion curve truncation and potentially nondiagnostic results. The knowledge of the impact of clinical variables on CTP may help better tailor the acquisition delays to improve diagnostic quality and avoid unnecessary radiation doses.

Background and purpose: Chronic multisymptom illness (CMI) includes symptoms of fatigue, pain, and sleep difficulties, as well as neurologic, respiratory, and gastrointestinal problems and is particularly common in veterans from the 1990-1991 Gulf War and the Afghanistan and Iraq Wars. Glymphatic system function may play an important role in the etiopathology of CMI but has not been addressed. DTI-derived analysis along the perivascular space provides a promising proxy for glymphatic system function by evaluating the status of perivascular space fluid flow. The objective of this study was to compare this DTI-derived glymphatic index in veterans with CMI and healthy controls, and to reveal possible correlations between this index and the severity of CMI symptoms.

Materials and methods: DTI-derived indices were extracted from imaging data of 203 veterans who met clinical diagnostic criteria for CMI, and 224 age-matched healthy control subjects from multiple public research databases. Severity of CMI, sleep difficulty, pain intensity, and the degree of chronic fatigue were based on self-report measures. MRI scanner and site variations were harmonized. Statistical analyses were performed adjusting for demographic confounding factors.

Results: Both healthy controls and veterans showed significantly reduced glymphatic indices associated with increased age. Compared with controls, veterans showed bilaterally lower indices (Cohen d = -0.47; P < .001) after adjusting for age, sex, and education. Across the entire sample of veterans, negative correlations were observed between glymphatic indices and pain intensities (r = -0.17; P = .01), sleep disturbances (r = -0.17; P = 0.02), degree of fatigue (r = -0.20; P = 0.006), severity of CMI (r = -0.17; P = 0.02), and the indices were positively correlated with medullar volumes (r = -0.19; P = .007). Note, these results showing significant outcomes for a group of patients do not guarantee the same outcome for individual patients.

Conclusions: This study suggests that impaired glymphatic functions are strongly associated with CMI. These findings improve our understanding of the pathologic mechanism underlying CMI and point to DTI-based metrics as a potential biomarker for disease severity in this condition.

Infants in the neonatal intensive care unit (NICU) can face barriers to access MRI of the brain including presence of assistive devices that are not MRI safe or conditional and staff resources needed to transport sick infants. An ultra-low-field (0.064T) portable MRI system has the ability to improve access by allowing infants to be scanned directly in the NICU eliminating the need to transport infants to the MRI scanner. However, the scanning parameters for ultra-low-field portable MRI were optimized for adult patient use and not for infants. This article describes an optimized scanning protocol for infant brain imaging by using ultra-low-field portable MRI.

Background and purpose: The relationship between digital ¹⁸F-FDG-PET findings and glucose metabolism-related genetic alterations remains unclear in primary CNS lymphoma (PCNSL). This study aimed to evaluate whether digital FDG-PET can serve as a noninvasive tool to detect MYD88 mutation-driven glycolytic activity in PCNSL.

Materials and methods: We retrospectively analyzed the imaging and molecular data of 54 patients with PCNSL (55 lesions). MRI and FDG-PET parameters, including the maximum standardized uptake value (SUVmax) and tumor-to-background ratio (TBR), were assessed. Tumor specimens were subjected to histopathologic and genomic evaluations, including the MYD88 mutation status.

Results: Among 55 tumors, 34 (61.8%) were examined with digital FDG-PET and 21 (38.2%) with analog ¹⁸F-FDG-PET. In the digital FDG-PET group, MYD88-mutant tumors showed statistically higher SUVmax (30.2 ± 9.9) and TBR (6.1 ± 1.5) compared with wild-type tumors (SUVmax: 19.3 ± 7.2, P = .006; TBR: 3.5 ± 1.3, P < .001). In the analog FDG-PET group, the SUVmax was higher in MYD88-mutant tumors (P = .01), whereas the TBR differences were not statistically significant (P = .38). Receiver operating characteristic analysis of TBR in digital FDG-PET yielded an area under the curve of 0.913 (95% CI, 0.954-1.000) with a cutoff value of 4.49, achieving 88% sensitivity and 88% specificity for MYD88 mutation detection. Multivariate logistic regression identified SUVmax and TBR from digital FDG-PET as independent predictors of MYD88 mutation status. The transcriptomic analysis confirmed the up-regulation of glycolysis-related genes, including hexokinase 2, in MYD88-mutant tumors, supporting increased glycolytic activity.

