Computational Risk Modeling of Carotid Atherosclerosis in Type 2 Diabetes: Insights from Doppler-Based Datat

Computational Risk Modeling of Carotid Atherosclerosis in Type 2 Diabetes: Insights from Doppler-Based Datat | 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 Computational Risk Modeling of Carotid Atherosclerosis in Type 2 Diabetes: Insights from Doppler-Based Datat Sónia Mateus, Patrícia Coelho, Francisco Rodrigues This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7160100/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Type 2 diabetes mellitus (T2DM) is a major risk factor for the development of atherosclerotic disease. Carotid Doppler ultrasound allows non-invasive detection of subclinical vascular alterations, offering a potential tool for early risk stratification. This study aimed to evaluate the association between T2DM and carotid atherosclerosis and to develop a computational predictive model integrating Doppler-derived data and clinical risk factors. Methods: A cross-sectional study involving 611 patients undergoing cervical Doppler ultrasound was conducted. Demographic data, cardiovascular risk factors, intima-media thickness (IMT), and carotid plaque presence were analyzed. Associations between T2DM and vascular alterations were assessed using chi-square and Mann-Whitney U tests. A multivariate logistic regression model was constructed to predict the presence of atherosclerotic changes, followed by ROC curve analysis to evaluate predictive performance. An ordinal regression analysis assessed predictors of stenosis severity. Results: T2DM was significantly associated with carotid plaque presence (82.4% vs. 67.6%; p=0.015), greater plaque burden (2.12 vs. 1.55 plaques/patient; p=0.001), and higher prevalence of stenosis >50% (13.7% vs. 4.9%; p=0.037). The multivariate predictive model identified T2DM (OR=2.56; 95% CI: 1.48–4.42; p<0.01), dyslipidemia (OR=1.87; p=0.02), and age (OR=1.04 per year; p<0.001) as independent predictors of carotid atherosclerotic alterations. The ROC curve analysis yielded an AUC of 0.836, demonstrating excellent discriminative ability. IMT, age, and hypertension were independently associated with increasing degrees of stenosis. Conclusion: T2DM significantly increases the risk of carotid atherosclerosis. The developed computational model, integrating clinical and Doppler-derived data, demonstrated robust predictive performance for vascular risk stratification in diabetic patients. These findings support the incorporation of vascular imaging and computational tools into preventive cardiovascular strategies. Type 2 diabetes mellitus carotid atherosclerosis Doppler ultrasound computational model vascular risk stratification predictive analytics cardiovascular prevention intima-media thickness Figures Figure 1 1. Introduction Diabetes mellitus (DM) is a chronic metabolic disorder characterized by persistent hyperglycemia resulting from insufficient insulin production or ineffective insulin action (Freitas, 2002 ). Insulin, secreted by pancreatic β-cells, is essential for glucose transport into cells, where it is utilized for energy production (Freitas, 2002 ). Two main types of DM are distinguished: type 1, characterized by autoimmune destruction of pancreatic β-cells leading to insulin deficiency, and type 2, which involves insulin resistance often accompanied by a relative insulin secretory defect (Rydén et al., 2007). According to the International Diabetes Federation (IDF) Diabetes Atlas, in 2021 approximately 537 million adults (aged 20–79 years) worldwide were living with diabetes, representing 10.5% of the global adult population. Over 90% of these cases are type 2 diabetes (International Diabetes Federation, 2021; Sun et al., 2022 ). Projections estimate this number will increase to 643 million by 2030 and 783 million by 2045 (Sun et al., 2022 ). Additionally, the World Health Organization (WHO) reported that approximately 830 million people had diabetes in 2022, marking a significant rise since 1990. This growth is attributed to urbanization, population aging, reduced physical activity, and increased prevalence of overweight and obesity (World Health Organization, 2022). Among the factors explaining this rise, obesity and arterial stiffness have been identified as independent predictors for developing type 2 diabetes (Zhou et al., 2023 ), contributing to the increasing global disease burden. In Portugal, data from 2014 indicated a prevalence of 13.1% among adults aged 20 to 79 years, corresponding to over one million affected individuals (Ramos et al., 2018). Complications associated with DM include retinopathy, nephropathy, neuropathy, and various macrovascular manifestations such as coronary artery disease, cerebrovascular disease, and peripheral arterial disease (Rydén et al., 2007; SBD, 2021). DM constitutes a major risk factor for cardiovascular events, including stroke, with 29% of stroke hospitalizations in Portugal in 2014 occurring in patients with diabetes (Ramos et al., 2018; Bonardi, 2005). Chronic hyperglycemia contributes to endothelial dysfunction and atherosclerosis development through mechanisms involving increased oxidative stress, inflammation, and impaired nitric oxide (NO) production by endothelial cells (Rydén et al., 2007; Sena et al., 2008). NO exerts vasoprotective effects by promoting vasodilation and inhibiting platelet aggregation and inflammatory responses. However, in the presence of risk factors such as