Abstract
This study aims to investigate the clinical effectiveness of integrating head and neck computed tomography angiography (CTA) with magnetic resonance diffusion-weighted imaging (MR-DWI) for diagnosing acute ischemic stroke (AIS), compared to using a single imaging modality alone. This retrospective, single-center study included 160 patients with confirmed AIS. Patients were divided into 2 groups: one group underwent a single imaging modality (CTA or MR-DWI), while the other group underwent both CTA and MR-DWI during the same hospitalization, with the results jointly evaluated for diagnostic purposes. Diagnostic indicators, including sensitivity, specificity, accuracy, lesion identification, multifocal lesion detection, vascular occlusion localization, and the correlation between imaging findings and clinical outcomes (National Institutes of Health Stroke Scale and Modified Rankin Scale scores), were assessed and compared between groups. The 2 groups showed no significant differences in baseline characteristics (P > .05). The combined-modality group demonstrated significantly better diagnostic performance, with higher sensitivity (95.0%), specificity (92.5%), and accuracy (93.8%) (P < .05). It also achieved superior lesion detection (98.8%) and multifocal lesion identification (93.8%) compared with the single-modality group. Importantly, imaging findings in the combined group showed stronger correlations with neurological severity (R = 0.78) and short-term prognosis (R = 0.81) (P < .01), and detection of brainstem and deep lesions was markedly improved. The integration of head and neck CTA with MR-DWI significantly enhances diagnostic precision for AIS. This approach improves lesion visualization, vascular occlusion localization, and prediction of clinical outcomes, supporting its broader application in early stroke diagnosis and management.
Keywords: acute ischemic stroke, combined imaging, computed tomography angiography, lesion detection, magnetic resonance diffusion-weighted imaging, prognosis prediction
1. Introduction
Acute cerebral infarction (ACI) is a leading cause of death and long-term disability,[1–3] with rising incidence and disability rates in China that impose a substantial social and economic burden.[4,5] Early and accurate diagnosis – particularly in assessing lesion extent and identifying vascular occlusion sites – is crucial for guiding timely interventions such as thrombolysis and mechanical thrombectomy.[6,7]
Imaging plays an irreplaceable role in this process. Head and neck computed tomography angiography (CTA) rapidly visualizes cerebrovascular structures and occlusion sites,[8,9] but has limited ability to directly detect early infarct lesions. Magnetic resonance diffusion-weighted imaging (MR-DWI) is highly sensitive for detecting infarct lesions,[10] yet it cannot fully evaluate intracranial and extracranial vessels or delineate lesion extent.
Recent studies have suggested that integrating CTA with MR-DWI may leverage the strengths of both modalities, offering more comprehensive evaluation of pathological features in ACI.[11,12] However, prior research has been limited by small sample sizes and inconsistent findings, particularly regarding the detection of multiple lesions, lesion extent, and prediction of vascular recanalization.[13,14]
Therefore, this study investigates the clinical value of integrating head and neck CTA with MR-DWI in diagnosing ACI in a cohort of 160 patients, with a focus on lesion identification, vascular occlusion localization, and diagnostic accuracy, thereby providing more robust evidence for clinical practice.
2. Materials and methods
2.1. Study subjects and sample source
This study was approved by the Ethics Committee of the Traditional Chinese medical hospital of Tongxiang city. This study was a single-center retrospective study conducted in a tertiary hospital’s neurology emergency department. A total of 160 patients with ACI admitted between January 2020 and December 2023 were enrolled. Inclusion criteria were: meeting the diagnostic criteria for ACI as outlined in the 2018 Chinese Guidelines for the Diagnosis and Treatment of Acute Ischemic Stroke; symptom onset ≤24 hours; age between 18 and 80 years; and informed consent obtained from the patients and their families. Exclusion criteria included: patients with cerebral hemorrhage, other types of cerebrovascular diseases, or intracranial tumors; severe liver, kidney, or cardiac dysfunction; contraindications for imaging, such as contrast agent allergy; and incomplete data or poor-quality imaging. Of the enrolled patients, 98 were male, and 62 were female, with a mean age of 64.3 ± 10.5 years.
