Artificial intelligence for volumetric measurement of cerebral white matter hyperintensities on thick-slice fluid-attenuated inversion recovery (FLAIR) magnetic resonance images from multiple centers.

We aimed to develop a new artificial intelligence software that can automatically extract and measure the volume of white matter hyperintensities (WMHs) in head magnetic resonance imaging (MRI) using only thick-slice fluid-attenuated inversion recovery (FLAIR) sequences from multiple centers. We enrolled 1092 participants in Japan, comprising the thick-slice Private Dataset. Based on 207 randomly selected participants, neuroradiologists annotated WMHs using predefined guidelines. The annotated images of participants were divided into training (n = 138) and test (n = 69) datasets. The WMH segmentation model comprised a U-Net ensemble and was trained using the Private Dataset. Two other models were trained for validation using either both thin- and thick-slice MRI datasets or the thin-slice dataset alone. The voxel-wise Dice similarity coefficient (DSC) was used as the evaluation metric. The model trained using only thick-slice MRI showed a DSC of 0.820 for the test dataset, which is comparable to the accuracy of human readers. The model trained with the additional thin-slice dataset showed only a slightly improved DSC of 0.822. This automatic WMH segmentation model comprising a U-Net ensemble trained on a thick-slice FLAIR MRI dataset is a promising new method. Despite some limitations, this model may be applicable in clinical practice.

Performance changes due to differences among annotating radiologists for training data in computerized lesion detection.

PURPOSE: The quality and bias of annotations by annotators (e.g., radiologists) affect the performance changes in computer-aided detection (CAD) software using machine learning. We hypothesized that the difference in the years of experience in image interpretation among radiologists contributes to annotation variability. In this study, we focused on how the performance of CAD software changes with retraining by incorporating cases annotated by radiologists with varying experience. METHODS: We used two types of CAD software for lung nodule detection in chest computed tomography images and cerebral aneurysm detection in magnetic resonance angiography images. Twelve radiologists with different years of experience independently annotated the lesions, and the performance changes were investigated by repeating the retraining of the CAD software twice, with the addition of cases annotated by each radiologist. Additionally, we investigated the effects of retraining using integrated annotations from multiple radiologists. RESULTS: The performance of the CAD software after retraining differed among annotating radiologists. In some cases, the performance was degraded compared to that of the initial software. Retraining using integrated annotations showed different performance trends depending on the target CAD software, notably in cerebral aneurysm detection, where the performance decreased compared to using annotations from a single radiologist. CONCLUSIONS: Although the performance of the CAD software after retraining varied among the annotating radiologists, no direct correlation with their experience was found. The performance trends differed according to the type of CAD software used when integrated annotations from multiple radiologists were used.

Axillary Lymphadenopathy after Pfizer-BioNTech and Moderna COVID-19 Vaccination: MRI Evaluation.

Background COVID-19 vaccination-related axillary lymphadenopathy has become an important problem in cancer imaging. Data are needed to update or support imaging guidelines for conducting appropriate follow-up. Purpose To investigate the prevalence, predisposing factors, and MRI characteristics of COVID-19 vaccination-related axillary lymphadenopathy. Materials and Methods Prospectively collected prevaccination and postvaccination chest MRI scans were secondarily analyzed. Participants who underwent two doses of either the Pfizer-BioNTech or Moderna COVID-19 vaccine and chest MRI from June to October 2021 were included. Enlarged axillary lymph nodes were identified on postvaccination MRI scans compared with prevaccination scans. The lymph node diameter, signal intensity with T2-weighted imaging, and apparent diffusion coefficient (ADC) of the largest enlarged lymph nodes were measured. These values were compared between prevaccination and postvaccination MRI by using the Wilcoxon signed-rank test. Results Overall, 433 participants (mean age, 65 years ± 11 [SD]; 300 men and 133 women) were included. The prevalence of axillary lymphadenopathy in participants 1-14 days after vaccination was 65% (30 of 46). Participants with lymphadenopathy were younger than those without lymphadenopathy (P < .001). Female sex and the Moderna vaccine were predisposing factors (P = .005 and P = .003, respectively). Five or more enlarged lymph nodes were noted in 2% (eight of 433) of participants. Enlarged lymph nodes greater than or equal to 10 mm in the short axis were noted in 1% (four of 433) of participants. The median signal intensity relative to the muscle on T2-weighted images was 4.0; enlarged lymph nodes demonstrated a higher signal intensity (P = .002). The median ADC of enlarged lymph nodes after vaccination in 90 participants was 1.1 × 10-3 mm2/sec (range, 0.6-2.0 × 10-3 mm2/sec), thus ADC values remained normal. Conclusion Axillary lymphadenopathy after the second dose of the Pfizer-BioNTech or Moderna COVID-19 vaccines was frequent within 2 weeks after vaccination, was typically less than 10 mm in size, and had a normal apparent diffusion coefficient. © RSNA, 2022.

