Nomogram of Intracranial Artery Calcification with Integration of CT-based Radiomics to Identify Culprit Lesions

Nomogram of Intracranial Artery Calcification with Integration of CT-based Radiomics to Identify Culprit Lesions | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Nomogram of Intracranial Artery Calcification with Integration of CT-based Radiomics to Identify Culprit Lesions Huan Yang, Yunchao Chen, Teng Ma, Jizhen Feng, Chencui Huang, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7763910/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background : To develop a machine learning model of intracranial artery calcification (IAC) based on computed tomography (CT) images and assess its value for improved identifying culprit lesions responsible for acute/subacute ischemic cerebral infarction (ASCI). Methods: Patients with intracranial atherosclerotic diseases in the vertebrobasilar artery or intracranial internal carotid artery who underwent vessel wall MRI and head CT examinations at two hospital centers were retrospectively assessed. Each calcified plaque was classified by the likelihood of having caused an ASCI as culprit or non-culprit. Machine learning technique was utilized to automatically select twenty top-ranked features from IAC segmentation and build a model using the logistic regression algorithm with fivefold stratified cross-validation. The added values of radiomic-based score (Radscore) to stenosis and clinical risk factors for identification of culprit lesions were evaluated using area under the curve (AUC). A nomogram was constructed by integrating the Radscore with clinical and imaging covariates. Results: One hundred and thirty-three ASCI patients with culprit plaques were identified in the training set (totally 282 patients), and 36 were identified in the external test set (totally 71 patients). Diabetes, smoking, coronary heart disease, and stenosis were found to be associated with the culprit lesions in the multivariate analysis. The diagnostic performance of Radscore was 0.674 and 0.609 for the training and external test data set. The nomogram, which includes clinical factors, stenosis, spotty calcification, and Radscore, demonstrated moderate values for the discrimination of symptomatic intracranial atherosclerotic lesions, with an AUC of 0.749 in the training set and an AUC of 0.736 in the external test set. Conclusion: Radiomics of intracranial artery calcification in the culprit lesion may provide added value for identifying ASCI beyond stenosis and clinical factors. The nomogram incorporating both conventional and radiomics variables may serve as a potential diagnostic tool for stroke risk assessment in clinical settings. Trial registration: Retrospectively registered. Radiomics intracranial atherosclerosis calcification computed tomography magnetic resonance imaging Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 BACKGROUND Intracranial artery calcification (IAC) is a frequent imaging finding on computed tomography (CT) scans, with incidence rates reaching 100% in individuals aged 90 and older( 1 ). The presence of IAC indicates a correlation with atherosclerosis sharing similar cardiovascular risk factors( 2 ). A significant IAC burden is associated with an increased risk of stroke across all ethnic groups( 3 ). It is widely acknowledged that coronary artery calcification in culprit coronary lesions, as verified by the histological evidence, could be used as a predictor for myocardial infarction( 4 ). Given the systemic and divergent nature of atherosclerosis, there is considerable interest in understanding the clinical implications of IAC in culprit lesions. The unstable effects of calcification on atherosclerotic plaques arise from inflammation and heightened biomechanical forces. Spotty calcification is recognized as a biomarker of plaque vulnerability in coronary arteries and is closely linked to plaque rupture( 5 ). Recent studies have indicated that the presence or number of spotty calcifications in carotid and intracranial arteries is associated with an increased risk of stroke( 6 ). However, this smallest calcification unit discernible on CT images provides limited diagnostic information, primarily regarding morphology. Additionally, visual assessments of calcification heavily rely on radiologists' interpretations and experience. Facilitating objective quantification of this type of radiologic data beyond human visual capabilities is crucial. Radiomics entails extracting extensive data from gray-level images, including features related to first-order statistics, shape, texture, and transform-based characteristics. We previously showed that texture analysis of intracranial artery calcification offered more information than the visual assessments and held significant potential for identifying culprit plaques( 7 ). Machine learning (ML) could transform complex, heterogeneous pathology into a straightforward radiomic-based parameter. A radiomic-based score (Radscore) for cardiac CT significantly enhanced the predictive accuracy compared to the conventional Agatston calcium scores in the community-based Framingham Heart Study( 8 ). Deep learning of coronary artery calcium scores from SPECT/CT adds significant values to SPECT myocardial perfusion alone for risk assessment( 9 ). We hypothesized that the artificial intelligence of IAC can provide additional value in stratifying plaque instability beyond traditional assessments. High-resolution vessel wall magnetic resonance imaging (VW-MRI) significantly improved our understanding of intracranial atherosclerosis as a primary cause of ischemic stroke. VW-MRI enables the assessment of plaque morphology and compositional characterization, i.e., intraplaque hemorrhage and fibrous cap( 10 ). Nevertheless, interpreting IAC can be challenged on routine MRI due to insufficient resolution to detect the complex signals surrounding calcium deposits. CT is the preferred imaging modality for visualizing calcifications and the first-line diagnostic tool for stroke patients. Combining CT and VW-MRI images could comprehensively characterize intracranial atherosclerosis and accompanied calcifications. The presence of vulnerable plaques in high-risk patients increases the likelihood of adverse cardiovascular events( 11 ). A nomogram is commonly designed for individual patient profiles to estimate the probability of clinical events using an intuitive graphical interface. The study aimed to develop a ML model using CT images of IAC and assess the diagnostic value of the incorporated nomogram in identifying culprit intracranial atherosclerotic lesions, using VW-MRI as the reference standard. Materials and methods Study Population The research was conducted in accordance with the Declaration of Helsinki. Clinical trial number: not applicable. The cohort for this study was drawn from patients who underwent VW-MRI and head CT scans within two weeks at two medical centers from January 2020 to August 2024. Patients were included if there were 30% to 99% atherosclerotic stenosis in the intracranial internal carotid artery (iICA) or vertebrobasilar artery (VBA) on VW-MRI images(12) and the concurrent presence of calcification on CT images. Exclusion criteria were (i) evidence of non-atherosclerotic intracranial vascular pathology (e.g., cardiac embolism, dissection, vasculitis, aneurysm, moya-moya disease); (ii) a history of stent or treatment of the target vessel; and (iii) inadequate image quality, incomplete clinical data or insufficient region of interest (ROI) for the radiomic analysis. The Institutional Review Boards approved this study, and the requirement for informed consent were waived. The flow chart of patient selection is shown in Figure 1 . MRI Examination MRI exams were performed on 3T MR imaging scanners (Ingenia, Philips Healthcare; MAGNETOM Prisma, Siemens Healthineers) with either an 8-channel or 16-channel head coil. The protocols of MRI scans are shown in Table 1 . A plaque was considered the culprit if the only or most stenotic lesion within the artery territory of acute/subacute cerebral infarction (ASCI)(13). A plaque was considered as non-culprit if it was the most stenotic lesion in non-ASCI patients. Luminal stenosis was measured using criteria established in the Warfarin-Aspirin Symptomatic Intracranial Disease trial(14) on pre-contrast VW-MRI images. Table 1. MRI Sequence Parameters Scan parameter PHILIPS SIEMENS T1-VISTA DWI T1-SPACE DWI FOV (mm 2 ) 220 × 220 250 × 250 160 × 160 220 × 220 Acquisition orientation Coronal/axial Axial Sagittal Axial Acquired spatial resolution (mm 3 ) 0.7 × 0.7 × 1.1 1.2× 1.2 × 5.0 0.6 × 0.6 × 0.6 1.1× 1.1 × 5.0 Matrix 316 × 312 152 × 122 256 × 256 192 x 192 TR/TE (ms) 425/19 2250/70 800/4.8 2800/55 b value (mm 2 /s) - 0/1000 - 0/1000 Echo spacing (ms) 6.3 1.08 4.76 0.32 Number of averages 2 1 1 1 Parallel acceleration factor 2 2 2 2 Acquisition time (min) 6.1 1.04 8.05 1.17 DWI, diffusion-weighted imaging; FOV, field-of-view; TE, echo time; TR, repetition time; T1-VISTA, T1-weighted volumetric isotropic turbo spin-echo acquisition; T1-SPACE, T1-sampling perfection with application-optimized contrasts using different flip angle evolutions. CT Imaging The CT scans were conducted on 128-slice dual-source CT scanners (SOMATOM Force, Siemens Healthineers; SOMATOM Definition Flash, Siemens Healthineers). The parameters of head CT scanning were as follows: tube voltage of 120 kVp, reconstructed slice thickness of 1 mm, reconstructed slice interval of 0.5 mm and scan range from the foramen magnum to the top of the skull. Image quality was evaluated based on the presence and severity of artifacts, including beam hardening, photon starvation, and noise, using a three-point scale: poor, adequate, or excellent. Images were determined to be of adequate or excellent quality for analysis. Calcification refers to areas with increased density having a CT attenuation value of 130 Hounsfield units (HU) or higher. Spotty calcification refers to calcium deposits that are less than 3 mm in length and are contained within an arc measuring less than 90 degrees. Two radiologists independently interpreted all CT and VW-MRI images using specific anatomical landmarks (e.g., carotid siphon, vertebrobasilar junction). The disagreement in data annotation was resolved by consensus. ML of Calcification Features All original CT images were stored in Neuroimaging Informatics Technology Initiative (NIfTI) format and loaded into Medical Imaging Interaction Toolkit (MITK, open-source software, https://www.mitk.org). Calcification radiomic analyses were performed by a radiologist with five years of experience in neuroradiology, who was blind to the clinical information. The ROI was delineated manually slice by slice in the axial plane, and this process was cross-checked with VW-MRI. In the presence of multiple calcifications, each must be distinctly delineated. Meanwhile, a radiologist with ten years of expertise in VW-MRI verified the delineation of the ROI. When differing opinions were presented, the final ROI was determined after discussion. The segmentations were used as mask images to isolate voxels containing calcifications in the corresponding plaque lesion. The integrated ML platform (Deepwise Multimodal Research Platform version 2.6.0) was designed to analyze annotated medical data. The radiomic features extracted from the ROIs were as follows: 18 first-order features; 14 size and shape features; 75 texture features; 1111 transform-based features, including Laplacian of Gaussian filtering and Wavelet features. The following steps were performed to select features for identifying culprit lesions. Firstly, each feature group was standardized by the z-score normalization. Secondly, Spearman's correlation coefficient was calculated to determine the correlation between features. If the correlation coefficient between any two independent variables exceeded 0.9, one of the features would be eliminated to reduce redundancy. Thirdly, the F-test was conducted on the remaining features, and twenty features with the lowest p-values were selected as effective predictors for the subsequent logistic regression. Based on ranking, this selection process ensured that the most informative features were retained. The radiomics model was developed using a logistic regression classifier. To ensure a valid and robust model based on the data, fivefold cross-validation was used， which entails dividing the dataset into five mutually exclusive subsets of roughly equal size. The final result is obtained by averaging the outcomes from all five training iterations. Figure 2 shows the workflow of calcification radiomic analysis. Univariate analysis of clinical factors, including age, sex, risk factors (diabetes