DM, NO bioavailability is reduced, favoring atherosclerotic progression (Sena et al., 2008; Sharma et al., 2007). The evaluation of subclinical atherosclerosis can be performed using cervical Doppler ultrasound, which allows measurement of the carotid intima-media thickness (IMT), an early marker of atherosclerotic disease, as well as detection of atheromatous plaques (Ebrahim et al., 1999). Doppler ultrasound has proven useful not only in the assessment of carotid arteries but also in peripheral vascular studies, including functional and hemodynamic analysis in specific populations (Duarte-Mendes et al., 2020 ), supporting its broader applicability in vascular diagnostics. Studies have shown a higher prevalence of carotid atheromatosis in individuals with diabetes, reinforcing the link between DM and early vascular alterations (Freitas et al., 2008; Hong et al., 2015; Kardys et al., 2007). Nevertheless, many of these studies exhibit methodological limitations such as small sample sizes, lack of control groups, and incomplete analysis of plaque morphological characteristics (Camilo et al., 2017; Freitas & Brandão, 2020). Furthermore, diabetes is strongly associated with an increased risk of infections, particularly urinary tract infections (UTIs), which represent a significant burden in diabetic populations due to altered immune responses and glycemic control (Rodrigues, Barroso, & A., 2011; Branco, Coelho, & Rodrigues, 2024 ). The higher susceptibility to UTIs in individuals with diabetes highlights the importance of comprehensive clinical monitoring to prevent complications. In recent years, advances in computational modeling and data analysis have transformed clinical research and risk assessment strategies. Computational methods—such as multivariate statistical modeling, machine learning, and artificial intelligence—are increasingly applied to biomedical datasets to enhance disease prediction, stratify patient risk, and support clinical decision-making (Esteva et al., 2019 ; Jordan et al., 2019). These approaches allow the integration of large, complex datasets, like Doppler ultrasound-derived parameters, into predictive models that can objectively quantify vascular risk. Predictive modeling, particularly using logistic regression and receiver operating characteristic (ROC) analysis, remains a robust and interpretable computational tool with direct clinical application (Fawcett, 2006 ). Therefore, the present study aims to evaluate the association between type 2 diabetes mellitus and the presence of atherosclerotic changes in carotid arteries through a retrospective analysis of cervical Doppler ultrasound studies performed in a hospital population. Additionally, this study seeks to develop a computational risk model based on Doppler-derived data and clinical parameters to support vascular risk stratification in patients with T2DM. This computational approach has potential relevance as a decision-support tool in clinical practice, contributing to early identification of high-risk patients and optimizing therapeutic strategies. 2. Materials and Methods A cross-sectional, analytical observational study was conducted, involving a non-probabilistic consecutive sample of 611 patients who underwent cervical carotid Doppler ultrasound between August 2011 and August 2024 at the ultrasound laboratory of a district hospital. The sample included adults aged 37 to 93 years, of both sexes, encompassing various cardiovascular risk profiles, including dyslipidemia and hypertension. Data collection involved retrospective consultation of the LUSCAN database and analysis of archived Doppler ultrasound examinations. The study aimed to identify carotid vascular alterations in individuals diagnosed with type 2 diabetes mellitus (T2DM) through extraction of the following variables: sex, age (continuous variable), cardiovascular risk factors, personal medical history, intima-media thickness (IMT; continuous variable), presence of vascular alterations in right and left carotid arteries, and classification of atherosclerotic plaques, when present (Table 1 ). Table 1 Collected qualitative variables and their respective categories Variables Cathegories Gender Male or female Race Caucasian or other Diabetes type 2 Yes or no Arterial hypertencion Yes or no Choresterol Yes or no Smoke smoke, ex-smoker, never Smoke Stroke personal history Yes or no Acute myocardial infarction Yes or no Atrial fibrillation Yes or no Carotid ultrasound Normal or phatological Vascular pathology Normal, < 70% stenosis, ≥ %70 stenosis or occlusion 2.1 Statistical and Computational Analysis Descriptive statistics were performed using IBM SPSS Statistics® v22.0 (IBM Corp., Armonk, NY). Continuous variables were described as means ± standard deviations, while categorical variables were expressed as absolute and relative frequencies. Associations between T2DM and carotid atherosclerotic changes were analyzed using Pearson’s chi-square test or Fisher’s exact test, as appropriate. Ordinal variables, including stenosis degree, were compared using the linear-by-linear association chi-square test. Normality of continuous variables (age and IMT) was assessed via the Kolmogorov-Smirnov test. Due to non-normal distribution, inter-group comparisons were performed using the Mann-Whitney U test. A computational predictive model was developed using multivariate binary logistic regression, incorporating T2DM, age, sex, hypertension, dyslipidemia, and smoking status