2.2. Sample size calculation
Based on previous studies, the sensitivity of CTA combined with MR-DWI for diagnosing ACI was estimated at 90%, compared to 80% for CTA alone. Assuming a significance level (α) of 0.05 and power (1−β) of 0.8, the sample size was calculated using the following formula:
The minimum required sample size for each group was calculated to be 72. Accounting for a 10% dropout rate, the total sample size was determined to be 160 patients.
2.3. Grouping strategy
Based on the actual imaging examination methods that patients received in clinical practice, rather than through random grouping, the patients were divided into 2 groups. The single-examination group (n = 80) included patients who only underwent head and neck CTA examination, mainly due to urgent diagnostic needs, contraindications for magnetic resonance imaging (MRI), and other reasons. The combined-examination group (n = 80) included patients who underwent both CTA and MR-DWI examinations, with the aim of conducting a more comprehensive assessment through clinical decision-making. Since this was a retrospective study, the grouping was based on the actual diagnostic path. To address potential selection bias issues, the baseline demographic and clinical characteristics of the 2 groups were compared to address them. The results showed no significant differences (P > .05).
2.4. Detection indicators
Primary detection indicators included: diagnostic sensitivity, specificity, and accuracy; lesion identification rate and detection rate of multiple lesions; accuracy of vascular occlusion localization; and correlation of imaging features with neurological deficits (National Institutes of Health Stroke Scale (NIHSS) score) and prognosis (modified rankin scale [mRS] score).
2.5. Detection methods
In this study, patients underwent head and neck CTA, MR-DWI, or a combination of both imaging modalities to comprehensively assess infarction characteristics and vascular lesions. For CTA, a Siemens SOMATOM Force dual-source computed tomography scanner was used. Patients were positioned supine, and an iodinated contrast agent (350 mg I/mL) was administered at 80 to 100 mL via an elbow vein at a rate of 4.5 mL/s, followed by head and neck CTA scanning. Post-scan image processing included maximum intensity projection (MIP), volume rendering, and curved planar reconstruction to clearly visualize the structure of major intracranial and extracranial vessels and detect potential occlusions or stenoses.
For MR-DWI, a GE Discovery MR750 3.0T MRI scanner was used. Patients were also positioned supine, and routine MRI sequences (T1-weighted, T2-weighted, and FLAIR) were performed, followed by diffusion-weighted imaging (DWI). The DWI parameters were set as TR = 3000 ms, TE = 60 ms, matrix size = 128 × 128, and slice thickness = 5 mm. Apparent diffusion coefficient maps were analyzed to identify the location, extent, and characteristics of infarction lesions, enabling accurate detection of ACI.
For the combined diagnosis, CTA images were integrated with MR-DWI results to enhance diagnostic accuracy and comprehensiveness. The combined evaluation provided both anatomical characteristics of infarction lesions and vascular abnormalities, overcoming the limitations of single imaging modalities and optimizing lesion detection, vascular occlusion localization, and pathophysiological assessments. This comprehensive imaging approach offered more precise and thorough evidence to guide clinical decision-making.
2.6. Statistical analysis
All statistical analyses were performed using SPSS 26.0 software. Continuous variables were expressed as mean ± standard deviation (mean ± SD) and compared using independent-sample t-tests. Categorical variables were expressed as frequencies and percentages and compared using χ2 tests. Multivariable logistic regression was used to analyze the correlation between infarction lesion characteristics and prognosis. A P-value < .05 was considered statistically significant.
3. Results
3.1. Comparison of general clinical characteristics
There were no significant differences between the 2 groups in terms of gender, age, time of onset, or underlying diseases (e.g., hypertension, diabetes) (P > .05), indicating comparability between the groups (Table 1).
Table 1.
