Search

BACKGROUND AND PURPOSE: Fast and accurate quantification of globe volumes in the event of an ocular trauma can provide clinicians with valuable diagnostic information. In this work, an automated workflow using a deep learning-based convolutional neural network is proposed for prediction of globe contours and their subsequent volume quantification in CT images of the orbits. MATERIALS AND METHODS: An automated workflow using a deep learning -based convolutional neural network is proposed for prediction of globe contours in CT images of the orbits. The network, 2D Modified Residual UNET (MRes-UNET2D), was trained on axial CT images from 80 subjects with no imaging or clinical findings of globe injuries. The predicted globe contours and volume estimates were compared with manual annotations by experienced observers on 2 different test cohorts. RESULTS: On the first test cohort (n = 18), the average Dice, precision, and recall scores were 0.95, 96%, and 95%, respectively. The average 95% Hausdorff distance was only 1.5 mm, with a 5.3% error in globe volume estimates. No statistically significant differences (P = .72) were observed in the median globe volume estimates from our model and the ground truth. On the second test cohort (n = 9) in which a neuroradiologist and 2 residents independently marked the globe contours, MRes-UNET2D (Dice = 0.95) approached human interobserver variability (Dice = 0.94). We also demonstrated the utility of inter-globe volume difference as a quantitative marker for trauma in 3 subjects with known globe injuries. CONCLUSIONS: We showed that with fast prediction times, we can reliably detect and quantify globe volumes in CT images of the orbits across a variety of acquisition parameters.

A 74-year-old woman with primary hyperparathyroidism diagnosed from routine laboratory tests described symptoms of fatigue and difficulty with concentration. During surgical consultation, the cervical and thoracic spine MRI scans from the preceding 10-year period, performed for relapsing-remitting multiple sclerosis, were reviewed. In this clinical context, the slowly enlarging left upper paraesophageal lesion, reported as a lateral proximal esophageal (Killian-Jamieson) diverticulum, was reevaluated for a potential parathyroid adenoma. 99mTc-sestamibi SPECT/CT demonstrated focal uptake in the paraesophageal lesion with surgical resection, confirming it to be a large parathyroid adenoma.

This article discusses instrumented spinal surgeries, the radiologic assessment of spinal fixation hardware, and the potential complications of spinal hardware. Radiography is the standard for the postoperative assessment of spinal hardware. Computed tomography and magnetic resonance imaging play a valuable role in the detection of hardware and postsurgical-related complications such as infection, pseudarthrosis, and malpositioned instrumentation. Familiarity with the normal imaging appearance of implanted spinal hardware along with the expected progression of normal postoperative osseous arthrodesis enables recognition of potential complications and helps facilitate appropriate clinical management.

BACKGROUND AND PURPOSE: Accurate and reliable detection of white matter hyperintensities and their volume quantification can provide valuable clinical information to assess neurologic disease progression. In this work, a stacked generalization ensemble of orthogonal 3D convolutional neural networks, StackGen-Net, is explored for improving automated detection of white matter hyperintensities in 3D T2-FLAIR images. MATERIALS AND METHODS: Individual convolutional neural networks in StackGen-Net were trained on 2.5D patches from orthogonal reformatting of 3D-FLAIR (n = 21) to yield white matter hyperintensity posteriors. A meta convolutional neural network was trained to learn the functional mapping from orthogonal white matter hyperintensity posteriors to the final white matter hyperintensity prediction. The impact of training data and architecture choices on white matter hyperintensity segmentation performance was systematically evaluated on a test cohort (n = 9). The segmentation performance of StackGen-Net was compared with state-of-the-art convolutional neural network techniques on an independent test cohort from the Alzheimer’s Disease Neuroimaging Initiative-3 (n = 20). RESULTS: StackGen-Net outperformed individual convolutional neural networks in the ensemble and their combination using averaging or majority voting. In a comparison with state-of-the-art white matter hyperintensity segmentation techniques, StackGen-Net achieved a significantly higher Dice score (0.76 [SD, 0.08], F1-lesion (0.74 [SD, 0.13]), and area under precision-recall curve (0.84 [SD, 0.09]), and the lowest absolute volume difference (13.3% [SD, 9.1%]). StackGen-Net performance in Dice scores (median = 0.74) did not significantly differ (P = .22) from interobserver (median = 0.73) variability between 2 experienced neuroradiologists. We found no significant difference (P = .15) in white matter hyperintensity lesion volumes from StackGen-Net predictions and ground truth annotations. CONCLUSIONS: A stacked generalization of convolutional neural networks, utilizing multiplanar lesion information using 2.5D spatial context, greatly improved the segmentation performance of StackGen-Net compared with traditional ensemble techniques and some state-of-the-art deep learning models for 3D-FLAIR.