as independent variables. The dependent outcome was defined as the presence or absence of carotid atherosclerotic alterations. Predictive outputs were expressed as Odds Ratios (OR ) with corresponding 95% confidence intervals (CI). Statistical significance was defined as p < 0.05. To evaluate the predictive performance of the model, a Receiver Operating Characteristic (ROC ) curve analysis was conducted, with calculation of the Area Under the Curve (AUC) to quantify discriminatory capacity. This ROC analysis functioned as a computational validation tool for assessing model accuracy and clinical applicability. The integration of multivariate modeling and ROC analysis was considered a computational approach to vascular risk stratification, leveraging Doppler-derived clinical and hemodynamic data. This study was approved by the Ethics Committee of the Local Health Unit of Central Alentejo and conducted in accordance with the principles of the Declaration of Helsinki. The authors declare no conflicts of interest. 3. Results 3.1 Sample Characterization A total of 611 patients who underwent cervical carotid Doppler ultrasound were included in the study. The mean age was 67.8 ± 11.2 years, with 51.2% males. All participants were Caucasian. Among the total population: · 261 patients (42.7%) had type 2 diabetes mellitus (T2DM). · The prevalence of hypertension was 81.2%, and dyslipidemia was recorded in 58.9% of participants. · Current or former smokers accounted for 29.3% of the sample. Detailed demographic and cardiovascular risk profiles for both the diabetic and non-diabetic groups are presented in Table 1. 3.2 Comparison of Carotid Atherosclerotic Findings Between T2DM and Control Groups A significant association was observed between the presence of T2DM and carotid atherosclerotic changes: · Atherosclerotic plaques were identified in 82.4% of diabetic patients compared to 67.6% in non-diabetic controls (p = 0.015). · The mean number of plaques per patient was significantly higher in the T2DM group (2.12 vs. 1.55, p = 0.001). · Stenosis >50% was present in 13.7% of diabetic patients versus 4.9% in the control group (p = 0.037). · No statistically significant difference was found in mean intima-media thickness (IMT) between the two groups (p = 0.101), although diabetic patients tended to present thicker IMT values. When stratifying vascular alterations: · Diabetic patients had higher rates of significant stenosis (≥70%) and arterial occlusion compared to non-diabetic individuals (Table 2). · The association between the presence of T2DM and more severe degrees of stenosis was statistically significant (p = 0.004). 3.3 Multivariate Computational Risk Model A computational predictive model was constructed using binary logistic regression, aiming to assess independent predictors of carotid atherosclerotic alterations (presence of plaques and/or stenosis). Independent variables included: · T2DM status · Age · Sex · Hypertension · Dyslipidemia · Smoking status Significant independent predictors identified: · Type 2 Diabetes Mellitus: OR = 2.56 (95% CI: 1.48–4.42), p < 0.01 · Age (per year): OR = 1.04 (95% CI: 1.02–1.06), p < 0.001 · Dyslipidemia: OR = 1.87 (95% CI: 1.10–3.21), p = 0.02 Other variables (sex, hypertension, smoking status) did not reach statistical significance in the final model. 3.4 Model Validation: ROC Curve Analysis The predictive performance of the multivariate model was assessed using a Receiver Operating Characteristic (ROC) curve. · The Area Under the Curve (AUC) was 0.836 (95% CI: 0.794–0.878), indicating excellent discriminative ability of the model in classifying patients according to carotid atherosclerotic risk (Figure 1). 3.5 Ordinal Analysis of Stenosis Severity To explore predictors of stenosis severity, an ordinal logistic regression model was developed, incorporating IMT and vascular risk factors: · Age, IMT, and hypertension emerged as significant predictors of increasing stenosis severity: o Age: OR = 1.041 (95% CI: 1.017–1.065), p = 0.001 o IMT: OR = 174.57 (95% CI: 44.4–682.1), p < 0.001 o Hypertension: p = 0.008 Interestingly, T2DM status did not remain a significant predictor of stenosis severity after multivariate adjustment (p = 0.361), suggesting that its contribution may be primarily related to earlier stages of atherosclerotic progression. The model demonstrated acceptable explanatory power (Nagelkerke R² = 0.192) and good calibration (Hosmer-Lemeshow p = 0.377). 3.6 Risk Stratification Based on the Computational Model By combining clinical data and Doppler ultrasound findings, the computational model enabled effective vascular risk stratification: · Patients with T2DM, dyslipidemia, and advanced age exhibited the highest predicted probabilities of significant carotid atherosclerotic alterations. · The predictive model, supported by ROC curve validation, provides a practical tool for early identification of high-risk individuals within the clinical setting. 3.7 Summary of Key Findings · T2DM was significantly associated with the presence and severity of carotid atherosclerosis. · Age and dyslipidemia contributed independently to vascular risk. · A computational model integrating clinical and Doppler-derived data demonstrated excellent performance for risk stratification. · IMT emerged as a key indicator of stenosis severity. 