Comparison of general clinical characteristics between the 2 groups.
| Clinical characteristics | Single modality group (n = 80) | Combined modality group (n = 80) | Statistic | P-value |
|---|---|---|---|---|
| Male (cases, %) | 48 (60.0%) | 50 (62.5%) | χ2 = 0.10 | .75 |
| Female (cases, %) | 32 (40.0%) | 30 (37.5%) | ||
| Mean age (yr) | 63.8 ± 11.2 | 64.7 ± 10.5 | t = 0.45 | .65 |
| Onset time (h) | 12.4 ± 3.6 | 12.1 ± 3.8 | t = 0.52 | .61 |
| Hypertension (cases, %) | 55 (68.8%) | 57 (71.3%) | χ2 = 0.12 | .73 |
| Diabetes (cases, %) | 28 (35.0%) | 29 (36.3%) | χ2 = 0.02 | .88 |
3.2. Comparison of diagnostic performance
The sensitivity, specificity, and accuracy of ACI diagnosis differed significantly between the single-modality and combined-modality groups (Table 2, Fig. 1). The combined-modality group achieved diagnostic sensitivity, specificity, and accuracy of 95.0%, 92.5%, and 93.8%, respectively, which were significantly higher than the single-modality group’s 83.8%, 81.3%, and 82.5% (P < .05).
Table 2.
Comparison of diagnostic sensitivity, specificity, and accuracy.
| Group | Sensitivity (%) | Specificity (%) | Accuracy (%) | AUC |
|---|---|---|---|---|
| Single modality group | 83.8 | 81.3 | 82.5 | 0.865 |
| Combined modality group | 95.0 | 92.5 | 93.8 | 0.945 |
| P-value | <.05 | <.05 | <.05 | <.05 |
AUC = area under curve.
Figure 1.
Comparative analysis of CTA and MR-DWI illustrating the improved diagnostic accuracy with combined application. (A) M1 distal occlusion; (B) 3D view of M1 distal occlusion; (C) Infarction not evident on plain CT scan; (D) CTA showing occlusion of the right vertebral artery at the V1 segment; (E) Infarction undetectable on plain CT scan; (F) MR-DWI revealing infarction in the cerebellar vermis. CT = computed tomography, CTA = computed tomography angiography, MR-DWI = magnetic resonance diffusion-weighted imaging.
3.3. Comparison of lesion identification and detection rates of multiple lesions
The overall lesion identification rate in the combined-modality group was 98.8%, significantly higher than the single-modality group’s 87.5% (P < .05). Regarding the detection of multiple lesions, the detection rate in the combined-modality group was 93.8%, markedly higher than the single-modality group’s 76.3% (P < .05, Table 3).
Table 3.
Comparison of lesion detection rate and multifocal lesion identification rate.
| Group | Lesion detection rate (%) | Multifocal lesion identification rate (%) |
|---|---|---|
| Single modality group | 87.5 | 76.3 |
| Combined modality group | 98.8 | 93.8 |
| P-value | <.05 | <.05 |
3.4. Comparison of accuracy in vascular occlusion localization
In evaluating the accuracy of vascular occlusion localization, the combined-modality group outperformed the single-modality group (Table 4). The combined-modality group achieved localization accuracies of over 95% for internal carotid artery occlusion, middle cerebral artery occlusion, and vertebrobasilar artery occlusion, while the single-modality group achieved only 84.6% accuracy for middle cerebral artery occlusion (P < .05).
Table 4.
Comparison of accuracy in vessel occlusion localization.
| Occlusion site | Single modality group accuracy (%) | Combined modality group accuracy (%) | P-value |
|---|---|---|---|
| Internal carotid artery | 91.3 | 97.5 | .02 |
| Middle cerebral artery | 84.6 | 96.3 | <.01 |
| Vertebrobasilar artery | 89.8 | 95.0 | .04 |
3.5. Correlation between imaging features and neurological deficits
A comparison of the correlation between imaging features and NIHSS scores revealed that in the combined-modality group, lesion volume was significantly positively correlated with NIHSS scores (R = 0.78, P < .01), while the correlation was weaker in the single-modality group (R = 0.65, P < .05). Furthermore, the combined-modality group demonstrated significantly higher accuracy in predicting vascular recanalization (Table 5).