mellitus, hypertension, smoking, hyperlipidemia and coronary heart disease) and medication uses, stenosis, spotty calcification, and Radscore, was performed. A multivariate logistic regression included variables with p-values less than 0.2 to develop a radiomic nomogram. The diagnostic performance of the nomogram was assessed using the area under the receiver operating characteristic curve (AUC). Statistical Analysis Categorical data were presented as frequencies, while continuous data were presented as either means ± standard deviations or medians with interquartile ranges. The differences between various groups were analyzed using the Student's t-test, the Mann-Whitney U test for continuous variables, and the Chi-squared (χ²) test for categorical variables. The univariate and multivariate logistic regression analyses were conducted to determine covariates associated with culprit lesions. The diagnostic performance of all models was assessed using AUC, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. A two-tailed p-value of less than 0.05 was considered statistically significant. Statistical analyses were performed using the Deepwise Multimodal Research Platform. RESULTS Clinical and Lesion Characteristics Between ASCI and non-ASCI Groups Table 2 presents the clinical and lesion characteristics. A total of 282 patients (195 male; mean age, 62.07 ± 9.12 years) were included in the training set, of which 133 (47.16%) were ASCI patients and 149 (52.84%) were non-ASCI patients. Age and sex were similar between ASCI and non-ASCI groups. The ASCI group exhibited a greater degree of stenosis (p<0.001) and stenosis grade (p=0.005) compared to the non-ASCI group. In the visual assessment of calcification characteristics, the ASCI group exhibited a higher prevalence of spotty calcification (45.86% vs. 30.20%, p=0.007) than the non-ASCI group. Among the 71 patients (52 male; mean age, 61.42 ± 9.62 years) included in the external test set, 36 (50.70%) had ASCI. Multivariate regression analysis showed that diabetes, smoking, coronary heart disease, and stenosis were associated with ASCI, with odds ratios of 2.37, 1.99, 3.05, and 1.67, respectively ( Table 3 ). The parameter of spotty calcification was insignificant (p=0.152). Table2. Clinical and Imaging Characteristics between ASCI and Non-ASCI Groups Training set P value External test set P value ASCI (n=133) Non-ASCI (n=149) ASCI (n=36) Non-ASCI (n=35) Age, mean±SD, year 61.47±9.91 62.61±8.36 0.297 60.42±10.02 62.46±9.22 0.376 Sex 0.790 0.381 Male 93(69.92) 102(68.46) 28(77.78) 24(68.57) Female 40(30.08) 47(31.54) 8(22.22) 11(31.43) DM 0.052 0.186 Yes 77(57.89) 69(46.31) 19(52.78) 13(37.14) No 56(42.11) 80(53.69) 17(47.22) 22(62.86) HTN 0.326 0.885 Yes 104(78.20) 109(73.15) 25(69.44) 25(71.43) No 29(21.80) 40(26.85) 11(30.56) 10(28.67) Smoking 0.112 0.275 Yes 55(41.35) 48(32.21) 18(50.00) 13(37.14) No 78(58.65) 101(67.79) 18(50.00) 22(62.86) HLD 0.482 0.211 Yes 36(27.07) 46(30.87) 5(13.89) 9(25.71) No 97(72.93) 103(69.13) 31(86.11) 26(74.29) CHD 0.127 0.417 Yes 43(32.33) 36(24.16) 9(25.00) 6(17.14) No 90(67.67) 113(75.84) 27(75.00) 29(82.86) History of statin use 0.264 0.350 Yes 39(29.32) 53(35.57) 6(16.67) 9(25.71) No 94(70.68) 96(64.43) 30(83.33) 26(74.29) History of aspirin use 0.241 0.168 Yes 42(31.58) 57(38.26) 8(22.22) 13(37.14) No 91(68.42) 92(61.74) 28(77.78) 22(62.86) Plaque location 0.734 0.614 Anterior circulation 84(63.16) 97(65.10) 33(94.37) 34(97.14) Posterior circulation 49(36.84) 52(34.90) 3(5.63) 1(2.86) Stenosis degree, median (IQR) 75.00(64.52-83.05) 68.00(54.98-77.34) ＜0.001 83.21(67.5-89.27) 70.00(49.00-85.00) 0.013 Stenosis grade 0.005 0.132 ＜30% 7(5.26) 18(12.08) 2(5.56) 9(25.72) 30%-69% 40(30.07) 63(42.28) 7(19.44) 6(17.14) 70%-89% 62(46.62) 55(36.91) 18(50.00) 14(40.00) ≥90% 24(18.05) 13(8.73) 9(25.00) 6(17.14) Spotty calcification 0.007 0.013 Yes 61(45.86) 45(30.20) 25(69.44) 14(40.00) No 72(54.14) 104(69.80) 11(30.56) 21(60.00) ASCI, acute/subacute cerebral infarction; CHD, coronary heart disease; DM, diabetes mellitus; HLD, hyperlipidemia; HTN, hypertension; IQR, interquartile range; SD, standard deviation. Table3. Univariable and Multivariable Regression Analysis for Culprit Lesions Univariate Multivariate OR 95%CI P value OR 95%CI P value Age 0.986 0.961-1.012 0.296 Sex 1.071 0.645-1.778 0.790 DM 1.594 0.995-2.554 0.052 2.374 1.362-4.139 0.002 HTN 1.316 0.761-2.277 0.326 Smoking 1.484 0.912-2.415 0.112 1.993 1.098-3.401 0.022 HLD 0.831 0.496-1.393 0.483 CHD 1.500 0.890-2.528 0.128 3.049 1.617-5.750 0.001 History of statin use 0.752 0.455-1.241 0.265 0.735 0.413-1.310 0.297 History of aspirin use 0.745 0.455-1.219 0.242 Stenosis 1.710 1.266-2.308 ＜0.001 1.671 1.198-2.332 0.003 Spotty calcification 1.958 1.201-3.191 0.007 1.508 0.859-2.649 0.152 Radscore (per 0.1 increase) 2.411 1.716-3.387 ＜0.001 2.899 1.957-4.294 ＜0.001 CHD, coronary heart disease; CI, confidence interval; DM, diabetes mellitus; HLD, hyperlipidemia; HTN, hypertension; OR, odds ratio. Radiomic Analysis Of the 1,218 radiomic features, 916 were removed due to a linear correlation coefficient threshold of 0.9. The remaining 302 features were retained using the F-test. Finally, the logistic regression method selected the twenty most significant features with the lowest p-value to build the prediction model. Figure 3 shows the results of feature weights with fivefold cross-validation. For the finally selected 20 features, 4 (20%) were first order, 12 (60%) were grey level co-occurrence matrix, 2 (10%) were gray-level dependence matrix, 1 (5%) were grey level run length matrix and 1 (5%) was gray-level size zone matrix parameters. ML Models for the Identification of A S CI The diagnostic performance of ML models is shown in Table 4 . The ROC curve showed an AUC of 0.674 (95% confidence interval [CI], 0.612-0.736) for the radiomic model in the training set, with an AUC of 0.609 (95% CI, 0.543-0.674) in the external test set. For the training set, the threshold was set at 0.476, yielding an accuracy of 0.606, a sensitivity of 0.526, a specificity of 0.678, and a Brier Score of 0.227. The fivefold cross-validation resulted in an accuracy of 0.617, a sensitivity of 0.504, a specificity of 0.718, and a Brier Score of 0.236 in the validation set. Table 4. Diagnostic Performances of Models to Identify Culprit Lesions Sensitivity Specificity Accuracy AUC (95%CI) NPV PPV Clinic model Training set 0.790 0.362 0.525 0.637(0.573-0.702 0.659 0.525 External test set 0.782 0.349 0.517 0.556(0.489-0.623) 0.642 0.517 Clinic+Stenosis model Training set 0.714 0.550 0.586 0.665(0.602-0.728) 0.683 0.586 External test set 0.714 0.571 0.638 0.648(0.584-0.712) 0.691 0.598 Nomogram model Training set 0.722 0.644 0.642 0.749(0.693-0.805) 0.722 0.644 External test set 0.714 0.624 0.629 0.736(0.678-0.793) 0.710 0.629 AUC, area under the curve; CI, confidence internal; NPV, negative predictive value; PPV, positive predictive value. A radiomics nomogram combined with clinical factors, stenosis, spotty calcification, and Radscore was established ( Figure 4 ). The Radscore accounted for the largest proportion among all the clinical-radiologic features. As shown in Figure 5, incorporating stenosis into clinical factors slightly increased the AUC for identifying culprit lesions compared with clinical factors alone. The nomogram with integration of the Radscore provided additional value beyond stenosis and clinical factors in the training set (AUC: 0.749, 95% CI: 0.693-0.805 vs. 0.665, 95% CI: 0.602-0.728, p=0.001) and in the external test set (AUC:0.736, 95%CI: 0.678-0.793 vs. 0.648, 95% CI: 0.584-0.712, p=0.001). DISCUSSION This study aimed to evaluate the effectiveness of ML models, developed by the most relevant radiomic features of calcified intracranial atherosclerotic lesions on CT images, to discriminate lesions that have caused acute/subacute infarctions. The association between Radscore and symptomatic intracranial atherosclerotic lesions was statistically significant, with an odds ratio of 2.9 for every 0.1 increase in the Radscore. The ML model based on IAC features on CT images offered additional value beyond stenosis and clinical variables for identifying symptomatic lesions. Moreover, the final nomogram incorporated patient-specific risk factors, stenosis, spotty calcification, and Radscore, demonstrating moderate discriminative ability for risk stratification in ASCI patients. Several studies suggested the associations of IAC with ischemic cerebrovascular events ( 3 ) and cognitive impairment( 15 ), yet the role of ICA in the culprit lesion remained understudied. The simultaneous interpretation of VW-MRI and CT images enables the identification of intracranial atherosclerotic plaques and their associated calcification presence. CT values are recognized as the gold standard for assessing calcified components, even within small intracranial atherosclerotic lesions. Our study expanded upon existing histological evidence to investigate the varied presence of calcification in culprit lesions within extracranial artery beds. The gray levels on CT images, derived from absolute voxel attenuation values without considering complex spatial relationships, provide appropriate data for radiomic analysis. Radiomics utilizes computational methods to derive numerous quantitative features for detailed calcification characterization, thereby improving diagnostics and risk assessment( 16 ). Of note, calcification was not exclusively confined within the intimal due to the limited resolution of the imaging modality, which precluded differentiation between the various layers of the artery( 17 ). Histology studies verified that calcification occurs frequently in the internal elastic lamina of the intracranial artery, with minimal luminal stenosis. Cases with mild stenosis (< 30%) have been excluded from our inclusion criteria. Moreover, calcification within the intimal and medial layers usually occurs concurrently and is frequently exacerbated by inflammation( 18 ). A thorough evaluation that includes the subintimal lesion and the surrounding area could provide more information in identifying culprit lesions. In the present study, CT-based ML of IAC reached an AUC of 0.674 in the training set. In the previous clinical scenarios to predict adverse clinical event risk, a model based on coronary artery calcium radiomic features achieved a higher AUC of 0.76( 8 ). This study was not comparable to ours as it evaluated the relationship between overall calcification burden in four main coronary arteries and the long-term adverse outcomes, aside from the different targeted arterial beds. Spotty calcification was suggested as a vulnerable plaque feature associated with clinical symptoms and poor prognosis. Homssi et al. demonstrated that spotty calcification in nonstenotic carotid atherosclerosis was linked to ischemic stroke on the ipsilateral side( 19 ). We previously reported a higher prevalence of spotty calcification in the vertebrobasilar artery among stroke patients( 20 ). In the present study, no association was found between spotty calcification and the presence of cerebral infarction after multivariate adjustment. Compared to the VBA, calcification in the iICA typically occurs within the internal elastic lamina( 21 ), with a lining and circular shape appearance. The crosstalk between atherosclerotic calcification in the intimal layer and non-atherosclerotic calcification in other layers( 22 ) may lead to the ambiguous observation of spotty calcification in several cases. Culprit plaques exhibit a higher degree of stenosis than non-culprit plaques; therefore, stenosis severity remains the predominant arbiter of decision-making regarding revascularization. Discrepancies between stenosis and ischemic events highlighted that atherosclerosis is the primary disease process( 23 ). Imaging advancements have shown that atherosclerosis without significant stenosis can still cause downstream ischemic events( 24 ). In our study, we found that stenosis was linked to the presence of culprit lesions even after adjusting for multiple variables. This is reasonable as hemodynamic insufficiency is the primary stroke mechanism in iICA and VBA( 25 ). The stenosis-focused approach, such as CT angiography, could achieve the characterization of calcification at stenotic lesions. Future studies should leverage artificial intelligence to analyze the additive effects of ICA quantification as an indicator of atherosclerosis, along with its consequences on lumen geometry (such as stenosis and positive remodeling) for stroke diagnosis, prognostication, and therapeutic decision-making. In the nomogram, diabetes, coronary heart disease, and smoking were identified as independent predictors of cerebral infarction in our study, with