4. Discussion The present study reinforces the strong association between type 2 diabetes mellitus (T2DM) and carotid atherosclerosis, confirming that individuals with T2DM exhibit a higher prevalence of carotid plaques, greater plaque burden, and more severe degrees of stenosis when compared to non-diabetic individuals. These findings align with previous evidence demonstrating the pro-atherogenic effects of chronic hyperglycemia, endothelial dysfunction, and systemic inflammation observed in diabetic populations (Lu et al., 2023 ; Wu et al., 2022 ). Notably, the identification of T2DM as an independent predictor of carotid atherosclerosis even after multivariate adjustment underlines its critical role in early vascular damage pathophysiology. A key contribution of this study lies in the integration of a computational predictive model for vascular risk stratification, combining clinical risk factors and Doppler ultrasound-derived data within a multivariate logistic regression framework. The model identified T2DM, dyslipidemia, and age as independent predictors of carotid atherosclerotic alterations, in agreement with established pathophysiological mechanisms underlying vascular disease progression in diabetic individuals (Telles et al., 2008 ; Ifrim & Vasilescu, 2004 ). These findings reinforce the importance of early vascular evaluation in T2DM patients, even in the absence of overt cardiovascular symptoms. The predictive performance of the developed model was further validated through ROC curve analysis, with an area under the curve (AUC) of 0.836, confirming its excellent discriminatory power. This result emphasizes the potential applicability of such computational tools in clinical practice, particularly when anchored in structured, objective imaging data like Doppler ultrasound findings (Dossabhoy et al., 2023 ; Lareyre et al., 2023 ). By enabling the stratification of vascular risk in a reproducible and interpretable manner, the model can support individualized patient assessment and targeted preventive strategies. From a computational standpoint, the use of logistic regression offers several practical advantages in clinical settings: it generates transparent, statistically robust, and easily interpretable models, thus facilitating clinical adoption without reliance on complex machine learning infrastructures (Cheng et al., 2022 ; Wu et al., 2024 ). Furthermore, its probabilistic output allows the generation of personalized risk estimates, enhancing clinical decision-making and resource allocation (Taslimitehrani et al., 2016 ; Lv et al., 2021 ). In this context, the incorporation of Doppler-derived morphological and hemodynamic parameters into computational models represents a significant advance beyond traditional risk factor-based assessments. The analysis using ordinal logistic regression provided additional insights, revealing that variables such as age, intima-media thickness (IMT), and hypertension were strongly associated with stenosis severity. Interestingly, T2DM did not independently predict more severe stenosis after multivariate adjustment, suggesting that its role may be more relevant in the initiation and early progression of carotid atherosclerosis. This highlights the complex interplay of factors in the progression of vascular disease and underlines the value of integrating imaging data, such as IMT measurements, into risk models for a more precise stratification. The early detection of subclinical atherosclerosis in high-risk populations, such as patients with T2DM, is fundamental for timely therapeutic intervention and the reduction of cardiovascular events (Robinson et al., 2009 ; Fuster et al., 2010 ). The computational model developed in this study could serve as a complementary clinical decision-support tool, assisting healthcare professionals in identifying high-risk patients and initiating personalized preventive measures. Nevertheless, the study presents some limitations. Its retrospective, single-center design may limit the generalizability of findings. While the computational model demonstrates good performance, it remains based on a conventional statistical approach. Future research should explore the integration of advanced machine learning algorithms, capable of managing larger datasets and capturing nonlinear interactions, which could enhance predictive accuracy and external applicability. Moreover, external validation in independent cohorts will be essential before clinical implementation of the proposed model. In conclusion, this study confirms the significant impact of T2DM on carotid atherosclerosis progression and introduces a clinically applicable, computationally validated risk stratification model integrating Doppler-derived data. These findings support the incorporation of vascular imaging and computational tools into cardiovascular risk assessment pathways for diabetic patients. 