Successful transcatheter arterial embolization for pseudoaneurysm of the deep femoral artery in a patient with presumptive ACTA2-related vasculopathy.

ACTA2-related vasculopathy is an autosomal dominant genetic disorder characterized by aortic aneurysms and dissection, and limb artery lesions are rare. We report a case of transcatheter arterial embolization for a pseudoaneurysm of a deep femoral artery in a patient with presumptive ACTA2-related vasculopathy. A 58-year-old woman was presumed to have an ACTA2 mutation based on her history of aortic diseases and family history of ACTA2 mutations. During follow-up, contrast-enhanced computed tomography for aortic diseases revealed occlusion and vessel wall abnormalities of the bilateral deep femoral arteries. Two weeks later, she complained of acute right inguinal pain without any triggering factors, and contrast-enhanced computed tomography revealed a pseudoaneurysm of the right deep femoral artery. Vascular fragility due to ACTA2 mutation was believed to be the cause of the pseudoaneurysm. Transcatheter arterial embolization was successfully performed and no rebleeding occurred during 1.5 years after the transcatheter arterial embolization.

MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction.

BACKGROUND: Unsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects. Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss. However, without considering continuity between multiple adjacent slices, they cannot directly discriminate diseases composed of the accumulation of subtle anatomical anomalies, such as Alzheimer's disease (AD). Moreover, no study has shown how unsupervised anomaly detection is associated with either disease stages, various (i.e., more than two types of) diseases, or multi-sequence magnetic resonance imaging (MRI) scans. RESULTS: We propose unsupervised medical anomaly detection generative adversarial network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect brain anomalies at different stages on multi-sequence structural MRI: (Reconstruction) Wasserstein loss with Gradient Penalty + 100 [Formula: see text] loss-trained on 3 healthy brain axial MRI slices to reconstruct the next 3 ones-reconstructs unseen healthy/abnormal scans; (Diagnosis) Average [Formula: see text] loss per scan discriminates them, comparing the ground truth/reconstructed slices. For training, we use two different datasets composed of 1133 healthy T1-weighted (T1) and 135 healthy contrast-enhanced T1 (T1c) brain MRI scans for detecting AD and brain metastases/various diseases, respectively. Our self-attention MADGAN can detect AD on T1 scans at a very early stage, mild cognitive impairment (MCI), with area under the curve (AUC) 0.727, and AD at a late stage with AUC 0.894, while detecting brain metastases on T1c scans with AUC 0.921. CONCLUSIONS: Similar to physicians' way of performing a diagnosis, using massive healthy training data, our first multiple MRI slice reconstruction approach, MADGAN, can reliably predict the next 3 slices from the previous 3 ones only for unseen healthy images. As the first unsupervised various disease diagnosis, MADGAN can reliably detect the accumulation of subtle anatomical anomalies and hyper-intense enhancing lesions, such as (especially late-stage) AD and brain metastases on multi-sequence MRI scans.

Differentiation of high-grade and low-grade intra-axial brain tumors by time-dependent diffusion MRI.