odds ratios of 2.37, 3.05, and 1.99, respectively. These risk factors also contributed to IAC( 26 , 27 ), indicating a potential association between IAC and atherosclerosis. Interestingly, hypertension, as a significant risk factor of stroke and major adverse cardiovascular events, showed no statistical differences between ASCI and non-ASCI patients. The degree of cerebral arterial stiffness, a marker of hypertension, was associated with calcification in the internal elastic lamina. Future studies with a larger sample size may clarify the relationship between stroke mechanisms and the characteristics of culprit lesions, particularly regarding the presence and morphology of calcifications. This study has several limitations. Firstly, our retrospective study, which includes both CT and VW-MRI, may introduce a possibility of selection bias. Secondly, the small region of interest for calcification led to the exclusion of fourteen cases from radiomic analysis, which may have contributed to the decreased diagnostic performance of spotty calcification. A nomogram that integrates visual assessment of calcification with radiomic analysis is essential. Lastly, manual segmentation was a prerequisite for generating regions of interest when focusing on artery calcification at the culprit lesion, although time-consuming and labor-intensive. Future research should explore the utility of auto-segmentation and deep learning( 28 ) to assess the vessel-specific burden of ICA in predicting ischemic cerebrovascular events. CONCLUSION The machine learning analyses based on CT image features of intracranial artery calcification aids in the identification of plaque instability. The nomogram could serve as a diagnostic tool for clinical decision-making, offering personalized information for stroke risk assessment and practical treatment guidance. Abbreviations ASCI: acute/subacute ischemic cerebral infarction AUC: area under the curve (AUC) CT: computed tomography HU: Hounsfield units (HU) ICA: intracranial artery calcification iICA: intracranial internal carotid artery ML: machine learning (ML) Radscore: radiomic-based score ROI: region of interest VBA: vertebrobasilar artery VW-MRI: vessel wall magnetic resonance imaging Declarations Ethics approval and consent to participate This retrospective study received approval from the local institutional review board at Shandong Provincial Hospital Affiliated with Shandong First Medical University and Qilu Hospital of Shandong University. The requirement for informed consent was waived. Consent for publication Not applicable Availability of data and materials Data are available from the authors upon reasonable request. Competing interests The authors declare that they have no competing interests. Funding This research was supported by the Natural Science Foundation of Shandong Province (grant numbers ZR2023MH320 and ZR2025QC980). Authors' contributions H.Y. and B.L. developed the concept; H.Y., Y.C., T.M., J.F., C.H., and J.X. performed the data analysis and drafted the manuscript; X.W. administered the project; B.L. was responsible for research ethics. All authors reviewed the final manuscript. Acknowledgements Not applicable References Kockelkoren R, De Vis JB, de Jong PA, Vernooij MW, Mali W, Hendrikse J, et al. Intracranial Carotid Artery Calcification From Infancy to Old Age. J Am Coll Cardiol. 2018;72(5):582-4. Bartstra JW, van den Beukel TC, Van Hecke W, Mali W, Spiering W, Koek HL, et al. Intracranial Arterial Calcification: Prevalence, Risk Factors, and Consequences: JACC Review Topic of the Week. J Am Coll Cardiol. 2020;76(13):1595-604. Berghout BP, Camarasa RY, Van Dam-Nolen DH, van der Lugt A, de Bruijne M, Koudstaal PJ, et al. Burden of intracranial artery calcification in white patients with ischemic stroke. Eur Stroke J. 2024;9(3):743-50. Virmani R, Burke AP, Farb A, Kolodgie FD. Pathology of the vulnerable plaque. J Am Coll Cardiol. 2006;47(8 Suppl):C13-8. Motoyama S, Kondo T, Sarai M, Sugiura A, Harigaya H, Sato T, et al. Multislice computed tomographic characteristics of coronary lesions in acute coronary syndromes. J Am Coll Cardiol. 2007;50(4):319-26. Zhang F, Yang L, Gan L, Fan Z, Zhou B, Deng Z, et al. Spotty Calcium on Cervicocerebral Computed Tomography Angiography Associates With Increased Risk of Ischemic Stroke. Stroke. 2019;50(4):859-66. Liu B, Xue C, Lu H, Wang C, Duan S, Yang H. CT texture analysis of vertebrobasilar artery calcification to identify culprit plaques. Front Neurol. 2024;15:1381370. Eslami P, Parmar C, Foldyna B, Scholtz JE, Ivanov A, Zeleznik R, et al. Radiomics of Coronary Artery Calcium in the Framingham Heart Study. Radiol Cardiothorac Imaging. 2020;2(1):e190119. Miller RJH, Pieszko K, Shanbhag A, Feher A, Lemley M, Killekar A, et al. Deep Learning Coronary Artery Calcium Scores from SPECT/CT Attenuation Maps Improve Prediction of Major Adverse Cardiac Events. J Nucl Med. 2023;64(4):652-8. Tao L, Li XQ, Hou XW, Yang BQ, Xia C, Ntaios G, et al. Intracranial Atherosclerotic Plaque as a Potential Cause of Embolic Stroke of Undetermined Source. J Am Coll Cardiol. 2021;77(6):680-91. Tomaniak M, Katagiri Y, Modolo R, de Silva R, Khamis RY, Bourantas CV, et al. Vulnerable plaques and patients: state-of-the-art. Eur Heart J. 2020;41(31):2997-3004. Qiao Y, Zeiler SR, Mirbagheri S, Leigh R, Urrutia V, Wityk R, et al. Intracranial plaque enhancement in patients with cerebrovascular events on high-spatial-resolution MR images. Radiology. 2014;271(2):534-42. Arnold Fiebelkorn C, Vemuri P, Rabinstein AA, Mielke MM, Przybelski SA, Kantarci K, et al. Frequency of Acute and Subacute Infarcts in a Population-Based Study. Mayo Clin Proc. 2018;93(3):300-6. Samuels OB, Joseph GJ, Lynn MJ, Smith HA, Chimowitz MI. A standardized method for measuring intracranial arterial stenosis. AJNR Am J Neuroradiol. 2000;21(4):643-6. van den Beukel TC, Wolters FJ, Siebert U, Spiering W, Ikram MA, Vernooij MW, et al. Intracranial arteriosclerosis and the risk of dementia: A population-based cohort study. Alzheimers Dement. 2024;20(2):869-79. Rogers MA, Aikawa E. Cardiovascular calcification: artificial intelligence and big data accelerate mechanistic discovery. Nat Rev Cardiol. 2019;16(5):261-74. Portanova A, Hakakian N, Mikulis DJ, Virmani R, Abdalla WM, Wasserman BA. Intracranial vasa vasorum: insights and implications for imaging. Radiology. 2013;267(3):667-79. Durham AL, Speer MY, Scatena M, Giachelli CM, Shanahan CM. Role of smooth muscle cells in vascular calcification: implications in atherosclerosis and arterial stiffness. Cardiovasc Res. 2018;114(4):590-600. Homssi M, Vora A, Zhang C, Baradaran H, Kamel H, Gupta A. Association Between Spotty Calcification in Nonstenosing Extracranial Carotid Artery Plaque and Ipsilateral Ischemic Stroke. J Am Heart Assoc. 2023;12(10):e028525. Yang H, Liu B, Yin Q, Zhang S, Shen Y, Ji C, et al. Comparison of symptomatic vertebrobasilar plaques between patients with and without Diabetes Mellitus using computed tomographic angiography and vessel wall magnetic resonance imaging. Diab Vasc Dis Res. 2022;19(1):14791641211073944. Vos A, Van Hecke W, Spliet WG, Goldschmeding R, Isgum I, Kockelkoren R, et al. Predominance of Nonatherosclerotic Internal Elastic Lamina Calcification in the Intracranial Internal Carotid Artery. Stroke. 2016;47(1):221-3. Bardeesi ASA, Gao J, Zhang K, Yu S, Wei M, Liu P, et al. A novel role of cellular interactions in vascular calcification. J Transl Med. 2017;15(1):95. Ahmadi A, Leipsic J, Ovrehus KA, Gaur S, Bagiella E, Ko B, et al. Lesion-Specific and Vessel-Related Determinants of Fractional Flow Reserve Beyond Coronary Artery Stenosis. JACC Cardiovasc Imaging. 2018;11(4):521-30. Ryoo S, Lee MJ, Cha J, Jeon P, Bang OY. Differential Vascular Pathophysiologic Types of Intracranial Atherosclerotic Stroke: A High-Resolution Wall Magnetic Resonance Imaging Study. Stroke. 2015;46(10):2815-21. Kim JS, Nah HW, Park SM, Kim SK, Cho KH, Lee J, et al. Risk factors and stroke mechanisms in atherosclerotic stroke: intracranial compared with extracranial and anterior compared with posterior circulation disease. Stroke. 2012;43(12):3313-8. Vos A, Kockelkoren R, de Vis JB, van der Schouw YT, van der Schaaf IC, Velthuis BK, et al. Risk factors for atherosclerotic and medial arterial calcification of the intracranial internal carotid artery. Atherosclerosis. 2018;276:44-9. van den Beukel TC, Lucci C, Hendrikse J, Spiering W, Koek HL, Geerlings MI, et al. Risk factors for calcification of the vertebrobasilar arteries in cardiovascular patients referred for a head CT, the SMART study. J Neuroradiol. 2021;48(4):248-53. Mu D, Bai J, Chen W, Yu H, Liang J, Yin K, et al. Calcium Scoring at Coronary CT Angiography Using Deep Learning. Radiology. 2022;302(2):309-16. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 13 Nov, 2025 Reviews received at journal 11 Nov, 2025 Reviews received at journal 07 Nov, 2025 Reviewers agreed at journal 01 Nov, 2025 Reviews received at journal 29 Oct, 2025 Reviewers agreed at journal 29 Oct, 2025 Reviewers agreed at journal 29 Oct, 2025 Reviewers invited by journal 29 Oct, 2025 Editor assigned by journal 12 Oct, 2025 Submission checks completed at journal 12 Oct, 2025 First submitted to journal 02 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7763910","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":541682817,"identity":"e7d14115-e954-49ec-870e-d1d57ac1c1c6","order_by":0,"name":"Huan Yang","email":"","orcid":"","institution":"Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Huan","middleName":"","lastName":"Yang","suffix":""},{"id":541682818,"identity":"33b5a3b7-9bf8-4f0b-ba5f-dcd100a697fa","order_by":1,"name":"Yunchao Chen","email":"","orcid":"","institution":"Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yunchao","middleName":"","lastName":"Chen","suffix":""},{"id":541682819,"identity":"4f8a1010-b113-4361-8a1b-bf366af08ddf","order_by":2,"name":"Teng Ma","email":"","orcid":"","institution":"Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Teng","middleName":"","lastName":"Ma","suffix":""},{"id":541682820,"identity":"463b8ce8-ed3c-47db-b4d3-16e08cd872c0","order_by":3,"name":"Jizhen Feng","email":"","orcid":"","institution":"Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jizhen","middleName":"","lastName":"Feng","suffix":""},{"id":541682823,"identity":"15fa4d44-1772-4ff9-b2d6-029e09adad43","order_by":4,"name":"Chencui Huang","email":"","orcid":"","institution":"Hangzhou Deepwise \u0026 League of PHD Technology Co., Ltd","correspondingAuthor":false,"prefix":"","firstName":"Chencui","middleName":"","lastName":"Huang","suffix":""},{"id":541682825,"identity":"85a3bf55-a1bb-40ad-8f31-896895d638c5","order_by":5,"name":"Jingxu Xu","email":"","orcid":"","institution":"Hangzhou Deepwise \u0026 League of PHD Technology Co., Ltd","correspondingAuthor":false,"prefix":"","firstName":"Jingxu","middleName":"","lastName":"Xu","suffix":""},{"id":541682828,"identity":"cb8555ac-ef92-47b5-8e39-78364c960fae","order_by":6,"name":"Ximing Wang","email":"","orcid":"","institution":"Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ximing","middleName":"","lastName":"Wang","suffix":""},{"id":541682830,"identity":"1089152a-c768-4da0-bd25-ef91637d3a4a","order_by":7,"name":"Bo Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAtElEQVRIiWNgGAWjYBACPmYGNhDN2MDe2PjwAzFa2OBaeA43G0sQpYUBpkUivU2Ahygt7DxmDz4w1MluuPmwjUGCwU5Ot4Ggw3jMDWcwHDbecDux7UEBQ7Kx2QHCWsykeRgOJAK1tBtIABnbiNRSl7jh5sE2CR4StDAnbrjBSLQWtnKwX2aeSQQGsgERfuHnP7wNHGJ9x48/fPihwk6OoBYwYPwHYxkQo3wUjIJRMApGAUEAAJtHOzGzjogbAAAAAElFTkSuQmCC","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":true,"prefix":"","firstName":"Bo","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-10-02 05:38:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7763910/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7763910/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95536221,"identity":"8e49af95-46b8-478d-a0eb-f7672dabed1d","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16152007,"visible":true,"origin":"","legend":"","description":"","filename":"20251010main.docx","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/96a084f1a058e21282a51647.docx"},{"id":95536223,"identity":"bd12513e-d05c-4475-bb28-06253b5b1f86","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32693,"visible":true,"origin":"","legend":"","description":"","filename":"Table20250202.docx","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/30bda06ae0c006ea4279dc5c.docx"},{"id":95536225,"identity":"70f04cb3-5e6e-42d7-99cf-4cd9ce4ff060","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"json","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9343,"visible":true,"origin":"","legend":"","description":"","filename":"14e865e15910491a88facc1df23a8f38.json","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/7b0e9c98d940baf47ec8c8a2.json"},{"id":95654510,"identity":"16ed1b6f-e261-480c-b94c-8e533c692116","added_by":"auto","created_at":"2025-11-11 