1
|
Seo M, Jung W, Jeong G, Yang S, Shin I, Lee JY, Ahn KJ, Kim BS, Jang J. Deep learning improves quality of intracranial vessel wall MRI for better characterization of potentially culprit plaques. Sci Rep 2024; 14:18983. [PMID: 39152167 PMCID: PMC11329665 DOI: 10.1038/s41598-024-69750-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/15/2024] [Accepted: 08/08/2024] [Indexed: 08/19/2024] Open
Abstract
Intracranial vessel wall imaging (VWI), which requires both high spatial resolution and high signal-to-noise ratio (SNR), is an ideal candidate for deep learning (DL)-based image quality improvement. Conventional VWI (Conv-VWI, voxel size 0.51 × 0.51 × 0.45 mm3) and denoised super-resolution DL-VWI (0.28 × 0.28 × 0.45 mm3) of 117 patients were analyzed in this retrospective study. Quality of the images were compared qualitatively and quantitatively. Diagnostic performance for identifying potentially culprit atherosclerotic plaques, using lesion enhancement and presence of intraplaque hemorrhage (IPH), was evaluated. DL-VWI significantly outperformed Conv-VWI in all image quality ratings (all P < .001). DL-VWI demonstrated higher SNR and contrast-to-noise ratio (CNR) than Conv-VWI, both in normal walls (basilar artery; SNR 4.83 ± 1.23 vs. 3.02 ± 0.59, P < .001) and lesions (contrast-enhanced images; SNR 22.12 ± 11.68 vs. 8.33 ± 3.26, P < .001). In the assessment of 86 lesions, DL-VWI showed higher confidence of detection (4.56 ± 0.55 vs. 2.62 ± 0.77, P < .001), more concordant IPH characterization (Cohen's Kappa 0.85 vs. 0.59) and greater enhancement. For culprit plaque identification, IPH exhibited higher sensitivity in DL-VWI compared to Conv-VWI (70.6% vs. 23.5%) and excellent specificity (94.3% vs. 94.3%). Deep learning application of intracranial vessel wall images successfully improved the quality and resolution of the images. This aided in detecting vessel wall lesions and intraplaque hemorrhage, and in identifying potentially culprit atherosclerotic plaques.
Collapse
Affiliation(s)
- Minkook Seo
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea
| | | | | | | | - Ilah Shin
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea
| | - Ji Young Lee
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea
| | - Kook-Jin Ahn
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea
| | - Bum-Soo Kim
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea
| | - Jinhee Jang
- Department of Radiology, College of Medicine, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul, 06591, Republic of Korea.
| |
Collapse
|
2
|
Kang DW, Kim DY, Kim J, Baik SH, Jung C, Singh N, Song JW, Bae HJ, Kim BJ. Emerging Concept of Intracranial Arterial Diseases: The Role of High Resolution Vessel Wall MRI. J Stroke 2024; 26:26-40. [PMID: 38326705 PMCID: PMC10850450 DOI: 10.5853/jos.2023.02481] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/29/2023] [Revised: 10/27/2023] [Accepted: 12/04/2023] [Indexed: 02/09/2024] Open
Abstract
Intracranial arterial disease (ICAD) is a heterogeneous condition characterized by distinct pathologies, including atherosclerosis. Advances in magnetic resonance technology have enabled the visualization of intracranial arteries using high-resolution vessel wall imaging (HR-VWI). This review summarizes the anatomical, embryological, and histological differences between the intracranial and extracranial arteries. Next, we review the heterogeneous pathophysiology of ICAD, including atherosclerosis, moyamoya or RNF213 spectrum disease, intracranial dissection, and vasculitis. We also discuss how advances in HR-VWI can be used to differentiate ICAD etiologies. We emphasize that one should consider clinical presentation and timing of imaging in the absence of pathology-radiology correlation data. Future research should focus on understanding the temporal profile of HR-VWI findings and developing quantitative interpretative approaches to improve the decision-making and management of ICAD.
