Background and purpose

Patients with unruptured intracranial aneurysms routinely undergo surveillance imaging to monitor growth. Angiography is the criterion standard for aneurysm diagnosis, but it is invasive. This study aimed to evaluate the accuracy and reproducibility of a 3D noncontrast black-blood MR imaging technique for unruptured intracranial aneurysm measurement in comparison with 3D-TOF and contrast-enhanced MRA, using 3D rotational angiography as a reference standard.

Materials and methods

Sixty-four patients (57.3 ± 10.9 years of age, 41 women) with 68 saccular unruptured intracranial aneurysms were recruited. Patients underwent 3T MR imaging with 3D-TOF-MRA, 3D black-blood MR imaging, and contrast-enhanced MRA, and they underwent 3D rotational angiography within 2 weeks. The neck, width, and height of the unruptured intracranial aneurysms were measured by 2 radiologists independently on 3D rotational angiography and 3 MR imaging sequences. The accuracy and reproducibility were evaluated by Bland-Altman plots, the coefficient of variance, and the intraclass correlation coefficient.

Results

3D black-blood MR imaging demonstrates the best agreement with DSA, with the smallest limits of agreement and measurement error (coefficients of variance range, 5.87%-7.04%). 3D-TOF-MRA had the largest limits of agreement and measurement error (coefficients of variance range, 12.73%-15.78%). The average coefficient of variance was 6.26% for 3D black-blood MR imaging, 7.03% for contrast-enhanced MRA, and 15.54% for TOF-MRA. No bias was found among 3 MR imaging sequences compared with 3D rotational angiography. All 3 MR imaging sequences had excellent interreader agreement (intraclass correlation coefficient, >0.95). 3D black-blood MR imaging performed the best for patients with intraluminal thrombus (n = 10).

Conclusions

3D black-blood MR imaging achieves better accuracy for aneurysm size measurements compared with 3D-TOF, using 3D rotational angiography as a criterion standard. This noncontrast technique is promising for surveillance of unruptured intracranial aneurysms.

The S=1/2 spin chain material SrCuO2 doped with 1% S=1 Ni impurities is studied by inelastic neutron scattering. At low temperatures, the spectrum shows a pseudogap Δ≈8  meV, absent in the parent compound, and not related to any structural phase transition. The pseudogap is shown to be a generic feature of quantum spin chains with dilute defects. A simple model based on this idea quantitatively accounts for the experimental data measured in the temperature range from 2 to 300 K, and allows us to represent the momentum-integrated dynamic structure factor in a universal scaling form.

Background and purpose

Convolutional neural networks are a powerful technology for image recognition. This study evaluates a convolutional neural network optimized for the detection and quantification of intraparenchymal, epidural/subdural, and subarachnoid hemorrhages on noncontrast CT.

Materials and methods

This study was performed in 2 phases. First, a training cohort of all NCCTs acquired at a single institution between January 1, 2017, and July 31, 2017, was used to develop and cross-validate a custom hybrid 3D/2D mask ROI-based convolutional neural network architecture for hemorrhage evaluation. Second, the trained network was applied prospectively to all NCCTs ordered from the emergency department between February 1, 2018, and February 28, 2018, in an automated inference pipeline. Hemorrhage-detection accuracy, area under the curve, sensitivity, specificity, positive predictive value, and negative predictive value were assessed for full and balanced datasets and were further stratified by hemorrhage type and size. Quantification was assessed by the Dice score coefficient and the Pearson correlation.

Results

A 10,159-examination training cohort (512,598 images; 901/8.1% hemorrhages) and an 862-examination test cohort (23,668 images; 82/12% hemorrhages) were used in this study. Accuracy, area under the curve, sensitivity, specificity, positive predictive value, and negative-predictive value for hemorrhage detection were 0.975, 0.983, 0.971, 0.975, 0.793, and 0.997 on training cohort cross-validation and 0.970, 0.981, 0.951, 0.973, 0.829, and 0.993 for the prospective test set. Dice scores for intraparenchymal hemorrhage, epidural/subdural hemorrhage, and SAH were 0.931, 0.863, and 0.772, respectively.

Conclusions

A customized deep learning tool is accurate in the detection and quantification of hemorrhage on NCCT. Demonstrated high performance on prospective NCCTs ordered from the emergency department suggests the clinical viability of the proposed deep learning tool.

Current guidelines for primary and secondary prevention of stroke in patients with carotid atherosclerosis are based on the quantification of the degree of stenosis and symptom status. Recent publications have demonstrated that plaque morphology and composition, independent of the degree of stenosis, are important in the risk stratification of carotid atherosclerotic disease. This finding raises the question as to whether current guidelines are adequate or if they should be updated with new evidence, including imaging for plaque phenotyping, risk stratification, and clinical decision-making in addition to the degree of stenosis. To further this discussion, this roadmap consensus article defines the limits of luminal imaging and highlights the current evidence supporting the role of plaque imaging. Furthermore, we identify gaps in current knowledge and suggest steps to generate high-quality evidence, to add relevant information to guidelines currently based on the quantification of stenosis.

Multiplicities of charged hadrons produced in deep inelastic muon scattering off a 6LiD target have been measured as a function of the DIS variables xBj , Q2, W2 and the final state hadron variables pT and z. The p2T distributions are fitted with a single exponential function at low values of p2T to determine the dependence of 〈p2T 〉 on xBj , Q2, W2 and z. The z-dependence of 〈p2T 〉 is shown to be a potential tool to extract the average intrinsic transverse momentum squared of partons, 〈k2⊥〉, as a function of xBj and Q2 in a leading order QCD parton model.