5. Conclusions This study confirms the significant association between type 2 diabetes mellitus (T2DM) and the development of carotid atherosclerosis, with diabetic individuals presenting a higher prevalence of plaques, increased plaque burden, and more severe degrees of carotid stenosis compared to non-diabetic controls. The integration of Doppler ultrasound-derived data and clinical variables into a computational predictive model allowed accurate risk stratification, identifying T2DM, dyslipidemia, and age as independent predictors of carotid vascular alterations. The logistic regression-based model, validated through ROC curve analysis (AUC = 0.836), demonstrated strong discriminatory power, highlighting its potential clinical applicability as a decision-support tool for early identification of high-risk individuals. Additionally, the incorporation of morphological markers, such as intima-media thickness (IMT), proved essential in assessing the severity of carotid stenosis. These findings underscore the value of integrating vascular imaging and computational risk models in cardiovascular risk assessment strategies for diabetic populations, supporting personalized and preventive approaches in clinical practice. Future studies should focus on external validation and the potential refinement of predictive models using advanced machine learning techniques to optimize risk prediction and improve clinical outcomes. Declarations Author Contributions : Conceptualization, S.M. and P.C.; methodology, S.M; software F.R.; validation, F.R., P.C. and S.M.; formal analysis, F.R; investigation, F.R.; data curation, P.C. and S.M.; writing—preparation of the original draft, S.M. and P.C.; writing—review and editing F.R.; visualization, P.C. and F.R; supervision, S.M.; project coordination, S.M. All authors have read and agreed to the published version of the manuscript. Funding : This research received no external funding. Institutional Review Board Statement : This work was approved by the Ethics Committee from the Local Health Unit of Central Alentejo, and all ethical precepts were scrupulously respected by the researchers. Informed consent was waived, given the retrospective nature and the fact that no user-identifying data were used. Consent Statement : Not applicable, as this was a data review. Data Availability Statement : Data may be shared, in case of a duly justified request, considering the necessary protection. Conflicts of Interest : The authors declare no conflicts of interest Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References Branco, M. C., Coelho, P., & Rodrigues, F. (2024). Urinary tract infections in a single hospital in central Portugal, a 5-year analysis. Microbiology Research , 15(2), 850–863. https://doi.org/10.3390/microbiolres15020055 Cheng, X., Han, W., Liang, Y., Lin, X., Luo, J., Zhong, W., & Chen, D. (2022). Risk Prediction of Coronary Artery Stenosis in Patients with Coronary Heart Disease Based on Logistic Regression and Artificial Neural Network. Computational and mathematical methods in medicine , 2022 , 3684700. https://doi.org/10.1155/2022/3684700 Dossabhoy, S. S., Ho, V. T., Ross, E. G., Rodriguez, F., & Arya, S. (2023). Artificial intelligence in clinical workflow processes in vascular surgery and beyond. Seminars in vascular surgery , 36 (3), 401–412. https://doi.org/10.1053/j.semvascsurg.2023.07.002 Duarte-Mendes, P., Paulo, R., Coelho, P., Rodrigues, F., Marques, V., & Mateus, S. (2020). Variability of Lower Limb Artery Systolic-Diastolic Velocities in Futsal Athletes and Non-Athletes: Evaluation by Arterial Doppler Ultrasound. International journal of environmental research and public health, 17(2), 570. https://doi.org/10.3390/ijerph17020570 Federação Internacional de Diabetes (IDF). (2021). IDF Diabetes Atlas (10th ed.). https://diabetesatlas.org/ Freitas, D. P. (2002). Diabetes Mellitus tipo 2 [Internet]. ALERT. http://www.alert-online.com/pt/medical-guide/diabetes-mellitus-tipo-2 Esteva, A., Robicquet, A., Ramsundar, B., et al. (2019). A guide to deep learning in healthcare. Nature Medicine, 25 (1), 24-29. https://doi.org/10.1038/s41591-018-0316-z Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27 (8), 861-874. https://doi.org/10.1016/j.patrec.2005.10.010 Freitas, P. (2008). Aterosclerose carotídea avaliada pelo eco-Doppler: associação com fatores de risco e doenças arteriais sistêmicas. Jornal Vascular Brasileiro, 7 (4), 298–307. Fuster, V., Lois, F., & Franco, M. (2010). Early identification of atherosclerotic disease by noninvasive imaging. Nature reviews. Cardiology , 7 (6), 327–333. https://doi.org/10.1038/nrcardio.2010.54 Ifrim, S., & Vasilescu, R. (2004). Early detection of atherosclerosis in type 2 diabetic patients by endothelial dysfunction and intima-media thickness. Romanian journal of internal medicine = Revue roumaine de medecine interne , 42 (2), 343–354. Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. Science, 349 (6245), 255-260. https://doi.org/10.1126/science.aaa8415 Lareyre, F., Behrendt, C. A., Chaudhuri, A., Lee, R., Carrier, M., Adam, C., Lê, C. D., & Raffort, J. (2023). Applications of artificial intelligence for patients with peripheral artery disease. Journal of vascular surgery , 77 (2), 650–658.e1. https://doi.org/10.1016/j.jvs.2022.07.160 Lu, S. X., Wu, T. W., Chou, C. L., Cheng, C. F., & Wang, L. Y. (2023). Combined effects of hypertension, hyperlipidemia, and diabetes mellitus on the presence and severity of carotid atherosclerosis in community-dwelling elders: A community-based study. Journal of the Chinese Medical Association : JCMA , 86 (2), 220–226. https://doi.org/10.1097/JCMA.0000000000000839 Lv, H., Yang, X., Wang, B., Wang, S., Du, X., Tan, Q., Hao, Z., Liu, Y., Yan, J., & Xia, Y. (2021). Machine Learning-Driven Models to Predict Prognostic Outcomes in Patients Hospitalized With Heart Failure Using Electronic Health Records: Retrospective Study. Journal of medical Internet research , 23 (4), e24996. https://doi.org/10.2196/24996 Organização Mundial da Saúde (OMS). (2022). Diabetes . https://www.who.int/news-room/fact-sheets/detail/diabetes Taslimitehrani, V., Dong, G., Pereira, N. L., Panahiazar, M., & Pathak, J. (2016). Developing EHR-driven heart failure risk prediction models using CPXR(Log) with the probabilistic loss function. Journal of biomedical informatics , 60 , 260–269. https://doi.org/10.1016/j.jbi.2016.01.009 Ryden, L., Standl, E., Bartnik, M., Van den Berghe, G., Betteridge, J., de Boer, M.