Table 5.
Correlation between imaging features and NIHSS scores.
| Group | Correlation coefficient (r) | P-value |
|---|---|---|
| Single modality group | 0.65 | <.05 |
| Combined modality group | 0.78 | <.01 |
NIHSS = National Institutes of Health Stroke Scale.
3.6. Correlation between imaging features and prognosis
Using mRS scores to assess short-term prognosis, imaging features (e.g., lesion size, location, type of vascular occlusion) in the combined-modality group showed a significant correlation with mRS scores (R = 0.81, P < .01). Imaging features in the combined-modality group were more precise in predicting patients’ short-term prognosis (Table 6).
Table 6.
Correlation between imaging features and mRS scores.
| Group | Correlation coefficient (r) | P-value |
|---|---|---|
| Single modality group | 0.69 | <.05 |
| Combined modality group | 0.81 | <.01 |
mRS = modified rankin scale.
3.7. Detection of different lesion types and locations
The combined-modality group demonstrated significantly higher detection rates for deep lesions, subcortical lesions, and brainstem lesions compared to the single-modality group (P < .05). Additionally, the combined-modality group exhibited a clear advantage in identifying microlesions in the brainstem, achieving a detection rate of 96.3% (Table 7).
Table 7.
Comparison of detection rates for different lesion types and locations.
| Lesion type or location | Single modality group detection rate (%) | Combined modality group detection rate (%) | P-value |
|---|---|---|---|
| Deep lesions | 88.8 | 97.5 | .03 |
| Subcortical lesions | 85.0 | 96.3 | .01 |
| Brainstem lesions | 81.3 | 96.3 | <.01 |
4. Discussion
Acute ischemic stroke (AIS) is a major disease that threatens human health, and its rapid diagnosis and precise treatment are critical for improving patient outcomes. This study analyzed the diagnostic efficacy of head and neck CTA combined with MR-DWI sequences in AIS. The results demonstrated that the combined examination outperformed single examinations in lesion detection, vascular occlusion localization, and prognosis evaluation, highlighting the potential value of combining these imaging techniques.
4.1. Diagnostic advantages of combined examination
The findings showed that head and neck CTA combined with MR-DWI exhibited significantly higher sensitivity, specificity, and accuracy in diagnosing AIS compared to single examination groups. CTA quickly reveals the vascular anatomy and occlusion status of the head and neck, compensating for the limitations of DWI in providing vascular lesion information. MR-DWI, on the other hand, has high sensitivity in detecting brain tissue lesions, especially for identifying early ischemic changes. The combination of these modalities not only accurately locates lesions but also comprehensively evaluates lesion size, distribution, and potential vascular abnormalities, providing more reliable imaging evidence for clinical decision-making.
4.2. Improved lesion detection rate and vascular occlusion localization
The study revealed that the combined examination group achieved significantly higher lesion detection and multifocal lesion identification rates than single examination groups. This advantage was particularly evident in the detection of small and deep brain tissue lesions. This improvement is likely due to MR-DWI’s high resolution for ischemic lesions and CTA’s precise visualization of vascular abnormalities.[15] Furthermore, the combined examination demonstrated superior accuracy in vascular occlusion localization, especially in complex conditions such as internal carotid artery, middle cerebral artery, and vertebrobasilar artery occlusions.[16,17] Accurate vascular occlusion localization is crucial for guiding revascularization therapy, and the advantages of combined examination provide essential support for optimizing clinical treatment plans.[18]
Importantly, the improved detection of deep and brainstem lesions has direct clinical implications. Brainstem infarctions are frequently associated with severe neurological dysfunctions such as impaired consciousness, respiratory compromise, and dysphagia, which require urgent management and close monitoring. Likewise, accurate recognition of subcortical and deep lesions is essential for predicting motor and sensory deficits, guiding early rehabilitation strategies, and tailoring secondary prevention. Therefore, beyond enhancing diagnostic accuracy, the integrated approach provides clinically actionable information that supports individualized treatment and improves patient outcomes.