Collapse
Affiliation(s)
- Dong-Wan Kang
- Department of Neurology, Seoul National University Bundang Hospital, Seongnam, Korea
- Headquarters for Public Health Care, Seoul National University Bundang Hospital, Seongnam, Korea
- Department of Neurology, Gyeonggi Provincial Medical Center, Icheon Hospital, Icheon, Korea
| | - Do Yeon Kim
- Department of Neurology, Seoul National University Bundang Hospital, Seongnam, Korea
- Headquarters for Public Health Care, Seoul National University Bundang Hospital, Seongnam, Korea
- Department of Neurology, Gyeonggi Provincial Medical Center, Icheon Hospital, Icheon, Korea
| | - Jonguk Kim
- Department of Neurology, Inha University Hospital, Incheon, Korea
| | - Sung Hyun Baik
- Department of Radiology, Seoul National University Bundang Hospital, Seongnam, Korea
| | - Cheolkyu Jung
- Department of Radiology, Seoul National University Bundang Hospital, Seongnam, Korea
| | - Nishita Singh
- Department of Internal Medicine-Neurology Division, Max Rady College of Medicine, University of Manitoba, Winnipeg, MB, Canada
| | - Jae W. Song
- Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA, USA
| | - Hee-Joon Bae
- Department of Neurology, Seoul National University Bundang Hospital, Seongnam, Korea
| | - Beom Joon Kim
- Department of Neurology, Seoul National University Bundang Hospital, Seongnam, Korea
| |
Collapse
|
3
|
Cheng X, Liu J, Li H, Yang J, Zhou C, Zhi B, Liu Q, Li Y, Xiao L, Zhu W, Lu G. Incremental value of enhanced plaque length for identifying intracranial atherosclerotic culprit plaques: a high-resolution magnetic resonance imaging study. Insights Imaging 2023; 14:99. [PMID: 37227551 DOI: 10.1186/s13244-023-01449-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2023] [Accepted: 05/14/2023] [Indexed: 05/26/2023] Open
Abstract
OBJECTIVES Besides plaque enhancement grade, the incremental value of enhancement-related high-resolution MRI features in defining culprit plaques needs further evaluation. This study was focused on assessing whether plaque enhancement features contribute to culprit plaque identification and further risk stratification. METHODS We retrospectively studied patients who experienced an acute ischaemic stroke and transient ischaemic attack due to intracranial atherosclerosis from 2016 to 2022. The enhancement features included enhancement grade, enhanced length, and enhancement quadrant. Associations between plaque enhancement features and culprit plaques, as well as diagnostic value, were investigated using logistic regression and receiver operating characteristic analyses. RESULTS Overall, 287 plaques were identified, of which 231 (80.5%) and 56 (19.5%) were classified as culprit and non-culprit plaques, respectively. Comparison of the pre- and post-enhancement images revealed enhanced length longer than the plaque length in 46.32% of the culprit plaques. Multivariate logistic regression showed that enhanced length longer than plaque length (OR 6.77; 95% CI 2.47-18.51) and grade II enhancement (OR 7.00; 95% CI 1.69-28.93) were independently associated with culprit plaques. The area under the curve value for the combination of stenosis and plaque enhancement grade for the diagnosis of culprit plaques was 0.787, which increased significantly to 0.825 on the addition of enhanced length longer than the plaque length (p = 0.026 for DeLong's test). CONCLUSIONS Enhanced length longer than the plaque length and grade II enhancement were independently associated with culprit plaques. The combination of the enhanced plaque features resulted in better culprit plaque identification.