-J., et al. (2007). Guidelines on diabetes, pre-diabetes, and cardiovascular diseases: full text. European Heart Journal Supplements, 9 (Suppl C), C3–C74. https://doi.org/10.1093/eurheartj/ehl261 Robinson, J. G., Fox, K. M., Bullano, M. F., Grandy, S., & SHIELD Study Group (2009). Atherosclerosis profile and incidence of cardiovascular events: a population-based survey. BMC cardiovascular disorders , 9 , 46. https://doi.org/10.1186/1471-2261-9-46 Rodrigues, J. B., Barroso, F. P. D., & A. (2011). Etiology and bacterial susceptibility to urinary tract infections | Etiologia e sensibilidade bacteriana em infecções do trato urinário. Revista Portuguesa de Saúde Pública , 29(2), 123–131. Sociedade Portuguesa de Aterosclerose. (2023). Highlights do Congresso Europeu de Aterosclerose 2023 . https://spaterosclerose.org/noticias/highlights-do-congresso-europeu-de-aterosclerose-2023/ Sun, H., Saeedi, P., Karuranga, S., Pinkepank, M., Ogurtsova, K., Duncan, B. B., ... & Williams, R. (2022). IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Research and Clinical Practice, 183 , 109119. https://doi.org/10.1016/j.diabres.2021.109119 Telles, R. W., Lanna, C. C., Ferreira, G. A., Souza, A. J., Navarro, T. P., & Ribeiro, A. L. (2008). Carotid atherosclerotic alterations in systemic lupus erythematosus patients treated at a Brazilian university setting. Lupus , 17 (2), 105–113. https://doi.org/10.1177/0961203307085312 Wu, T. W., Chou, C. L., Cheng, C. F., Lu, S. X., & Wang, L. Y. (2022). Prevalences of diabetes mellitus and carotid atherosclerosis and their relationships in middle-aged adults and elders: a community-based study. Journal of the Formosan Medical Association = Taiwan yi zhi , 121 (6), 1133–1140. https://doi.org/10.1016/j.jfma.2021.10.005 Wu, Y., Xin, B., Wan, Q., Ren, Y., & Jiang, W. (2024). Risk factors and prediction models for cardiovascular complications of hypertension in older adults with machine learning: A cross-sectional study. Heliyon , 10 (6), e27941. https://doi.org/10.1016/j.heliyon.2024.e27941 Zhou, Z., et al. (2023). Arterial stiffness and obesity as predictors of diabetes. JMIR Public Health and Surveillance, 10 , e46088. https://doi.org/10.2196/46088 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-7160100","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":489802887,"identity":"1e45efc3-6006-4c36-90a0-1283e90ac22a","order_by":0,"name":"Sónia Mateus","email":"","orcid":"","institution":"Polytechnic Institute of Castelo Branco","correspondingAuthor":false,"prefix":"","firstName":"Sónia","middleName":"","lastName":"Mateus","suffix":""},{"id":489802888,"identity":"f8f70b61-b051-44ac-a30d-6016472511f7","order_by":1,"name":"Patrícia Coelho","email":"","orcid":"","institution":"Polytechnic Institute of Castelo Branco","correspondingAuthor":false,"prefix":"","firstName":"Patrícia","middleName":"","lastName":"Coelho","suffix":""},{"id":489802889,"identity":"d466911f-1c09-4452-9d0c-9ee39d06e53a","order_by":2,"name":"Francisco Rodrigues","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIiWNgGAWjYBAC9gYQaQDEEkDMAyYZGB8kMDAk4NLCcwCLFmYDiBZmPFoY4FrAgA3Exq1F+vCzBz8K6uT5pXsfPni7x8Juw/H2ZxUPd9jlMUj3H8CqhS/N3LDH4LDhzDnHjQ3nPJNI3nDmjNmNxDPJxQwyh7HaYs/DYCbNYHAgweBGGps0zwGJZIMbOWw3EtsOJDZIJGN3GA/7N6CWOiQt958/K8CvhQdkCzNci53BDQYzBgJayiQhfjnGbDjngESC5JkcYwmQX9hkDhvgcNg2iR9/QCHWxvjgzYE6e77jxx9+/AkMMX7pxgdYrUEHiQ0gkrEBGjvEAHsGmBYGYrWMglEwCkbBcAcAAGBdbK8MgJoAAAAASUVORK5CYII=","orcid":"","institution":"Polytechnic Institute of Castelo Branco","correspondingAuthor":true,"prefix":"","firstName":"Francisco","middleName":"","lastName":"Rodrigues","suffix":""}],"badges":[],"createdAt":"2025-07-18 18:38:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7160100/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7160100/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87705642,"identity":"752112ff-c00f-4a10-8105-8540a27b59b3","added_by":"auto","created_at":"2025-07-28 07:57:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eThis image is not available with this version.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7160100/v1/35b72fb7b339e51944ee6e49.png"},{"id":87913087,"identity":"c0236da1-895c-4fd7-b796-75547c01a9cd","added_by":"auto","created_at":"2025-07-30 10:17:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":600243,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7160100/v1/32c65949-737f-4177-a86c-bed79fdf0269.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Computational Risk Modeling of Carotid Atherosclerosis in Type 2 Diabetes: Insights from Doppler-Based Datat","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eDiabetes mellitus (DM) is a chronic metabolic disorder characterized by persistent hyperglycemia resulting from insufficient insulin production or ineffective insulin action (Freitas, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Insulin, secreted by pancreatic β-cells, is essential for glucose transport into cells, where it is utilized for energy production (Freitas, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Two main types of DM are distinguished: type 1, characterized by autoimmune destruction of pancreatic β-cells leading to insulin deficiency, and type 2, which involves insulin resistance often accompanied by a relative insulin secretory defect (Ryd\u0026eacute;n et al., 2007).