4.3. Prognostic value of combined examination
This study further explored the correlation between imaging features and patient prognosis. The results indicated that lesion size, location, and vascular occlusion type identified by the combined examination were significantly correlated with neurological deficits (NIHSS scores) and short-term outcomes (mRS scores). The combined examination provided more precise predictions of disease progression and prognosis, which is crucial for developing individualized treatment strategies.[19] For instance, patients with larger lesion volumes or involvement of critical functional areas may have poorer prognoses, and combined examination can offer imaging-based support for early intervention in these high-risk patients.[20–22]
4.4. Limitations of combined examination
Despite the significant advantages of the integrated use of CTA and MR-DWI demonstrated in this study, certain limitations should be acknowledged.[23,24] First, the integrated examination involves higher time and financial costs compared to single examinations, which may limit its applicability in primary healthcare settings. Second, CTA requires the use of contrast agents that may cause allergic reactions or renal impairment, restricting its suitability for specific patient populations. In addition, MR-DWI requires a longer scanning time and patient cooperation, which can be challenging for critically ill patients in the acute phase.
Beyond these technical considerations, several methodological limitations inherent to the retrospective design must be noted. Patient selection bias may have been introduced, as individuals with MRI contraindications or incomplete/poor-quality imaging data were excluded. This exclusion could affect the generalizability of our findings to broader AIS populations. Information bias is also possible due to variability in the completeness and quality of clinical records. Furthermore, although imaging interpretation followed standardized protocols, it was not conducted in a fully blinded manner, which may introduce subjective bias. With respect to missing data, cases with incomplete records were excluded from analysis to maintain data reliability, but this approach may have further contributed to selection bias.
Finally, the external validity of our findings may be limited. As this study was conducted in a single tertiary hospital with advanced imaging facilities, the results may not fully translate to primary care or smaller hospitals where MR-DWI may be less accessible, and resource constraints could affect the feasibility of implementing integrated CTA and MR-DWI. Therefore, future prospective, multicenter studies across diverse healthcare settings are needed to confirm the broader applicability of this approach and to explore strategies for adapting it to environments with limited resources.
4.5. Comparison with previous studies
Compared to previous studies focusing on single CTA or MR-DWI in AIS diagnosis, this study further confirmed the advantages of combining the 2 techniques. For example, some studies have shown that CTA has high sensitivity in assessing vascular lesions but limited ability to detect small lesions, while MR-DWI is sensitive to lesion detection but lacks detailed vascular anatomical information.[25,26] By integrating these 2 modalities, this study addresses the shortcomings of single examinations and provides important insights for optimizing imaging strategies.[27]
4.6. Clinical significance and future prospects
The combined use of head and neck CTA with MR-DWI demonstrates not only high efficacy in diagnosing AIS but also significant clinical implications. In AIS management, “time is brain,” and rapid, accurate diagnosis facilitates early implementation of revascularization therapy, reducing disability and mortality rates. The results of this study suggest that combined examination can serve as an important imaging strategy for AIS diagnosis. In the future, artificial intelligence (AI) technology could be incorporated to deeply analyze CTA and MR-DWI imaging data, further improving diagnostic efficiency and accuracy. Moreover, larger multicenter studies are needed to validate the clinical efficacy of the combined examination and explore its applications in guiding treatment strategies and improving prognosis.
In conclusion, head and neck CTA combined with MR-DWI sequences demonstrates significant advantages in AIS diagnosis, providing essential support for precision medicine in acute stroke care. However, further optimization of its application strategies is needed to address cost and suitability for specific populations, promoting its widespread use in clinical practice.