INTRODUCTION: Oscillating gradient spin-echo (OGSE) sequences enable acquisitions with shorter diffusion times. There is growing interest in the effect of diffusion time on apparent diffusion coefficient (ADC) values in patients with cancer. However, little evidence exists regarding its usefulness for differentiating between high-grade and low-grade brain tumors. The purpose of this study is to investigate the utility of changes in the ADC value between short and long diffusion times in distinguishing low-grade and high-grade brain tumors. MATERIAL AND METHODS: Eleven patients with high-grade brain tumors and ten patients with low-grade brain tumors were scanned using a 3 T magnetic resonance imaging with diffusion-weighted imaging (DWI) using OGSE and PGSE (effective diffusion time [Δeff]: 6.5 ms and 35.2 ms) and b-values of 0 and 1000 s/mm2. Using a region of interest (ROI) analysis of the brain tumors, we measured the ADC for two Δeff (ADCΔeff) values and computed the subtraction ADC (ΔADC = ADC6.5 ms - ADC35.2 ms) and the relative ADC (ΔADC = (ADC6.5 ms - ADC35.2 ms) / ADC35.2 ms × 100). The maximum values for the subtraction ADC (ΔADCmax) and the relative ADC (rADCmax) on the ROI were compared between low-grade and high-grade tumors using the Wilcoxon rank-sum test. A P-value <.05 was considered significant. The ROIs were also placed in the normal white matter of patients with high- and low-grade brain tumors, and ΔADCmax values were determined. RESULTS: High-grade tumors had significantly higher ΔADCmax and rADCmax than low-grade tumors. The ΔADCmax values of the normal white matter were lower than the ΔADCmax of high- and low-grade brain tumors. CONCLUSION: The dependence of ADC values on diffusion time between 6.5 ms and 35.2 ms was stronger in high-grade tumors than in low-grade tumors, suggesting differences in internal tissue structure. This finding highlights the importance of reporting diffusion times in ADC evaluations and might contribute to the grading of brain tumors using DWI.

Signal Intensity within Cerebral Venous Sinuses on Synthetic MRI.

PURPOSE: Flowing blood sometimes appears bright on synthetic T1-weighted images, which could be misdiagnosed as a thrombus. This study aimed to investigate the frequency of hyperintensity within cerebral venous sinuses on synthetic MR images and to evaluate the influence of increasing flow rates on signal intensity using a flow phantom. MATERIALS AND METHODS: Imaging data, including synthetic and conventional MRI scans, from 22 patients were retrospectively analyzed. Signal intensities at eight locations of cerebral venous sinuses on synthetic images were graded using the following three-point scale: 0, "dark vessel"; 1, "hyperintensity within the walls"; and 2, "hyperintensity within the lumen." A phantom with gadolinium solution inside a U-shaped tube was acquired without flow and then with increasing flow rates (60, 100, 200, 300, 400 ml/min). RESULTS: Considering all sinus locations, the venous signal intensity on synthetic T1-weighted images was graded as 2 in 79.8% of the patients. On synthetic T2-weighted images, all sinuses were graded as 0. On fluid-attenuated inversion recovery (FLAIR) images, sinuses were almost always graded as 0 (99.4%). In the phantom study, the signal initially became brighter on synthetic T1-weighted images as the flow rate increased. Above a certain flow rate, the signal started to decrease. CONCLUSION: High signal intensity within the cerebral venous sinuses is a frequent finding on synthetic T1-weighted images. This corresponds to the hyperintensity noted at certain flow rates in the phantom experiment.

Review of synthetic MRI in pediatric brains: Basic principle of MR quantification, its features, clinical applications, and limitations.

Quantitative magnetic resonance imaging (MRI) with multislice, multi-echo, and multi-delay acquisition enables simultaneous quantification of R1 and R2 relaxation rates, proton density, and the B1 field in a single acquisition, and requires only about 6 minutes for full-head coverage. Using dedicated SyMRI software, radiologists can generate any contrast-weighted image by manipulating the acquisition parameters, including repetition time, echo time, and inversion time. Moreover, automatic brain tissue segmentation, volumetry, and myelin measurement can also be performed. Using the SyMRI approach, a shorter scan time, an objective examination, and personalized MR imaging parameters can be obtained in daily clinical pediatric imaging. Here we summarize and review the use of SyMRI in imaging of the pediatric brain, including the basic principles of MR quantification along with its features, clinical applications, and limitations.