16:12:15","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157719,"visible":true,"origin":"","legend":"","description":"","filename":"14e865e15910491a88facc1df23a8f381enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/93a2feb3670f5756c0b72e3b.xml"},{"id":95536231,"identity":"ff79e5ef-79e1-4b05-9031-f529a700dbc5","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4184250,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/14b0665261a21ef43f93c678.jpeg"},{"id":95653963,"identity":"0da51218-da17-4785-b871-8c6d23700d59","added_by":"auto","created_at":"2025-11-11 16:06:53","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2582920,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/cae3835befe0dd440a91cb39.jpeg"},{"id":95653918,"identity":"671c561b-d244-4e4a-85b3-50004de482b7","added_by":"auto","created_at":"2025-11-11 16:05:57","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3196418,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/19c2e4727ece5f5ec427c99a.jpeg"},{"id":95536229,"identity":"8e381f1c-7329-491d-a184-957968c41c1b","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3437374,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/12311cca23ad53eba3a207f7.jpeg"},{"id":95653943,"identity":"18307757-5a7e-470e-8842-71028754ef4d","added_by":"auto","created_at":"2025-11-11 16:06:32","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2641810,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/6de9f64666044f940a940c3d.jpeg"},{"id":95536230,"identity":"5c6b2636-3d0f-4a14-9526-e48f873caa22","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59683,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/2af70e920582415a59571138.png"},{"id":95536236,"identity":"4c6320de-9e0d-415b-a254-ff5e14dd8a5b","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62713,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/cc049d6c1c7c6349ffa85896.png"},{"id":95536233,"identity":"6d32e65d-e8f6-4ef9-abc7-60ce94f3cae3","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25164,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/e5dcd2124eba46c1929f38bc.png"},{"id":95536226,"identity":"0e8d371e-cfe3-4a2f-abaa-73b2cdc0e8b5","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23149,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/a0d65afbc262c16c6ee4c331.png"},{"id":95536238,"identity":"4bd24878-7dfd-40d6-8272-ce4c91ad5992","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35593,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/dd9878d0ecc774c490d81c74.png"},{"id":95654018,"identity":"70958a40-7fe5-4f91-aa71-98dee1243ecd","added_by":"auto","created_at":"2025-11-11 16:08:28","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157635,"visible":true,"origin":"","legend":"","description":"","filename":"14e865e15910491a88facc1df23a8f381structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/90b52c11a63a5f4accf7370f.xml"},{"id":95536240,"identity":"376a1885-446f-439e-be69-4f00e9dec727","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"html","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":165799,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/165d34bbb35aa4d8e27c2482.html"},{"id":95654127,"identity":"14d40358-dba3-491e-be5f-36be91a9b784","added_by":"auto","created_at":"2025-11-11 16:09:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":181100,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of this study. ROC, receiver operating characteristic curve.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/39d2a034e0e4bb367422e78b.png"},{"id":95536219,"identity":"3548fb82-81a0-4b2f-ba53-f691ef6bbf28","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":214021,"visible":true,"origin":"","legend":"\u003cp\u003eRadiomic workflow of this study.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/e64cf86e2732332af64c8708.png"},{"id":95536220,"identity":"a559805b-3ec4-4e17-8315-46d79919e44a","added_by":"auto","created_at":"2025-11-10 10:39:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":116789,"visible":true,"origin":"","legend":"\u003cp\u003eInformation of twenty features selected and their corresponding weights in this study.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/b0505bba2c38d3d118ac18e0.png"},{"id":95654336,"identity":"104afcc0-7e82-4172-bd37-8a97a1c42c87","added_by":"auto","created_at":"2025-11-11 16:11:07","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":106686,"visible":true,"origin":"","legend":"\u003cp\u003eA nomogram integrating the radiomics score and clinical-imaging covariates.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/25dd9f282dd825e654cb4786.png"},{"id":95654057,"identity":"9391bff8-acef-4519-8b7f-3784caed0bd8","added_by":"auto","created_at":"2025-11-11 16:09:26","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":158980,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of the receiver operating characteristic curves of machine learning models in both the training set (A) and the external test set (B).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/c69eff306ef8ebf617997035.png"},{"id":95659931,"identity":"56d7985b-bb3a-4445-881a-87fd2fc4abf6","added_by":"auto","created_at":"2025-11-11 16:30:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1794772,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7763910/v1/fd044458-ac58-431d-a54f-360d1c2ec075.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Nomogram of Intracranial Artery Calcification with Integration of CT-based Radiomics to Identify Culprit Lesions","fulltext":[{"header":"BACKGROUND","content":"\u003cp\u003eIntracranial artery calcification (IAC) is a frequent imaging finding on computed tomography (CT) scans, with incidence rates reaching 100% in individuals aged 90 and older(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). The presence of IAC indicates a correlation with atherosclerosis sharing similar cardiovascular risk factors(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). A significant IAC burden is associated with an increased risk of stroke across all ethnic groups(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). It is widely acknowledged that coronary artery calcification in culprit coronary lesions, as verified by the histological evidence, could be used as a predictor for myocardial infarction(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Given the systemic and divergent nature of atherosclerosis, there is considerable interest in understanding the clinical implications of IAC in culprit lesions.\u003c/p\u003e\u003cp\u003eThe unstable effects of calcification on atherosclerotic plaques arise from inflammation and heightened biomechanical forces. Spotty calcification is recognized as a biomarker of plaque vulnerability in coronary arteries and is closely linked to plaque rupture(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Recent studies have indicated that the presence or number of spotty calcifications in carotid and intracranial arteries is associated with an increased risk of stroke(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). However, this smallest calcification unit discernible on CT images provides limited diagnostic information, primarily regarding morphology. Additionally, visual assessments of calcification heavily rely on radiologists' interpretations and experience. Facilitating objective quantification of this type of radiologic data beyond human visual capabilities is crucial.\u003c/p\u003e\u003cp\u003eRadiomics entails extracting extensive data from gray-level images, including features related to first-order statistics, shape, texture, and transform-based characteristics. We previously showed that texture analysis of intracranial artery calcification offered more information than the visual assessments and held significant potential for identifying culprit plaques(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Machine learning (ML) could transform complex, heterogeneous pathology into a straightforward radiomic-based parameter. A radiomic-based score (Radscore) for cardiac CT significantly enhanced the predictive accuracy compared to the conventional Agatston calcium scores in the community-based Framingham Heart Study(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Deep learning of coronary artery calcium scores from SPECT/CT adds significant values to SPECT myocardial perfusion alone for risk assessment(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). We hypothesized that the artificial intelligence of IAC can provide additional value in stratifying plaque instability beyond traditional assessments.\u003c/p\u003e\u003cp\u003eHigh-resolution vessel wall magnetic resonance imaging (VW-MRI) significantly improved our understanding of intracranial atherosclerosis as a primary cause of ischemic stroke. VW-MRI enables the assessment of plaque morphology and compositional characterization, i.e., intraplaque hemorrhage and fibrous cap(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Nevertheless, interpreting IAC can be challenged on routine MRI due to insufficient resolution to detect the complex signals surrounding calcium deposits. CT is the preferred imaging modality for visualizing calcifications and the first-line diagnostic tool for stroke patients. Combining CT and VW-MRI images could comprehensively characterize intracranial atherosclerosis and accompanied calcifications.\u003c/p\u003e\u003cp\u003eThe presence of vulnerable plaques in high-risk patients increases the likelihood of adverse cardiovascular events(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). A nomogram is commonly designed for individual patient profiles to estimate the probability of clinical events using an intuitive graphical interface. The study aimed to develop a ML model using CT images of IAC and assess the diagnostic value of the incorporated nomogram in identifying culprit intracranial atherosclerotic lesions, using VW-MRI as the reference standard.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Population\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research was conducted in accordance with the Declaration of Helsinki. Clinical trial number: not applicable. The cohort for this study was drawn from patients who underwent VW-MRI and head CT scans within two weeks at two medical centers from January 2020 to August 2024. Patients were included if there were 30% to 99% atherosclerotic stenosis in the intracranial internal carotid artery (iICA) or vertebrobasilar artery (VBA) on VW-MRI images(12) and the concurrent presence of calcification on CT images. Exclusion criteria were (i) evidence of non-atherosclerotic intracranial vascular pathology (e.g., cardiac embolism, dissection, vasculitis, aneurysm, moya-moya disease); (ii) a history of stent or treatment of the target vessel; and (iii) inadequate image quality, incomplete clinical data or insufficient region of interest (ROI) for the radiomic analysis. The Institutional Review Boards approved this study, and the requirement for informed consent were waived. The flow chart of patient selection is shown in \u003cstrong\u003eFigure 1\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI Examination\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMRI exams were performed on 3T MR imaging scanners (Ingenia, Philips Healthcare; MAGNETOM Prisma, Siemens Healthineers) with either an 8-channel or 16-channel head coil. The protocols of MRI scans are shown in \u003cstrong\u003eTable 1\u003c/strong\u003e.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eA plaque was considered the culprit if the only or most stenotic lesion within the artery territory of acute/subacute cerebral infarction (ASCI)(13). A plaque was considered as non-culprit if it was the most stenotic lesion in non-ASCI patients. Luminal stenosis was measured using criteria established in the Warfarin-Aspirin Symptomatic Intracranial Disease trial(14) on pre-contrast VW-MRI images.