Author contributions
Conceptualization: Zhen Shen, Deming Sun, Jingen Chai.
Data curation: Zhen Shen, Deming Sun, Jingen Chai, Xianfeng Shao.
Formal analysis: Zhen Shen, Deming Sun, Jingen Chai, Xianfeng Shao.
Investigation: Zhen Shen.
Methodology: Zhen Shen.
Software: Jingen Chai.
Validation: Zhen Shen, Deming Sun, Xianfeng Shao, Xianchao Feng.
Visualization: Zhen Shen, Deming Sun, Xianfeng Shao, Xianchao Feng.
Writing – original draft: Zhen Shen, Deming Sun, Jingen Chai, Xianfeng Shao, Xianchao Feng.
Writing – review & editing: Zhen Shen, Deming Sun, Jingen Chai, Xianfeng Shao.
Abbreviations:
- ACI
- acute cerebral infarction
- AIS
- acute ischemic stroke
- CTA
- computed tomography angiography
- MR-DWI
- magnetic resonance diffusion-weighted imaging
- MRI
- magnetic resonance imaging
- mRS
- modified rankin scale
- NIHSS
- National Institutes of Health Stroke Scale
The authors have no funding and conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
How to cite this article: Sun D, Chai J, Shao X, Feng X, Shen Z. Enhanced diagnostic value of sequential head and neck CTA and MR-DWI in acute ischemic stroke. Medicine 2025;104:52(e45967).
Contributor Information
Deming Sun, Email: sundeming9699@163.com.
Jingen Chai, Email: 947309852@qq.com.
Xianfeng Shao, Email: 470402569@qq.com.
Xianchao Feng, Email: sunzhanhe-love@163.com.
References
- [1].Maimaitiming M, Li S, Huang K, et al. Risk factors for cardiogenic shock incidence and mortality after acute myocardial infarction: a systematic review and meta-analysis. Commun Med (Lond). 2025;5:200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Herpich F, Rincon F. Management of acute ischemic stroke. Crit Care Med. 2020;48:1654–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Fan Y, Chen F, Yan S, et al. Clinical study of the prognostic value of the Essen score for acute cerebral infarction. Ir J Med Sci. 2024;193:2495–500. [DOI] [PubMed] [Google Scholar]
- [4].Wang YJ, Li ZX, Gu HQ, et al. ; China Stroke Statistics 2019 Writing Committee. China stroke statistics 2019: a report from the National Center for Healthcare Quality Management in neurological diseases, China National Clinical Research Center for neurological diseases, the Chinese Stroke Association, National Center for Chronic and Non-communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention and Institute for Global Neuroscience and Stroke Collaborations. Stroke Vasc Neurol. 2020;5:211–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Lai X, Rong H, Ma X, et al. The economic burden of influenza-like illness among children, chronic disease patients, and the elderly in China: a National Cross-sectional Survey. Int J Environ Res Public Health. 2021;18:6277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Antman EM, Anbe DT, Armstrong PW, et al. ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1999 Guidelines for the Management of Patients with Acute Myocardial Infarction). ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction – executive summary: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (writing committee to revise the 1999 guidelines for the management of patients with acute myocardial infarction). Circulation. 2004;110:588–636. [DOI] [PubMed] [Google Scholar]
- [7].Bajaj A, Sethi A, Rathor P, Suppogu N, Sethi A. Acute complications of myocardial infarction in the current era: diagnosis and management. J Investig Med. 2015;63:844–55. [DOI] [PubMed] [Google Scholar]
- [8].Chen G, Zhou C, Yang Y. Carotid artery CT angiography. In: Whole Body Computed Tomography Angiography. Springer Nature; 2024:161–89. [Google Scholar]
- [9].Wu C, Xu X, Wang R. Application of CT angiography in the diagnosis of acute cerebrovascular disease in neurology. J Med Imaging Health Inf. 2021;11:878–85. [Google Scholar]