We describe the unique MR imaging characteristics of intraocular perfluoro-n-octane, a liquid used for intraoperative and postoperative tamponade in the context of complex retinal detachment repair, and contrast it with other intraocular pathologies. Because trace amounts of perfluoro-n-octane may be left in the globe postoperatively, it may be confused for other abnormalities, such as foreign bodies or tumors.

Upper respiratory tract infections are the most frequent outpatient disease processes treated by the pediatricians. Although upper respiratory tract infections are typically self-limited viral illnesses, a subset of children will develop acute bacterial sinusitis, which is differentiated from the former viral nasopharyngitis by prolonged illness, worsening course, or severe onset of symptoms. Although diagnostic imaging is not indicated in the differentiation of a viral upper respiratory tract infection from acute bacterial sinusitis, contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI) are critical in the detection and classification of potentially serious complications of acute bacterial sinusitis resulting from intraorbital or intracranial spread of infection. Orbital infections are the most frequently associated complications of paranasal sinus inflammatory disease given the proximity of the paranasal sinuses and orbits. Imaging is essential in differentiating preseptal cellulitis from orbital cellulitis and guiding appropriate therapy. Cross-sectional imaging is also important in the characterization of noninvasive fungal sinusitis, forms of chronic sinusitis in the setting of an intact immune system, and invasive fungal sinusitis, an acute fulminant sinusitis induced by angioinvasive fungi in immunocompromised children requiring emergent surgical management.

A pedagogical review of fluid-attenuated inversion recovery (FLAIR) and double inversion recovery (DIR) imaging is conducted in this article. The basics of the two pulse sequences are first described, including the details of the inversion preparation and imaging sequences with accompanying mathematical formulae for choosing the inversion time in a variety of scenarios for use on clinical MRI scanners. Magnetization preparation (or T2prep), a strategy for improving image signal-to-noise ratio and contrast and reducing T1 weighting at high field strengths, is also described. Lastly, image artifacts commonly associated with FLAIR and DIR are described with clinical examples, to help avoid misdiagnosis. Level of evidence 5 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2017;46:1590-1600.

Despite the popularity of flow-diverting stents for the treatment of cerebral aneurysms, there is no widely accepted scale for the characterization of results. We present an outcomes-based grading scale that considers factors related to failure of flow diversion.The grading scale was developed using the results from consecutive patients at two institutions who were treated with flow diversion for a cerebral aneurysm. The initial treatment results were graded on patient, aneurysm, and treatment characteristics. A 6-point grading scale was developed based on these data.One hundred and seventy-one patients were included in the patient cohort. When compared by multivariate analysis with patients without residuals, patients with aneurysm residuals were found to be older (age ≥60 years, p=0.01, OR 1.17, 95% CI 1.03 to 1.33), to have larger aneurysms (size ≥15 mm, p0.01, OR 1.38, 95% CI 1.17 to 1.62), to have aneurysms with associated side branches (p=0.02, OR 1.17, 95% CI 1.03 to 1.33), and to have a post-treatment Raymond score of 2 or 3 (p=0.01, OR 1.28, 95% CI 1.06 to 1.56). Using the Raymond score (1-3) as the foundation for the grading scale, additional points (0 or 1) were given for the other three identified factors, creating a 6-point scale. We found that patients with residual aneurysms had statistically higher final tabulated scores (p0.01).We propose a novel straightforward outcomes-based scale to characterize results after flow diversion treatment of cerebral aneurysms. This scale may provide the basis for the common reporting of results in future studies.