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 1. MRI Sequence Parameters\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 29px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eScan parameter\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 34px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePHILIPS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSIEMENS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT1-VISTA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDWI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT1-SPACE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDWI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eFOV (mm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e220\u0026thinsp;\u0026times;\u0026thinsp;220\u0026thinsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e250\u0026thinsp;\u0026times;\u0026thinsp;250\u0026thinsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e160 \u0026times; 160\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e220\u0026thinsp;\u0026times;\u0026thinsp;220\u0026thinsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eAcquisition orientation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eCoronal/axial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003eAxial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eSagittal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003eAxial\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eAcquired spatial resolution (mm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.7\u0026thinsp;\u0026times;\u0026thinsp;0.7\u0026thinsp;\u0026times;\u0026thinsp;1.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.2\u0026times;\u0026thinsp;1.2\u0026thinsp;\u0026times;\u0026thinsp;5.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e0.6 \u0026times; 0.6 \u0026times; 0.6\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.1\u0026times;\u0026thinsp;1.1\u0026thinsp;\u0026times;\u0026thinsp;5.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eMatrix\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e316 \u0026times; 312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e152 \u0026times; 122\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e256 \u0026times; 256\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e192 x 192\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eTR/TE (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e425/19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e2250/70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e800/4.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e2800/55\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eb value (mm\u003csup\u003e2\u003c/sup\u003e/s)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0/1000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0/1000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eEcho spacing (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e6.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e4.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eNumber of averages\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eParallel acceleration factor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 29px;\"\u003e\n \u003cp\u003eAcquisition time (min)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e6.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 3px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e8.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 100px;\"\u003e\n \u003cp\u003eDWI, diffusion-weighted imaging; FOV, field-of-view; TE, echo time; TR, repetition time; T1-VISTA, T1-weighted volumetric isotropic turbo spin-echo acquisition; T1-SPACE, T1-sampling perfection with application-optimized contrasts using different flip angle evolutions.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eCT Imaging\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe CT scans were conducted on 128-slice dual-source CT scanners (SOMATOM Force, Siemens Healthineers; SOMATOM Definition Flash, Siemens Healthineers). The parameters of head CT scanning were as follows: tube voltage of 120 kVp, reconstructed slice thickness of 1 mm, reconstructed slice interval of 0.5 mm and scan range from the foramen magnum to the top of the skull. Image quality was evaluated based on the presence and severity of artifacts, including beam hardening, photon starvation, and noise, using a three-point scale: poor, adequate, or excellent. Images were determined to be of adequate or excellent quality for analysis. Calcification refers to areas with increased density having a CT attenuation value of 130 Hounsfield units (HU) or higher. Spotty calcification refers to calcium deposits that are less than 3 mm in length and are contained within an arc measuring less than 90 degrees. Two radiologists independently interpreted all CT and VW-MRI images using specific anatomical landmarks (e.g., carotid siphon, vertebrobasilar junction). The disagreement in data annotation was resolved by consensus.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eML of Calcification Features\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;All original CT images were stored in Neuroimaging Informatics Technology Initiative (NIfTI) format and loaded into Medical Imaging Interaction Toolkit (MITK, open-source software, https://www.mitk.org). Calcification radiomic analyses were performed by a radiologist with five years of experience in neuroradiology, who was blind to the clinical information. The ROI was delineated manually slice by slice in the axial plane, and this process was cross-checked with VW-MRI. In the presence of multiple calcifications, each must be distinctly delineated. Meanwhile, a radiologist with ten years of expertise in VW-MRI verified the delineation of the ROI. When differing opinions were presented, the final ROI was determined after discussion. The segmentations were used as mask images to isolate voxels containing calcifications in the corresponding plaque lesion.\u003c/p\u003e\n\u003cp\u003eThe integrated ML platform (Deepwise Multimodal Research Platform version 2.6.0) was designed to analyze annotated medical data. The radiomic features extracted from the ROIs were as follows: 18 first-order features; 14 size and shape features; 75 texture features; 1111 transform-based features, including Laplacian of Gaussian filtering and Wavelet features. The following steps were performed to select features for identifying culprit lesions. Firstly, each feature group was standardized by the z-score normalization. Secondly, Spearman\u0026apos;s correlation coefficient was calculated to determine the correlation between features. If the correlation coefficient between any two independent variables exceeded 0.9, one of the features would be eliminated to reduce redundancy. Thirdly, the F-test was conducted on the remaining features, and twenty features with the lowest p-values were selected as effective predictors for the subsequent logistic regression. Based on ranking, this selection process ensured that the most informative features were retained.\u003c/p\u003e\n\u003cp\u003eThe radiomics model was developed using a logistic regression classifier. To ensure a valid and robust model based on the data, fivefold cross-validation was used， which entails dividing the dataset into five mutually exclusive subsets of roughly equal size. The final result is obtained by averaging the outcomes from all five training iterations. \u003cstrong\u003eFigure 2\u0026nbsp;\u003c/strong\u003eshows the workflow of calcification radiomic analysis. Univariate analysis of clinical factors, including age, sex, risk factors (diabetes mellitus, hypertension, smoking, hyperlipidemia and coronary heart disease) and medication uses, stenosis, spotty calcification, and Radscore, was performed. A multivariate logistic regression included variables with p-values less than 0.2 to develop a radiomic nomogram. The diagnostic performance of the nomogram was assessed using the area under the receiver operating characteristic curve (AUC).\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCategorical data were presented as frequencies, while continuous data were presented as either means \u0026plusmn; standard deviations or medians with interquartile ranges. The differences between various groups were analyzed using the Student\u0026apos;s t-test, the Mann-Whitney U test for continuous variables, and the Chi-squared (\u0026chi;\u0026sup2;) test for categorical variables. The univariate and multivariate logistic regression analyses were conducted to determine covariates associated with culprit lesions. The diagnostic performance of all models was assessed using AUC, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. A two-tailed p-value of less than 0.05 was considered statistically significant. Statistical analyses were performed using the Deepwise Multimodal Research Platform.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003eClinical and Lesion Characteristics Between ASCI and non-ASCI Groups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e presents the clinical and lesion characteristics. A total of 282 patients (195 male; mean age, 62.07 \u0026plusmn; 9.12 years) were included in the training set, of which 133 (47.16%) were ASCI patients and 149 (52.84%) were non-ASCI patients. Age and sex were similar between ASCI and non-ASCI groups. The ASCI group exhibited a greater degree of stenosis (p\u0026lt;0.001) and stenosis grade (p=0.005) compared to the non-ASCI group. In the visual assessment of calcification characteristics, the ASCI group exhibited a higher prevalence of spotty calcification (45.86% vs. 30.20%, p=0.007) than the non-ASCI group. Among the 71 patients (52 male; mean age, 61.42 \u0026plusmn; 9.62 years) included in the external test set, 36 (50.70%) had ASCI. Multivariate regression analysis showed that diabetes, smoking, coronary heart disease, and stenosis were associated with ASCI, with odds ratios of 2.37, 1.99, 3.05, and 1.67, respectively (\u003cstrong\u003eTable 3\u003c/strong\u003e). The parameter of spotty calcification was insignificant (p=0.152).\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\" style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable2. Clinical and Imaging Characteristics between ASCI and Non-ASCI Groups\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTraining set\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 8px;\"\u003e\n \u003cp\u003eP value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 35px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eExternal test set\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 8px;\"\u003e\n \u003cp\u003eP value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003eASCI (n=133)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003eNon-ASCI (n=149)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eASCI (n=36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eNon-ASCI (n=35)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eAge, mean\u0026plusmn;SD, year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e61.47\u0026plusmn;9.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e62.61\u0026plusmn;8.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.297\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e60.42\u0026plusmn;10.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e62.46\u0026plusmn;9.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.376\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eSex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.790\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.381\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e93(69.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e102(68.46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e28(77.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e24(68.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e40(30.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e47(31.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e8(22.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e11(31.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eDM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.052\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.186\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e77(57.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e69(46.31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e19(52.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e13(37.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e56(42.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e80(53.69)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e17(47.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e22(62.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eHTN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.326\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.885\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e104(78.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e109(73.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e25(69.44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e25(71.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e29(21.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e40(26.85)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e11(30.