- [10].Cui L, Fan Z, Yang Y, et al. Deep learning in ischemic stroke imaging analysis: a comprehensive review. Biomed Res Int. 2022;2022:2456550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Wang C, Shi Z, Yang M, et al. Deep learning-based identification of acute ischemic core and deficit from non-contrast CT and CTA. J Cereb Blood Flow Metab. 2021;41:3028–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].JUSTIFIED O U R R B. 14th World Stroke Congress, Singapore, 26–29 October 2022. Int J Stroke. 2022;17(3S):3–288. [Google Scholar]
- [13].Harrison KA. Factors Affecting Focal Cerebral Ischemic Infarction: Inferences from Experimental Modelling in Macaques and Murines. Queen’s University; 2022. [Google Scholar]
- [14].Kattah JC, Talkad AV, Wang DZ, Hsieh YH, Newman-Toker DE. HINTS to diagnose stroke in the acute vestibular syndrome: three-step bedside oculomotor examination more sensitive than early MRI diffusion-weighted imaging. Stroke. 2009;40:3504–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Hu X, Yuan Z, Liang K, et al. Application of spinal subtraction and bone background fusion CTA in the accurate diagnosis and evaluation of spinal vascular malformations. AJNR Am J Neuroradiol. 2024;45:351–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Brugnara G, Baumgartner M, Scholze ED, et al. Deep-learning based detection of vessel occlusions on CT-angiography in patients with suspected acute ischemic stroke. Nat Commun. 2023;14:4938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Hu T, Ling R, Zhu Y. Advancements in imaging of intracranial atherosclerotic disease: beyond the arterial lumen to the vessel wall. Rev Neurosci. 2024;36:229–41. [DOI] [PubMed] [Google Scholar]
- [18].Beckman JA, Schneider PA, Conte MS. Advances in revascularization for peripheral artery disease: revascularization in PAD. Circ Res. 2021;128:1885–912. [DOI] [PubMed] [Google Scholar]
- [19].Tran KA, Kondrashova O, Bradley A, Williams ED, Pearson JV, Waddell N. Deep learning in cancer diagnosis, prognosis and treatment selection. Genome Med. 2021;13:1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Liu S, Liu C, Pan H, et al. Magnetic resonance imaging-based nomograms predict high-risk cytogenetic abnormalities in multiple myeloma: a two-centre study. Clin Radiol. 2025;82:106768. [DOI] [PubMed] [Google Scholar]
- [21].Mallio CA, Vertulli D, Di Gennaro G, et al. Relationship between DWI-based acute ischemic stroke volume, location and severity of dysphagia. Brain Sci. 2024;14:1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Parihar AS, Dehdashti F, Wahl RL. FDG PET/CT–based response assessment in malignancies. Radiographics. 2023;43:e220122. [DOI] [PubMed] [Google Scholar]
- [23].Koeppel DR, Boehm IB. Shortage of iodinated contrast media: status and possible chances – a systematic review. Eur J Radiol. 2023;164:110853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Jost G, McDermott M, Gutjahr R, Nowak T, Schmidt B, Pietsch H. New contrast media for K-edge imaging with photon-counting detector CT. Invest Radiol. 2023;58:515–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Guggenberger K, Krafft AJ, Ludwig U, et al. High-resolution compressed-sensing T1 black-blood MRI: a new multipurpose sequence in vascular neuroimaging? Clin Neuroradiol. 2021;31:207–16. [DOI] [PubMed] [Google Scholar]
- [26].Zhao H, Li C, Jin Z, et al. Risk prediction of preoperative acute ischemic stroke in acute type A aortic dissection. Eur Radiol. 2023;33:7250–9. [DOI] [PubMed] [Google Scholar]
- [27].Najjar R. Redefining radiology: a review of artificial intelligence integration in medical imaging. Diagnostics (Basel). 2023;13:2760. [DOI] [PMC free article] [PubMed] [Google Scholar]