Collapse
Affiliation(s)
- XiaoQing Cheng
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China.
| | - Jia Liu
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China
| | - HongXia Li
- Department of Medical Imaging, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, 210002, Jiangsu, China
| | - JiaLuo Yang
- Department of Medical Imaging, Jinling Hospital, Nanjing University of Chinese Medicine, Nanjing, 210002, Jiangsu, China
| | - ChangSheng Zhou
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China
| | - BeiBei Zhi
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China
| | - QuanHui Liu
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China
| | - YingLe Li
- Department of Neurology, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, 210002, Jiangsu, China
| | - LuLu Xiao
- Department of Neurology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China
| | - WuSheng Zhu
- Department of Neurology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China.
| | - GuangMing Lu
- Department of Medical Imaging, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, 210002, Jiangsu, China.
- Department of Medical Imaging, Jinling Hospital, The First School of Clinical Medicine, Southern Medical University, Nanjing, 210002, Jiangsu, China.
| |
Collapse
|
4
|
Jiang S, Liu Q, Zhang C, Chen K, Dou W, Wang X. High-Resolution Vessel Wall MRI in Assessing Postoperative Restenosis of Intracranial Atherosclerotic Disease Before Drug-Coated Balloon Treatment: An Outcome Prediction Study. J Magn Reson Imaging 2022. [PMID: 36259524 DOI: 10.1002/jmri.28490] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/18/2022] [Revised: 09/30/2022] [Accepted: 09/30/2022] [Indexed: 01/01/2023] Open
Abstract
BACKGROUND Postoperative restenosis frequently occurs in intracranial atherosclerotic disease (ICAD) patients after drug-coated balloon (DCB) treatment. However, high-risk plaques associated with postoperative restenosis remain to be explored. PURPOSE To assess whether high-resolution vessel wall MRI (HR-VWI) contributes to the identification of high-risk plaques associated with postoperative restenosis before DCB treatment. STUDY TYPE Retrospective. SUBJECTS A total of 70 patients with ICAD who underwent DCB treatment. FIELD STRENGTH/SEQUENCE 3.0 T; magnetic resonance angiography, HR-VWI. ASSESSMENT All patients underwent HR-VWI examination prior to DCB treatment. Digital subtraction angiography (DSA) measurement was assessed 6 months (±1 month) after operation to determine the vessel restenosis, classifying patients into three groups of no stenosis, mild stenosis (<50%), and restenosis (>50%). Clinical factors and HR-VWI characteristics, including vessel and lumen area at maximal lumen narrowing (MLN), plaque area and length, degree of stenosis, plaque burden, remodeling index, and enhancement amplitude, were compared among three groups. Clinical factors and HR-VWI characteristics were separately evaluated for the association with postoperative restenosis. STATISTICAL TESTS Kolmogorov-Smirnov test, intra-class correlation coefficient, Kruskal Wallis H test, Mann-Whitney U test, receiver operating characteristic (ROC) curve, multivariable linear regression analysis. P-values <0.05 was considered statistically significant. RESULTS During the follow-up DSA measurement, 13 lesions (18.5%) showed restenosis. With HR-VWI, significant differences among three groups were observed in plaque length, lumen area of MLN, degree of stenosis, enhancement amplitude, and plaque burden. In ROC analysis, plaque length (area under the curve [AUC] = 0.809), and enhancement amplitude (AUC = 0.880) provided higher efficacy in identification of high-risk plaques associated with postoperative restenosis than degree of stenosis (AUC = 0.746) and plaque burden (AUC = 0.759). Multivariable linear regression analysis showed that plaque length and enhancement amplitude were independent prognostic factors of postoperative restenosis. DATA CONCLUSION HR-VWI has the potential to identify high-risk plaques in ICAD patients before DCB treatment. LEVEL OF EVIDENCE 4 TECHNICAL EFFICACY: Stage 2.
Collapse
Affiliation(s)
- Shu Jiang
- Shandong Provincial Qianfoshan Hospital, Shandong University, Jinan, China
| | - Qingwei Liu
- Department of Radiology, Shandong Provincial Hospital, Shandong University, Jinan, China
| | - Chao Zhang
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Kunjian Chen
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| | | | - Xinyi Wang
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| |
Collapse
|