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e10(28.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eSmoking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.112\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.275\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e55(41.35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e48(32.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e18(50.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e13(37.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e78(58.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e101(67.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e18(50.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e22(62.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eHLD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.482\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.211\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e36(27.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e46(30.87)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e5(13.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e9(25.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e97(72.93)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e103(69.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e31(86.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e26(74.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eCHD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.127\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.417\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e43(32.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e36(24.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e9(25.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e6(17.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e90(67.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e113(75.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e27(75.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e29(82.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eHistory of statin use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.264\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.350\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e39(29.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e53(35.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e6(16.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e9(25.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e94(70.68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e96(64.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e30(83.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e26(74.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eHistory of aspirin use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.241\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.168\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e42(31.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e57(38.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e8(22.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e13(37.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e91(68.42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e92(61.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e28(77.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e22(62.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003ePlaque location\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.734\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.614\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Anterior circulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e84(63.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e97(65.10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e33(94.37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e34(97.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Posterior circulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e49(36.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e52(34.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e3(5.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1(2.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eStenosis degree, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e75.00(64.52-83.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e68.00(54.98-77.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e＜0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e83.21(67.5-89.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e70.00(49.00-85.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eStenosis grade\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.005\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; ＜30%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e7(5.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e18(12.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e2(5.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e9(25.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; 30%-69%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e40(30.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e63(42.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e7(19.44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e6(17.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; 70%-89%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e62(46.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e55(36.91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e18(50.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e14(40.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026ge;90%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e24(18.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e13(8.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e9(25.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e6(17.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003eSpotty calcification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.007\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e61(45.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e45(30.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e25(69.44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e14(40.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e72(54.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 13px;\"\u003e\n \u003cp\u003e104(69.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e11(30.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e21(60.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\" style=\"width: 100px;\"\u003e\n \u003cp\u003eASCI, acute/subacute cerebral infarction; CHD, coronary heart disease; DM, diabetes mellitus; HLD, hyperlipidemia; HTN, hypertension; IQR, interquartile range; SD, standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" style=\"width: 93px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable3. Univariable and Multivariable Regression Analysis for Culprit Lesions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 39px;\"\u003e\n \u003cp\u003eUnivariate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" style=\"width: 36px;\"\u003e\n \u003cp\u003eMultivariate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003eOR\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e\u0026nbsp;95%CI\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eP value\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003eOR\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e95%CI\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003eP value\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eAge\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0.986\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.961-1.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.296\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eSex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.071\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.645-1.778\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.790\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eDM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.594\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.995-2.554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e2.374\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e1.362-4.139\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eHTN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.316\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.761-2.277\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.326\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eSmoking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.484\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.912-2.415\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.112\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e1.993\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e1.098-3.401\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.022\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eHLD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0.831\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.496-1.393\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.483\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eCHD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.500\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.890-2.528\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.128\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e3.049\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e1.617-5.750\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eHistory of statin use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0.752\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.455-1.241\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.265\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e0.735\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e0.413-1.310\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.297\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eHistory of aspirin use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e0.745\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.455-1.219\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.242\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eStenosis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.710\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.266-2.308\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e＜0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e1.671\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e1.198-2.332\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eSpotty calcification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e1.958\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.201-3.191\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e1.508\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e0.859-2.649\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e0.152\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 17px;\"\u003e\n \u003cp\u003eRadscore (per 0.1 increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003e2.411\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e1.716-3.387\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e＜0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 8px;\"\u003e\n \u003cp\u003e2.899\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e1.957-4.294\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 11px;\"\u003e\n \u003cp\u003e＜0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" style=\"width: 93px;\"\u003e\n \u003cp\u003eCHD, coronary heart disease; CI, confidence interval; DM, diabetes mellitus; HLD, hyperlipidemia; HTN, hypertension; OR, odds ratio.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eRadiomic Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOf the 1,218 radiomic features, 916 were removed due to a linear correlation coefficient threshold of 0.9. The remaining 302 features were retained using the F-test. Finally, the logistic regression method selected the twenty most significant features with the lowest p-value to build the prediction model. \u003cstrong\u003eFigure 3\u003c/strong\u003e shows the results of feature weights with fivefold cross-validation. For the finally selected 20 features, 4 (20%) were first order, 12 (60%) were grey level co-occurrence matrix, 2 (10%) were gray-level dependence matrix, 1 (5%) were grey level run length matrix and 1 (5%) was gray-level size zone matrix parameters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eML Models for the Identification of A\u003c/strong\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003eCI\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe diagnostic performance of ML models is shown in \u003cstrong\u003eTable 4\u003c/strong\u003e. The ROC curve showed an AUC of 0.674 (95% confidence interval [CI], 0.612-0.736) for the radiomic model in the training set, with an AUC of 0.609 (95% CI, 0.543-0.674) in the external test set. For the training set, the threshold was set at 0.476, yielding an accuracy of 0.606, a sensitivity of 0.526, a specificity of 0.678, and a Brier Score of 0.227. The fivefold cross-validation resulted in an accuracy of 0.617, a sensitivity of 0.504, a specificity of 0.718, and a Brier Score of 0.236 in the validation set.