The automatic evaluation of steno-occlusive changes in time-of-flight magnetic resonance angiography of moyamoya patients using a 3D coordinate attention residual network

Abstract

Background: Moyamoya disease (MMD) is a rare cerebrovascular occlusive disease with progressive stenosis of the terminal portion of internal cerebral artery (ICA) and its main branches, which can cause complications, such as high risks of disability and increased mortality. Accurate and timely diagnosis may be difficult for physicians who are unfamiliar to MMD. Therefore, this study aims to achieve a preoperative deep-learning-based evaluation of MMD by detecting steno-occlusive changes in the middle cerebral artery or distal ICA areas.Methods: A fine-tuned deep learning model was developed using a three-dimensional (3D) coordinate attention residual network (3D CA-ResNet). This study enrolled 50 preoperative patients with MMD and 50 controls, and the corresponding time of flight magnetic resonance angiography (TOF-MRA) imaging data were acquired. The 3D CA-ResNet was trained based on sub-volumes and tested using patch-based and subject-based methods. The performance of the 3D CA-ResNet, as evaluated by the area under the curve (AUC) of receiving-operator characteristic, was compared with that of three other conventional 3D networks.Results: With the resulting network, the patch-based test achieved an AUC value of 0.94 for the 3D CA-ResNet in 480 patches from 10 test patients and 10 test controls, which is significantly higher than the results of the others. The 3D CA-ResNet correctly classified the MMD patients and normal healthy controls, and the vascular lesion distribution in subjects with the disease was investigated by generating a stenosis probability map and 3D vascular structure segmentation. Conclusions:The results demonstrated the reliability of the proposed 3D CA-ResNet in detecting stenotic areas on TOF-MRA imaging, and it outperformed three other models in identifying vascular steno-^ ORCID: 0000-0002-3711-5173.

Accuracy of 3.0T magnetic resonance angiography for the detection of arteriovenous fistula dysfunction in hemodialysis patients requiring interventional therapy: a prospective study

“…TOF-MRA, as a nonenhanced MR angiography technique, has been carried out for the assessment of cerebrovascular disease ( 25 , 26 ) and is seldom used for hemodialysis access since TOF-MRA is unable to adequately depict the venous vessels. When TOF-MRA was applied for cerebrovasculature, a presaturation pulse was used inferior to the slice to suppress the signal from venous blood, and the presaturation bands were omitted in our study to allow simultaneous visualization of both arterial and venous blood flow.…”

High-resolution magnetic resonance vessel wall imaging provides new insights into Moyamoya disease

“…Combine emerging technologies, like AI, to improve HR-VWI performance ( Kim et al, 2019 ). A Pseudo-Three-Dimensional Residual Network has been proposed to process spatio-temporal information and combined with DSA to recognize MMD ( Xu et al, 2023 ), and a 3D coordinate attention residual network has also been used in combination with MRA to evaluate stenotic occlusion changes in MMD ( Zhang Z. et al, 2023 ). In the future, it is possible to use these techniques in combination with HR-VWI.…”