BackgroundWith increasing use of flow-diverting stents for the treatment of intracranial aneurysms, standardized methods and a common language to evaluate angiographic outcomes are needed. Multiple grading scales have been developed for this purpose but none has been widely adopted.ObjectiveTo analyze these scales to determine interobserver reliability.MethodsFour independent assessors scored the intraprocedural angiograms of patients who underwent flow-diverting stent deployment for an intracranial saccular or fusiform aneurysm at our institution between October 2012 and June 2015. Angiographic outcome immediately after flow-diverting stent deployment was scored using three grading scales (Kamran–Byrne (KB), Simple Measurement of Aneurysm Residual after Treatment (SMART), and O'Kelley, Krings, Marotta (OKM)). Statistical analysis was performed using Light's κ for multiple raters (κ), Kendall's coefficient of concordance (W), and intraclass correlation (ICC).ResultsWe included the angiograms of 50 consecutive patients (mean age 58 years, range 30–79) who underwent flow-diverting stent deployment for an intracranial aneurysm (40 saccular, 10 fusiform). Six aneurysms were located in the posterior circulation. The inter-rater reliability was typically poor or fair: SMART aneurysm filling (κ=0.30, W=0.36, ICC=0.12), SMART parent vessel stenosis (κ=0.07, W=0.33, ICC=0.12), KB axis I (κ=0.24, W=0.50, ICC=0.25), KB axis II (κ=0.07, W=0.30, ICC=0.06), OKM aneurysm filling (κ=0.23, W=0.45, ICC=0.13), OKM contrast stasis (κ=0.36,W=0.71, ICC=0.54).ConclusionsExisting flow-diverting stent grading scales have low inter-rater reliability for most categories.

A 56-y-old man underwent total thyroidectomy and bilateral central and right lateral neck dissection for papillary thyroid carcinoma with lymph nodes metastases. Before radioiodine ablation, a 123I scan established the diagnosis of primary nasolacrimal duct obstruction (dacryostenosis).

BACKGROUND AND PURPOSE: TSE-based T2-weighted imaging of the spine has long scan times. This work proposes a fast imaging protocol using variable refocusing flip angles, optimized for blurring and specific absorption rate. MATERIALS AND METHODS: A variable refocusing flip angle echo-train was optimized for the spine to improve the point spread function and minimize the specific absorption rate, yielding images with improved spatial resolution and SNR compared with the constant flip angle sequence. Data were acquired from 51 patients (35 lumbar, 16 whole-spine) using conventional TSE and the proposed sequence, with a single-shot variant for whole-spine. Noninferiority analysis was performed to evaluate the efficiency of the proposed technique. RESULTS: The proposed multishot sequence resulted in a 2× shorter scan time with a >1.5× lower specific absorption rate. The variable flip angle sequence was noninferior to the conventional TSE (P < .025) for all image-quality and clinical criteria except signal-to-noise ratio for the lumbar spine protocol. However, mean image scores for the TSE-variable refocusing flip angle were ≥4.3 for all criteria, and concordance analysis showed high agreement (>90%) with the TSE, indicating clinical equivalence. The single-shot sequence resulted in 4× shorter whole-spine scans, and image scores were ≥4.4 for all criteria, attesting to its clinical utility. CONCLUSIONS: We present a fast T2-weighted spine protocol using variable refocusing flip angles, including a single-shot variant. The sequences have better point spread function behavior than their constant flip angle counterparts and, being faster, should be less sensitive to patient motion, often seen in the longer TSE scans.

A 63-year-old woman with epilepsy and chronic headaches was admitted for status epilepticus. A lumbar puncture revealed increased opening pressure of 320 mm H2O and bacterial meningitis. MRI brain demonstrated a partial empty sella, tortuous optic nerve sheaths, flattening of the optic papillae, and numerous prominent arachnoid granulations, consistent with chronic idiopathic intracranial hypertension (IIH), as well as bright CSF signal and brain parenchyma within a left meningoencephalocele (figure, A). A CT cisternogram confirmed communication of the subarachnoid space of the meningoencephalocele with the left sphenoid sinus through a small osseous defect (figure, B). While most CSF leaks are related to trauma, IIH is increasingly recognized as a cause of spontaneous leaks.1,2 While the increased opening pressure may have been due to her acute meningitis, her headache symptoms and MRI findings are suggestive of longstanding intracranial hypertension. Monitoring and medical treatment of IIH must accompany surgical repair to avoid recurrence.2