Conclusions: Digital FDG-PET may serve as a valuable noninvasive imaging technique to detect MYD88 mutation-driven enhanced glycolysis in patients with PCNSL.

Background and purpose: Intra-arterial chemotherapy (IAC) is a potent treatment for patients with pediatric retinoblastoma (RB). Ophthalmic artery (OA) navigation via the ICA can be technically challenging; and in cases with robust anastomoses to the OA from the middle meningeal artery (MMA), the latter may serve as an alternative route. Our study evaluates the ability of preprocedural MRA to identify OA variants and the presence of MMA-to-OA/lacrimal artery (LA) anastomoses to enhance successful delivery of IAC in this population.

Materials and methods: Consecutive patients with RB admitted for IAC treatment at a tertiary pediatric hospital were identified from October 2018 to January 2025. The morphology of the OA origin and the presence of MMA-OA anastomoses were assessed on preprocedural MRA and confirmed on DSA.

Results: A total of 13 patients who underwent 42 procedures were included, with a median age of 11 months (interquartile range, 11-21 months). The OA origin was visualized on MRA in 12 (92%) cases. MMA-OA/LA anastomoses were seen on MRA for 6 (46%) patients, all confirmed on DSA. Among patients with no MRA evidence of anastomoses, none were found to have anastomoses on DSA. MMA-OA anastomoses were used as the primary delivery route of IAC for 3 patients (50% of patients with a confirmed presence of MMA-OA anastomoses). Overall procedural success was 98% (41/42).

Conclusions: Preoperative MRA is an accurate tool for identifying OA origin variants and MMA-OA anastomoses, which may assist in optimizing IAC delivery for patients with pediatric RB, while minimizing radiation.

Background and purpose: Sphingosine-1-phosphate receptor 1 (S1PR1) is a key regulator of neuroinflammation and plays a crucial role in multiple neurodegenerative diseases. [¹¹C]CS1P1 is a novel PET tracer for measuring expression levels of S1PR1 in humans. Before widespread application, its quantification must be established and evaluated in healthy young and old adults through characterization of binding topographies, kinetics, and tracer metabolism rates.

Materials and methods: We acquired dynamic [¹¹C]CS1P1 emission data from 29 healthy controls and investigated the topography of [¹¹C]CS1P1 uptake, radio-labeled metabolites of the tracer, an image-derived input function estimation, and tissue compartment modeling.

Results: The image-derived input function approximated the arterially sampled input function. Further, radio-labeled metabolites of the tracer accumulated linearly throughout the scan and demonstrated consistency across participants. A 2-tissue compartment model fitted the observed emission data well, consistent with previously reported nonhuman primate studies. Kinetic modeling using the image-derived input functions, corrected by population estimates of tracer metabolism, provided a good fit for tissue activity curves. Graphical Logan analysis reliably estimated volume of distribution (Vt), and Vt closely reproduced S1PR1 distribution in the brain.

Conclusions: In this study, we have established a quantitative ¹¹C]CS1P1 PET processing approach by using a 2-tissue compartment model and imaging-derived input function with population metabolite correction. [¹¹C]CS1P1 PET reflects S1PR1 topography and supports its use for investigating neuroinflammation in humans.

Background and purpose: Neonatal brachial plexus imaging faces challenges with extended scan times and motion artifacts. This study assessed whether compressed sensitivity encoding (CS) acceleration could achieve image quality comparable with conventional sensitivity encoding (SENSE) while significantly reducing scanning time, potentially enhancing diagnostic accuracy and success rates in neonatal brachial plexus imaging.

Materials and methods: Forty-five neonates (18 male, 27 female; mean age 14.82 ± 9.62 days) with clinical suspicion of brachial plexus nerve injury were examined by using both CS and SENSE 3D NerveVIEW sequences on a 3T MRI scanner. The parallel acquisition acceleration factor was 1.3 for SENSE and 6 for CS. Image quality was evaluated quantitatively by using SNR and nerve-to-muscle contrast-to-noise ratio (CNR), and qualitatively through a 5-point semiquantitative scale assessment by 2 senior pediatric radiologists.