Choroid plexus cysts analyzed using diffusion-weighted imaging with short diffusion-time.

INTRODUCTION: Oscillating gradient spin-echo (OGSE) sequences can shorten diffusion times by replacing the long-lasting diffusion-sensitizing gradients used in pulsed gradient spin-echo (PGSE) methods with rapidly oscillating gradients. To obtain information regarding the internal structure of choroid plexus cysts that appear hyperintense on diffusion-weighted imaging (DWI), we investigated the apparent diffusion coefficient (ADC) values acquired with a shorter diffusion time using an OGSE sequence. MATERIAL AND METHODS: Twenty-seven patients with choroid plexus cysts were scanned using a 3 T magnetic resonance scanner. DWI was performed with both OGSE and PGSE, with effective diffusion times (Δeff) of 6.5 and 35.2 ms, respectively. ADC values for choroid plexus cysts, white matter (WM), and cerebrospinal fluid (CSF) were measured. The ADC values obtained with the shorter and longer diffusion times were compared using the Wilcoxon signed-rank test. P < .05 was considered significant. RESULTS: The ADC values of choroid plexus cysts and WM were significantly higher at the Δeff of 6.5 ms on OGSE than with the Δeff of 35.2 ms on PGSE. The ADC values of CSF were significantly lower at the Δeff of 6.5 ms on OGSE than with the Δeff of 35.2 ms on PGSE. The ADC values of choroid plexus cysts were lower than the ADC values of CSF with Δeff of 35.2 and 6.5 ms. CONCLUSIONS: The dependence of ADC values on the diffusion time in choroid plexus cysts suggested spatially restricted diffusion. In measurements obtained with short diffusion times, the lower ADC values for choroid plexus cysts in comparison with the CSF indicated the presence of spatially restricted diffusion and increased cyst viscosity.

Differentiating Alzheimer's Disease from Dementia with Lewy Bodies Using a Deep Learning Technique Based on Structural Brain Connectivity.

PURPOSE: Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) are representative disorders of dementia of the elderly and the neuroimaging has contributed to early diagnosis by estimation of alterations of brain volume, blood flow and metabolism. A brain network analysis by MR imaging (MR connectome) is a recently developed technique and can estimate the dysfunction of the brain network in AD and DLB. A graph theory which is a major technique of network analysis is useful for a group study to extract the feature of disorders, but is not necessarily suitable for the disorder differentiation at the individual level. In this investigation, we propose a deep learning technique as an alternative method of the graph analysis for recognition and classification of AD and DLB at the individual subject level. MATERIALS AND METHODS: Forty-eight brain structural connectivity data of 18 AD, 8 DLB and 22 healthy controls were applied to the machine learning consisting of a six-layer convolution neural network (CNN) model. Estimation of the deep learning model to classify AD, DLB and non-AD/DLB was performed using the 4-fold cross-validation method. RESULTS: The accuracy, average precision and recall of our CNN model were 0.73, 0.78 and 0.73, and the specificity precision and recall were 0.68 and 0.79 in AD, 0.94 and 0.65 in DLB and 0.73 and 0.75 in non-AD/DLB. The triangular probability map of the MR connectome revealed the probability of AD, DLB and non-AD/DLB in each subject. CONCLUSION: Our preliminary investigation revealed the adaptation of deep learning to the MR connectome and proposed its utility in the differentiation of dementia disorders at the individual subject level.

Myelin Measurement: Comparison Between Simultaneous Tissue Relaxometry, Magnetization Transfer Saturation Index, and T1w/T2w Ratio Methods.

Magnetization transfer (MT) imaging has been widely used for estimating myelin content in the brain. Recently, two other approaches, namely simultaneous tissue relaxometry of R1 and R2 relaxation rates and proton density (SyMRI) and the ratio of T1-weighted to T2-weighted images (T1w/T2w ratio), were also proposed as methods for measuring myelin. SyMRI and MT imaging have been reported to correlate well with actual myelin by histology. However, for T1w/T2w ratio, such evidence is limited. In 20 healthy adults, we examined the correlation between these three methods, using MT saturation index (MTsat) for MT imaging. After calibration, white matter (WM) to gray matter (GM) contrast was the highest for SyMRI among these three metrics. Even though SyMRI and MTsat showed strong correlation in the WM (r = 0.72), only weak correlation was found between T1w/T2w and SyMRI (r = 0.45) or MTsat (r = 0.38) (correlation coefficients significantly different from each other, with p values < 0.001). In subcortical and cortical GM, these measurements showed moderate to strong correlations to each other (r = 0.54 to 0.78). In conclusion, the high correlation between SyMRI and MTsat indicates that both methods are similarly suited to measure myelin in the WM, whereas T1w/T2w ratio may be less optimal.