\u003c/p\u003e\u003cp\u003eAccording to the International Diabetes Federation (IDF) Diabetes Atlas, in 2021 approximately 537\u0026nbsp;million adults (aged 20\u0026ndash;79 years) worldwide were living with diabetes, representing 10.5% of the global adult population. Over 90% of these cases are type 2 diabetes (International Diabetes Federation, 2021; Sun et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Projections estimate this number will increase to 643\u0026nbsp;million by 2030 and 783\u0026nbsp;million by 2045 (Sun et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Additionally, the World Health Organization (WHO) reported that approximately 830\u0026nbsp;million people had diabetes in 2022, marking a significant rise since 1990. This growth is attributed to urbanization, population aging, reduced physical activity, and increased prevalence of overweight and obesity (World Health Organization, 2022). Among the factors explaining this rise, obesity and arterial stiffness have been identified as independent predictors for developing type 2 diabetes (Zhou et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), contributing to the increasing global disease burden.\u003c/p\u003e\u003cp\u003eIn Portugal, data from 2014 indicated a prevalence of 13.1% among adults aged 20 to 79 years, corresponding to over one million affected individuals (Ramos et al., 2018). Complications associated with DM include retinopathy, nephropathy, neuropathy, and various macrovascular manifestations such as coronary artery disease, cerebrovascular disease, and peripheral arterial disease (Ryd\u0026eacute;n et al., 2007; SBD, 2021). DM constitutes a major risk factor for cardiovascular events, including stroke, with 29% of stroke hospitalizations in Portugal in 2014 occurring in patients with diabetes (Ramos et al., 2018; Bonardi, 2005).\u003c/p\u003e\u003cp\u003eChronic hyperglycemia contributes to endothelial dysfunction and atherosclerosis development through mechanisms involving increased oxidative stress, inflammation, and impaired nitric oxide (NO) production by endothelial cells (Ryd\u0026eacute;n et al., 2007; Sena et al., 2008). NO exerts vasoprotective effects by promoting vasodilation and inhibiting platelet aggregation and inflammatory responses. However, in the presence of risk factors such as DM, NO bioavailability is reduced, favoring atherosclerotic progression (Sena et al., 2008; Sharma et al., 2007).\u003c/p\u003e\u003cp\u003eThe evaluation of subclinical atherosclerosis can be performed using cervical Doppler ultrasound, which allows measurement of the carotid intima-media thickness (IMT), an early marker of atherosclerotic disease, as well as detection of atheromatous plaques (Ebrahim et al., 1999). Doppler ultrasound has proven useful not only in the assessment of carotid arteries but also in peripheral vascular studies, including functional and hemodynamic analysis in specific populations (Duarte-Mendes et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), supporting its broader applicability in vascular diagnostics. Studies have shown a higher prevalence of carotid atheromatosis in individuals with diabetes, reinforcing the link between DM and early vascular alterations (Freitas et al., 2008; Hong et al., 2015; Kardys et al., 2007). Nevertheless, many of these studies exhibit methodological limitations such as small sample sizes, lack of control groups, and incomplete analysis of plaque morphological characteristics (Camilo et al., 2017; Freitas \u0026amp; Brand\u0026atilde;o, 2020).\u003c/p\u003e\u003cp\u003eFurthermore, diabetes is strongly associated with an increased risk of infections, particularly urinary tract infections (UTIs), which represent a significant burden in diabetic populations due to altered immune responses and glycemic control (Rodrigues, Barroso, \u0026amp; A., 2011; Branco, Coelho, \u0026amp; Rodrigues, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The higher susceptibility to UTIs in individuals with diabetes highlights the importance of comprehensive clinical monitoring to prevent complications.\u003c/p\u003e\u003cp\u003eIn recent years, advances in computational modeling and data analysis have transformed clinical research and risk assessment strategies. Computational methods\u0026mdash;such as multivariate statistical modeling, machine learning, and artificial intelligence\u0026mdash;are increasingly applied to biomedical datasets to enhance disease prediction, stratify patient risk, and support clinical decision-making (Esteva et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jordan et al., 2019). These approaches allow the integration of large, complex datasets, like Doppler ultrasound-derived parameters, into predictive models that can objectively quantify vascular risk. Predictive modeling, particularly using logistic regression and receiver operating characteristic (ROC) analysis, remains a robust and interpretable computational tool with direct clinical application (Fawcett, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTherefore, the present study aims to evaluate the association between type 2 diabetes mellitus and the presence of atherosclerotic changes in carotid arteries through a retrospective analysis of cervical Doppler ultrasound studies performed in a hospital population. Additionally, this study seeks to develop a computational risk model based on Doppler-derived data and clinical parameters to support vascular risk stratification in patients with T2DM. This computational approach has potential relevance as a decision-support tool in clinical practice, contributing to early identification of high-risk patients and optimizing therapeutic strategies.