Results: CS reduced acquisition time by approximately 30% (3:36 versus 5:08 minutes) compared with SENSE, without compromising image quality. No significant differences were found in SNR and nerve-to-muscle CNR between CS and SENSE, with equivalence testing confirming comparable image quality (SNR: t(44) = 3.109, P = .002; nerve-to-muscle CNR: t(44) = 1.984, P = .03). Radiologists' subjective evaluations revealed no significant difference in image quality scores between CS and SENSE, with strong interrater agreement for both methods (CS: κ = 0.773; SENSE: κ = 0.617).

Conclusions: Implementation of CS technology in 3D NerveVIEW sequences for neonatal brachial plexus imaging is feasible and effective, providing image quality comparable to SENSE while significantly reducing scanning time. This advancement potentially improves patient outcomes through higher success rates in imaging examinations.

Background and purpose: Radiation therapy (RT), a primary treatment for head and neck cancer (HNC), increases the risk of radiation-induced carotid stenosis (RICS). This study examines the progression of carotid artery total occlusion (CATO) in patients with HNC after RT, focusing on low-density plaque (LDP) as a predictor, an aspect underexplored in prior research.

Materials and methods: A retrospective cohort study assessed 44 patients with HNC who underwent RT, using 2021 follow-up data. Carotid stenosis progression was quantified via CT scans before and after LDP detection. Patients with irregular follow-up or RT within 2 years, insufficient for long-term evaluation, were excluded.

Results: CATO occurred in 11 patients (25%) and 12 vessels (14.3%), with a median onset of 12.6 years (interquartile range [IQR], 7.6-22.5) after RT. The stenosis progression rate increased significantly from 0.00%/year (IQR, 0.0%-1.0%) before LDP detection to 4.7%/year (IQR, 3.4%-8.1%) afterward (P < .001), with a marked acceleration to 14.8%/year (IQR, 9.6%-24.0%) in the CATO group. Neither calcification nor age at RT was significantly associated with CATO, reinforcing LDP as a critical high-risk marker.

Conclusions: This study identifies LDP as a critical predictor of accelerated RICS, proposing a novel biphasic progression model with a slow phase before LDP detection and a rapid phase thereafter. Calcification showed no significant association with RICS progression. These findings support intensified surveillance, such as annual CT scans, and timely interventions to prevent occlusion in high-risk patients.

Background and purpose: Spinal cord (SC) lesions are critical in MS diagnosis and progression, yet their detection remains challenging. Conventional sequences such as 2D T2-weighted FSE and STIR have suboptimal sensitivity, both in the cervical and thoracic spine. This study evaluated the diagnostic performance of a new imaging technique for SC lesion detection, the 3D white-matter-nulled (WMn) MPRAGE sequence, compared with conventional MRI sequences.

Materials and methods: Thirty-eight patients with MS or clinically isolated syndrome were prospectively evaluated with 3T SC MRI, acquiring 2D T2-weighted FSE, 2D STIR, 3D MPRAGE, and 3D WMn. A deep learning denoising method was applied to 3D WMn to compensate the inherent low signal. Four blinded neuroradiologists independently assessed lesion count, confidence in lesion detection, and image quality (eg, artifacts). Contrast-to-noise ratio (CNR) was also computed for all lesions detected on all sequences. Statistical comparisons were performed across sequences.

Results: In the cervicothoracic spine, the 3D WMn sequence detected significantly more lesions than 2D T2-weighted FSE (+62%; P < .001), STIR (+47%; P < .05), and 3D MPRAGE (+50%; P < .01). In the thoracolumbar spine, the 3D WMn sequence detected significantly more lesions than 2D T2-weighted FSE (+53%, P < .05). The 3D WMn sequence demonstrated a higher CNR and improved lesion conspicuity while exhibiting fewer artifacts than STIR. These advantages contributed to greater rater confidence in lesion detection by using the 3D WMn sequence, with over 75% of strong confidence reported by the 2 most experienced raters, along with the highest interreader agreement.

Conclusions: The 3D WMn sequence associated with deep learning-based denoising significantly improves SC lesion detection, outperforming all conventional sequences. Three-dimensional WMn represents, then, a promising alternative for SC imaging in MS.