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 4. Diagnostic Performances of Models to Identify Culprit Lesions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSensitivity\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eAUC (95%CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNPV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003ePPV\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eClinic model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.790\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.362\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.525\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.637(0.573-0.702\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.659\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.525\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eExternal test set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.782\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.349\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.517\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.556(0.489-0.623)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.642\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.517\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eClinic+Stenosis model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.714\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.550\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.586\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.665(0.602-0.728)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.683\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.586\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eExternal test set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.714\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.571\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.638\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.648(0.584-0.712)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.691\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.598\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eNomogram model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.722\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.644\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.642\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.749(0.693-0.805)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.722\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.644\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eExternal test set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.714\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.624\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.629\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.736(0.678-0.793)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.710\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.629\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\"\u003e\n \u003cp\u003eAUC, area under the curve; CI, confidence internal; NPV, negative predictive value; PPV, positive predictive value.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eA radiomics nomogram combined with clinical factors, stenosis, spotty calcification, and Radscore was established (\u003cstrong\u003eFigure 4\u003c/strong\u003e). The Radscore accounted for the largest proportion among all the clinical-radiologic features.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eAs shown in \u003cstrong\u003eFigure 5,\u0026nbsp;\u003c/strong\u003eincorporating stenosis into clinical factors slightly increased the AUC for identifying culprit lesions compared with clinical factors alone. The nomogram with integration of the Radscore provided additional value beyond stenosis and clinical factors in the training set (AUC: 0.749, 95% CI: 0.693-0.805 vs. 0.665, 95% CI: 0.602-0.728, p=0.001) and in the external test set (AUC:0.736, 95%CI: 0.678-0.793 vs. 0.648, 95% CI: 0.584-0.712, p=0.001).\u0026nbsp;\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study aimed to evaluate the effectiveness of ML models, developed by the most relevant radiomic features of calcified intracranial atherosclerotic lesions on CT images, to discriminate lesions that have caused acute/subacute infarctions. The association between Radscore and symptomatic intracranial atherosclerotic lesions was statistically significant, with an odds ratio of 2.9 for every 0.1 increase in the Radscore. The ML model based on IAC features on CT images offered additional value beyond stenosis and clinical variables for identifying symptomatic lesions. Moreover, the final nomogram incorporated patient-specific risk factors, stenosis, spotty calcification, and Radscore, demonstrating moderate discriminative ability for risk stratification in ASCI patients.\u003c/p\u003e\u003cp\u003eSeveral studies suggested the associations of IAC with ischemic cerebrovascular events (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) and cognitive impairment(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e), yet the role of ICA in the culprit lesion remained understudied. The simultaneous interpretation of VW-MRI and CT images enables the identification of intracranial atherosclerotic plaques and their associated calcification presence. CT values are recognized as the gold standard for assessing calcified components, even within small intracranial atherosclerotic lesions. Our study expanded upon existing histological evidence to investigate the varied presence of calcification in culprit lesions within extracranial artery beds.\u003c/p\u003e\u003cp\u003eThe gray levels on CT images, derived from absolute voxel attenuation values without considering complex spatial relationships, provide appropriate data for radiomic analysis. Radiomics utilizes computational methods to derive numerous quantitative features for detailed calcification characterization, thereby improving diagnostics and risk assessment(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Of note, calcification was not exclusively confined within the intimal due to the limited resolution of the imaging modality, which precluded differentiation between the various layers of the artery(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Histology studies verified that calcification occurs frequently in the internal elastic lamina of the intracranial artery, with minimal luminal stenosis. Cases with mild stenosis (\u0026lt;\u0026thinsp;30%) have been excluded from our inclusion criteria. Moreover, calcification within the intimal and medial layers usually occurs concurrently and is frequently exacerbated by inflammation(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). A thorough evaluation that includes the subintimal lesion and the surrounding area could provide more information in identifying culprit lesions.\u003c/p\u003e\u003cp\u003eIn the present study, CT-based ML of IAC reached an AUC of 0.674 in the training set. In the previous clinical scenarios to predict adverse clinical event risk, a model based on coronary artery calcium radiomic features achieved a higher AUC of 0.76(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). This study was not comparable to ours as it evaluated the relationship between overall calcification burden in four main coronary arteries and the long-term adverse outcomes, aside from the different targeted arterial beds.\u003c/p\u003e\u003cp\u003eSpotty calcification was suggested as a vulnerable plaque feature associated with clinical symptoms and poor prognosis. Homssi et al. demonstrated that spotty calcification in nonstenotic carotid atherosclerosis was linked to ischemic stroke on the ipsilateral side(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). We previously reported a higher prevalence of spotty calcification in the vertebrobasilar artery among stroke patients(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). In the present study, no association was found between spotty calcification and the presence of cerebral infarction after multivariate adjustment. Compared to the VBA, calcification in the iICA typically occurs within the internal elastic lamina(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), with a lining and circular shape appearance. The crosstalk between atherosclerotic calcification in the intimal layer and non-atherosclerotic calcification in other layers(\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e) may lead to the ambiguous observation of spotty calcification in several cases.\u003c/p\u003e\u003cp\u003eCulprit plaques exhibit a higher degree of stenosis than non-culprit plaques; therefore, stenosis severity remains the predominant arbiter of decision-making regarding revascularization. Discrepancies between stenosis and ischemic events highlighted that atherosclerosis is the primary disease process(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Imaging advancements have shown that atherosclerosis without significant stenosis can still cause downstream ischemic events(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). In our study, we found that stenosis was linked to the presence of culprit lesions even after adjusting for multiple variables. This is reasonable as hemodynamic insufficiency is the primary stroke mechanism in iICA and VBA(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). The stenosis-focused approach, such as CT angiography, could achieve the characterization of calcification at stenotic lesions. Future studies should leverage artificial intelligence to analyze the additive effects of ICA quantification as an indicator of atherosclerosis, along with its consequences on lumen geometry (such as stenosis and positive remodeling) for stroke diagnosis, prognostication, and therapeutic decision-making.\u003c/p\u003e\u003cp\u003eIn the nomogram, diabetes, coronary heart disease, and smoking were identified as independent predictors of cerebral infarction in our study, with odds ratios of 2.37, 3.05, and 1.99, respectively. These risk factors also contributed to IAC(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e), indicating a potential association between IAC and atherosclerosis. Interestingly, hypertension, as a significant risk factor of stroke and major adverse cardiovascular events, showed no statistical differences between ASCI and non-ASCI patients. The degree of cerebral arterial stiffness, a marker of hypertension, was associated with calcification in the internal elastic lamina. Future studies with a larger sample size may clarify the relationship between stroke mechanisms and the characteristics of culprit lesions, particularly regarding the presence and morphology of calcifications.\u003c/p\u003e\u003cp\u003eThis study has several limitations. Firstly, our retrospective study, which includes both CT and VW-MRI, may introduce a possibility of selection bias. Secondly, the small region of interest for calcification led to the exclusion of fourteen cases from radiomic analysis, which may have contributed to the decreased diagnostic performance of spotty calcification. A nomogram that integrates visual assessment of calcification with radiomic analysis is essential. Lastly, manual segmentation was a prerequisite for generating regions of interest when focusing on artery calcification at the culprit lesion, although time-consuming and labor-intensive. Future research should explore the utility of auto-segmentation and deep learning(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) to assess the vessel-specific burden of ICA in predicting ischemic cerebrovascular events.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eThe machine learning analyses based on CT image features of intracranial artery calcification aids in the identification of plaque instability. The nomogram could serve as a diagnostic tool for clinical decision-making, offering personalized information for stroke risk assessment and practical treatment guidance.