Changes in the ADC of diffusion-weighted MRI with the oscillating gradient spin-echo (OGSE) sequence due to differences in substrate viscosities.

PURPOSE: Compared with the conventional pulsed gradient spin-echo (PGSE) sequence, diffusion-weighted imaging (DWI) with the oscillating gradient spin-echo (OGSE) sequence can shorten the diffusion time by changing the frequency. The purpose was to investigate whether n-alkanes are suitable as isotropic phantoms for estimating the diffusion coefficient with the OGSE sequence. MATERIALS AND METHODS: We investigated changes in the apparent diffusion coefficient (ADC) due to differences in the viscosities of nine n-alkane phantoms with different numbers of carbon atoms from C8H18 to C16H34 using OGSE and PGSE sequences at 21 °C. Effective diffusion times of 4.3, 5.1, 6.5, 9.3, 20, 40, and 60 ms were used. The T2 relaxation times of each n-alkane phantom were measured using quantitative synthetic magnetic resonance imaging (MRI). Circular regions of interest were placed manually within the alkane phantoms on ADC and T2 maps. RESULTS: In each alkane phantom, changes in mean ADC values were almost constant with changes in diffusion times. Viscosities and ADC values showed inverse proportionality, as expected theoretically. CONCLUSION: The ADC values of alkanes do not depend on diffusion times. The n-alkanes can be useful phantoms for assessing the accuracy of clinical protocols of DWI with the OGSE sequence.

Spatial Restriction within Intracranial Epidermoid Cysts Observed Using Short Diffusion-time Diffusion-weighted Imaging.

We report two cases of pathologically proven intracranial epidermoid cysts. Both cases were scanned with diffusion-weighted imaging using pulsed gradient spin-echo (PGSE) and oscillating gradient spin-echo (OGSE; 50 Hz) prototype sequences with diffusion times of 47.3 ms and 8.5 ms, respectively. The apparent diffusion coefficient measured by OGSE was higher than that measured by PGSE, indicating the spatial restriction of water diffusion in the laminated keratin layers within the cyst as demonstrated by histopathology.

Thalamic control of human attention driven by memory and learning.

The role of the thalamus in high-level cognition-attention, working memory (WM), rule-based learning, and decision making-remains poorly understood, especially in comparison to that of cortical frontoparietal networks [1-3]. Studies of visual thalamus have revealed important roles for pulvinar and lateral geniculate nucleus in visuospatial perception and attention [4-10] and for mediodorsal thalamus in oculomotor control [11]. Ventrolateral thalamus contains subdivisions devoted to action control as part of a circuit involving the basal ganglia [12, 13] and motor, premotor, and prefrontal cortices [14], whereas anterior thalamus forms a memory network in connection with the hippocampus [15]. This connectivity profile suggests that ventrolateral and anterior thalamus may represent a nexus between mnemonic and control functions, such as action or attentional selection. Here, we characterize the role of thalamus in the interplay between memory and visual attention. We show that ventrolateral lesions impair the influence of WM representations on attentional deployment. A subsequent fMRI study in healthy volunteers demonstrates involvement of ventrolateral and, notably, anterior thalamus in biasing attention through WM contents. To further characterize the memory types used by the thalamus to bias attention, we performed a second fMRI study that involved learning of stimulus-stimulus associations and their retrieval from long-term memory to optimize attention in search. Responses in ventrolateral and anterior thalamic nuclei tracked learning of the predictiveness of these abstract associations and their use in directing attention. These findings demonstrate a key role for human thalamus in higher-level cognition, notably, in mnemonic biasing of attention.