The potential role of IDEAL MRI for identification of lipids and hemorrhage in carotid artery plaques

Deep learning-based fully automatic screening of carotid artery plaques in computed tomography angiography: a multicenter study

AIM

To develop and validate a deep learning (DL) algorithm for the automated detection and classification of carotid artery plaques (CAPs) on computed tomography angiography (CTA) images.

MATERIALS AND METHODS

This retrospective study enrolled 400 patients (300 in the Center Ⅰ and 100 in Ⅱ). Three radiologists co-labeled CAPs, and their revised calcification status (noncalcified, mixed, and calcified) was regarded as ground truth. Center Ⅰ patients were randomly divided into training and internal validation datasets, while Center Ⅱ patients served as the external validation dataset. Carotid artery regions were segmented using a modified 3D-UNet network, followed by CAPs detection and classification using a ResUNet-based architecture in a two-step DL system. The DL model's detection and classification performance were evaluated on the validation dataset using precision-recall curve, free-response receiver operating characteristic (fROC) curve, Cohen's kappa, and ROC curve analysis.

RESULTS

The DL model had achieved 83.4% sensitivity at 3.0 false positives (FPs)/CTA scan in internal validation and 78.9% in external validation. F1-scores were 0.764 and 0.769 at the optimal threshold, and area under fROC curves were 0.756 and 0.738, respectively, indicating good overall accuracy for CAP detection. The DL model also showed good performance for the ternary classification of CAPs, with Cohen's kappa achieved 0.728 and 0.703 in both validation datasets.

CONCLUSION

This study demonstrated the feasibility of using a fully automated DL-based algorithm for the detection and ternary classification of CAPs, which could be helpful for the workloads of radiologists.

A Comparative Study of Carotid Magnetic Resonance Plaque Imaging in Predicting the Early Progression of Acute Anterior Circulation Mild Stroke

To analyze the biological properties of carotid lumen stenosis and plaque before and after bifurcation of the bilateral carotid arteries in patients with acute anterior circulation mild stroke (AACMS) by 3.0 T high-resolution magnetic resonance imaging (HR-MRI) with the aim to clarify the predictive effect of 3.0 T HR-MRI on early progression and poor prognosis in patients with AACMS. Random division of 96 patients with AACMS into a stroke progression group and a non-progression group. The bilateral cervical vessels of the patients were detected by HR-MRI. The degree of carotid artery stenosis before and after bilateral carotid bifurcation was evaluated using a vascular plaque imaging diagnostic system. There were significant differences in the maximum and average wall standardization index, maximum wall thickness, maximum wall area, plaque composition, proportion of plaque fibrous cap rupture, and proportion of VI complex plaques in the progressive group. There was no significant difference in the related indexes of bilateral vessels in the non-progressive group (P >0.05). There are significant differences in MRI morphological characteristics of bilateral carotid plaques in patients with AACMS progression. Through a comparative analysis of the plaque load and plaque composition of bilateral carotid arteries using 3.0 T HR-MRI in patients with early-stage AACMS, the type and stability of complex plaques can be identified, which serve as prognostic factors in predicting the early progression of stroke and guiding clinical treatment.

The diagnostic value of the IDEAL-T2WI sequence in dysthyroid optic neuropathy: a quantitative analysis of the optic nerve and cerebrospinal fluid in the optic nerve sheath

OBJECTIVES

To evaluate the optic nerve and CSF in the optic nerve sheath as imaging markers of dysthyroid optic neuropathy (DON).

METHODS

In this single-centre retrospective study, orbital images of 30 consecutive participants (54 orbits) with DON, 30 patients (60 orbits) with thyroid-associated ophthalmopathy (TAO) without DON, and 19 healthy controls (HCs; 38 orbits) were analysed. The diameter and cross-sectional area of the optic nerve and its sheath, water fraction of the optic nerve, and volume of the fluid in the optic nerve sheath were measured and compared. The associations between MR parameters and clinical measures were assessed using correlation analysis.

RESULTS

The diameter and water fraction of the optic nerve (3 mm and 6 mm behind the eyeball), optic nerve subarachnoid space (ONSS) (3 mm and 6 mm behind the eyeball), and subarachnoid fluid volume in the optic nerve sheath were significantly greater in the DON group than in the TAO group (p < 0.01) or HC group (p < 0.01). ROC analysis showed that ONSS 3 mm behind the eyeball (ONSS3) was a robust predictor of DON (AUC = 0.957, sensitivity = 0.907, specificity = 0.9). Water fraction of the optic nerve 3 mm behind the eyeball (water fraction3) had the best specificity (0.967). Water fraction3, fluid volume in the optic nerve sheath, and optic nerve diameter (3 mm behind the eyeball) were correlated with clinical measures (i.e. clinical activity score, mean defect, and pattern standard deviation).

CONCLUSIONS

Increased water fraction of the optic nerve and ONSS3 are promising and easily accessible radiological markers for diagnosing DON.

KEY POINTS

• The water fraction of the optic nerve and optic nerve subarachnoid space (ONSS) are greater in patients with dysthyroid optic neuropathy (DON) than in patients with thyroid-associated ophthalmopathy (TAO) without DON. • The optic nerve and the cerebrospinal fluid in the optic nerve sheath measures are associated with visual dysfunction. • The water fraction of the optic nerve and ONSS may be promising imaging markers for diagnosing DON.