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eASCI: acute/subacute ischemic cerebral infarction\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAUC: area under the curve (AUC)\u003c/p\u003e\n\u003cp\u003eCT: computed tomography\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHU: Hounsfield units (HU)\u003c/p\u003e\n\u003cp\u003eICA: intracranial artery calcification\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eiICA: intracranial internal carotid artery\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eML: machine learning (ML)\u003c/p\u003e\n\u003cp\u003eRadscore: radiomic-based score\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eROI: region of interest\u003c/p\u003e\n\u003cp\u003eVBA: vertebrobasilar artery\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eVW-MRI: vessel wall magnetic resonance imaging\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective study received approval from the local institutional review board at Shandong Provincial Hospital Affiliated with Shandong First Medical University and Qilu Hospital of Shandong University. The requirement for informed consent was waived.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData are available from the authors upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Natural Science Foundation of Shandong Province (grant numbers ZR2023MH320 and ZR2025QC980).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eH.Y. and B.L. developed the concept; H.Y., Y.C., T.M., J.F., C.H., and J.X. performed the data analysis and drafted the manuscript; X.W. administered the project; B.L. was responsible for research ethics. All authors reviewed the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eKockelkoren R, De Vis JB, de Jong PA, Vernooij MW, Mali W, Hendrikse J, et al. Intracranial Carotid Artery Calcification From Infancy to Old Age. J Am Coll Cardiol. 2018;72(5):582-4.\u003c/li\u003e\n\u003cli\u003eBartstra JW, van den Beukel TC, Van Hecke W, Mali W, Spiering W, Koek HL, et al. Intracranial Arterial Calcification: Prevalence, Risk Factors, and Consequences: JACC Review Topic of the Week. J Am Coll Cardiol. 2020;76(13):1595-604.\u003c/li\u003e\n\u003cli\u003eBerghout BP, Camarasa RY, Van Dam-Nolen DH, van der Lugt A, de Bruijne M, Koudstaal PJ, et al. Burden of intracranial artery calcification in white patients with ischemic stroke. Eur Stroke J. 2024;9(3):743-50.\u003c/li\u003e\n\u003cli\u003eVirmani R, Burke AP, Farb A, Kolodgie FD. Pathology of the vulnerable plaque. J Am Coll Cardiol. 2006;47(8 Suppl):C13-8.\u003c/li\u003e\n\u003cli\u003eMotoyama S, Kondo T, Sarai M, Sugiura A, Harigaya H, Sato T, et al. Multislice computed tomographic characteristics of coronary lesions in acute coronary syndromes. J Am Coll Cardiol. 2007;50(4):319-26.\u003c/li\u003e\n\u003cli\u003eZhang F, Yang L, Gan L, Fan Z, Zhou B, Deng Z, et al. Spotty Calcium on Cervicocerebral Computed Tomography Angiography Associates With Increased Risk of Ischemic Stroke. Stroke. 2019;50(4):859-66.\u003c/li\u003e\n\u003cli\u003eLiu B, Xue C, Lu H, Wang C, Duan S, Yang H. CT texture analysis of vertebrobasilar artery calcification to identify culprit plaques. Front Neurol. 2024;15:1381370.\u003c/li\u003e\n\u003cli\u003eEslami P, Parmar C, Foldyna B, Scholtz JE, Ivanov A, Zeleznik R, et al. Radiomics of Coronary Artery Calcium in the Framingham Heart Study. Radiol Cardiothorac Imaging. 2020;2(1):e190119.\u003c/li\u003e\n\u003cli\u003eMiller RJH, Pieszko K, Shanbhag A, Feher A, Lemley M, Killekar A, et al. Deep Learning Coronary Artery Calcium Scores from SPECT/CT Attenuation Maps Improve Prediction of Major Adverse Cardiac Events. J Nucl Med. 2023;64(4):652-8.\u003c/li\u003e\n\u003cli\u003eTao L, Li XQ, Hou XW, Yang BQ, Xia C, Ntaios G, et al. Intracranial Atherosclerotic Plaque as a Potential Cause of Embolic Stroke of Undetermined Source. J Am Coll Cardiol. 2021;77(6):680-91.\u003c/li\u003e\n\u003cli\u003eTomaniak M, Katagiri Y, Modolo R, de Silva R, Khamis RY, Bourantas CV, et al. Vulnerable plaques and patients: state-of-the-art. Eur Heart J. 2020;41(31):2997-3004.\u003c/li\u003e\n\u003cli\u003eQiao Y, Zeiler SR, Mirbagheri S, Leigh R, Urrutia V, Wityk R, et al. Intracranial plaque enhancement in patients with cerebrovascular events on high-spatial-resolution MR images. Radiology. 2014;271(2):534-42.\u003c/li\u003e\n\u003cli\u003eArnold Fiebelkorn C, Vemuri P, Rabinstein AA, Mielke MM, Przybelski SA, Kantarci K, et al. Frequency of Acute and Subacute Infarcts in a Population-Based Study. Mayo Clin Proc. 2018;93(3):300-6.\u003c/li\u003e\n\u003cli\u003eSamuels OB, Joseph GJ, Lynn MJ, Smith HA, Chimowitz MI. A standardized method for measuring intracranial arterial stenosis. AJNR Am J Neuroradiol. 2000;21(4):643-6.\u003c/li\u003e\n\u003cli\u003evan den Beukel TC, Wolters FJ, Siebert U, Spiering W, Ikram MA, Vernooij MW, et al. Intracranial arteriosclerosis and the risk of dementia: A population-based cohort study. Alzheimers Dement. 2024;20(2):869-79.\u003c/li\u003e\n\u003cli\u003eRogers MA, Aikawa E. Cardiovascular calcification: artificial intelligence and big data accelerate mechanistic discovery. Nat Rev Cardiol. 2019;16(5):261-74.\u003c/li\u003e\n\u003cli\u003ePortanova A, Hakakian N, Mikulis DJ, Virmani R, Abdalla WM, Wasserman BA. Intracranial vasa vasorum: insights and implications for imaging. Radiology. 2013;267(3):667-79.\u003c/li\u003e\n\u003cli\u003eDurham AL, Speer MY, Scatena M, Giachelli CM, Shanahan CM. Role of smooth muscle cells in vascular calcification: implications in atherosclerosis and arterial stiffness. Cardiovasc Res. 2018;114(4):590-600.\u003c/li\u003e\n\u003cli\u003eHomssi M, Vora A, Zhang C, Baradaran H, Kamel H, Gupta A. Association Between Spotty Calcification in Nonstenosing Extracranial Carotid Artery Plaque and Ipsilateral Ischemic Stroke. J Am Heart Assoc. 2023;12(10):e028525.\u003c/li\u003e\n\u003cli\u003eYang H, Liu B, Yin Q, Zhang S, Shen Y, Ji C, et al. Comparison of symptomatic vertebrobasilar plaques between patients with and without Diabetes Mellitus using computed tomographic angiography and vessel wall magnetic resonance imaging. Diab Vasc Dis Res. 2022;19(1):14791641211073944.\u003c/li\u003e\n\u003cli\u003eVos A, Van Hecke W, Spliet WG, Goldschmeding R, Isgum I, Kockelkoren R, et al. Predominance of Nonatherosclerotic Internal Elastic Lamina Calcification in the Intracranial Internal Carotid Artery. Stroke. 2016;47(1):221-3.\u003c/li\u003e\n\u003cli\u003eBardeesi ASA, Gao J, Zhang K, Yu S, Wei M, Liu P, et al. A novel role of cellular interactions in vascular calcification. J Transl Med. 2017;15(1):95.\u003c/li\u003e\n\u003cli\u003eAhmadi A, Leipsic J, Ovrehus KA, Gaur S, Bagiella E, Ko B, et al. Lesion-Specific and Vessel-Related Determinants of Fractional Flow Reserve Beyond Coronary Artery Stenosis. JACC Cardiovasc Imaging. 2018;11(4):521-30.\u003c/li\u003e\n\u003cli\u003eRyoo S, Lee MJ, Cha J, Jeon P, Bang OY. Differential Vascular Pathophysiologic Types of Intracranial Atherosclerotic Stroke: A High-Resolution Wall Magnetic Resonance Imaging Study. Stroke. 2015;46(10):2815-21.\u003c/li\u003e\n\u003cli\u003eKim JS, Nah HW, Park SM, Kim SK, Cho KH, Lee J, et al. Risk factors and stroke mechanisms in atherosclerotic stroke: intracranial compared with extracranial and anterior compared with posterior circulation disease. Stroke. 2012;43(12):3313-8.\u003c/li\u003e\n\u003cli\u003eVos A, Kockelkoren R, de Vis JB, van der Schouw YT, van der Schaaf IC, Velthuis BK, et al. Risk factors for atherosclerotic and medial arterial calcification of the intracranial internal carotid artery. Atherosclerosis. 2018;276:44-9.\u003c/li\u003e\n\u003cli\u003evan den Beukel TC, Lucci C, Hendrikse J, Spiering W, Koek HL, Geerlings MI, et al. Risk factors for calcification of the vertebrobasilar arteries in cardiovascular patients referred for a head CT, the SMART study. J Neuroradiol. 2021;48(4):248-53.\u003c/li\u003e\n\u003cli\u003eMu D, Bai J, Chen W, Yu H, Liang J, Yin K, et al. Calcium Scoring at Coronary CT Angiography Using Deep Learning. Radiology. 2022;302(2):309-16.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-neurology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nurl","sideBox":"Learn more about [BMC Neurology](http://bmcneurol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/nurl","title":"BMC Neurology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Radiomics, intracranial atherosclerosis, calcification, computed tomography, magnetic resonance imaging","lastPublishedDoi":"10.21203/rs.3.rs-7763910/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7763910/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eTo develop a machine learning model of intracranial artery calcification (IAC) based on computed tomography (CT) images and assess its value for improved identifying culprit lesions responsible for acute/subacute ischemic cerebral infarction (ASCI).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatients with intracranial atherosclerotic diseases in the vertebrobasilar artery or intracranial internal carotid artery who underwent vessel wall MRI and head CT examinations at two hospital centers were retrospectively assessed. Each calcified plaque was classified by the likelihood of having caused an ASCI as culprit or non-culprit. Machine learning technique was utilized to automatically select twenty top-ranked features from IAC segmentation and build a model using the logistic regression algorithm with fivefold stratified cross-validation. The added values of radiomic-based score (Radscore) to stenosis and clinical risk factors for identification of culprit lesions were evaluated using area under the curve (AUC). A nomogram was constructed by integrating the Radscore with clinical and imaging covariates.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOne hundred and thirty-three ASCI patients with culprit plaques were identified in the training set (totally 282 patients), and 36 were identified in the external test set (totally 71 patients). Diabetes, smoking, coronary heart disease, and stenosis were found to be associated with the culprit lesions in the multivariate analysis. The diagnostic performance of Radscore was 0.674 and 0.609 for the training and external test data set. The nomogram, which includes clinical factors, stenosis, spotty calcification, and Radscore, demonstrated moderate values for the discrimination of symptomatic intracranial atherosclerotic lesions, with an AUC of 0.749 in the training set and an AUC of 0.736 in the external test set.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRadiomics of intracranial artery calcification in the culprit lesion may provide added value for identifying ASCI beyond stenosis and clinical factors. The nomogram incorporating both conventional and radiomics variables may serve as a potential diagnostic tool for stroke risk assessment in clinical settings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrial registration:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRetrospectively registered.\u003c/p\u003e","manuscriptTitle":"Nomogram of Intracranial Artery Calcification with Integration of CT-based Radiomics to Identify Culprit Lesions","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-10 10:39:21","doi":"10.21203/rs.3.rs-7763910/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-13T07:44:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-11T20:08:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-07T17:03:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160243356122420208989361751128354941660","date":"2025-11-01T22:35:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-29T14:03:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"228272593575138059815223833450628393893","date":"2025-10-29T13:33:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"314721432715432415158790323921794839204","date":"2025-10-29T10:48:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-29T06:51:34+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-13T01:59:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-13T01:58:50+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Neurology","date":"2025-10-02T05:30:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-neurology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nurl","sideBox":"Learn more about [BMC Neurology](http://bmcneurol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/nurl","title":"BMC Neurology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5930f2cd-82b9-4ba6-9ad8-9fe7b08ec164","owner":[],"postedDate":"November 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-10T05:08:31+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-10 10:39:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7763910","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7763910","identity":"rs-7763910","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}