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eA cross-sectional, analytical observational study was conducted, involving a non-probabilistic consecutive sample of 611 patients who underwent cervical carotid Doppler ultrasound between August 2011 and August 2024 at the ultrasound laboratory of a district hospital. The sample included adults aged 37 to 93 years, of both sexes, encompassing various cardiovascular risk profiles, including dyslipidemia and hypertension.\u003c/p\u003e\u003cp\u003eData collection involved retrospective consultation of the LUSCAN database and analysis of archived Doppler ultrasound examinations. The study aimed to identify carotid vascular alterations in individuals diagnosed with type 2 diabetes mellitus (T2DM) through extraction of the following variables: sex, age (continuous variable), cardiovascular risk factors, personal medical history, intima-media thickness (IMT; continuous variable), presence of vascular alterations in right and left carotid arteries, and classification of atherosclerotic plaques, when present (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCollected qualitative variables and their respective categories\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCathegories\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMale or female\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRace\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCaucasian or other\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDiabetes type 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eArterial hypertencion\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChoresterol\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoke\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003esmoke, ex-smoker, never Smoke\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStroke personal history\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcute myocardial infarction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAtrial fibrillation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes or no\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCarotid ultrasound\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNormal or phatological\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVascular pathology\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNormal, \u0026lt;\u0026thinsp;70% stenosis, \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026ge;\u003c/span\u003e%70 stenosis or occlusion\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Statistical and Computational Analysis\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eDescriptive statistics were performed using IBM SPSS Statistics\u0026reg; v22.0 (IBM Corp., Armonk, NY). Continuous variables were described as means\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviations, while categorical variables were expressed as absolute and relative frequencies.\u003c/p\u003e\u003cp\u003eAssociations between T2DM and carotid atherosclerotic changes were analyzed using Pearson\u0026rsquo;s chi-square test or Fisher\u0026rsquo;s exact test, as appropriate. Ordinal variables, including stenosis degree, were compared using the linear-by-linear association chi-square test. Normality of continuous variables (age and IMT) was assessed via the Kolmogorov-Smirnov test. Due to non-normal distribution, inter-group comparisons were performed using the Mann-Whitney U test.\u003c/p\u003e\u003cp\u003eA computational predictive model was developed using multivariate binary logistic regression, incorporating T2DM, age, sex, hypertension, dyslipidemia, and smoking status as independent variables. The dependent outcome was defined as the presence or absence of carotid atherosclerotic alterations. Predictive outputs were expressed as Odds Ratios (OR\u003cb\u003e)\u003c/b\u003e with corresponding 95% confidence intervals (CI). Statistical significance was defined as p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\u003cp\u003eTo evaluate the predictive performance of the model, \u003cb\u003ea\u003c/b\u003e Receiver Operating Characteristic (ROC\u003cb\u003e)\u003c/b\u003e curve analysis was conducted, with calculation of the Area Under the Curve (AUC) to quantify discriminatory capacity. This ROC analysis functioned as a computational validation tool for assessing model accuracy and clinical applicability.\u003c/p\u003e\u003cp\u003eThe integration of multivariate modeling and ROC analysis was considered a computational approach to vascular risk stratification, leveraging Doppler-derived clinical and hemodynamic data.\u003c/p\u003e\u003cp\u003e This study was approved by the Ethics Committee of the Local Health Unit of Central Alentejo and conducted in accordance with the principles of the Declaration of Helsinki. The authors declare no conflicts of interest.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003ch3\u003e\u003cu\u003e3.1 Sample Characterization\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eA total of 611 patients who underwent cervical carotid Doppler ultrasound were included in the study. The mean age was 67.8 \u0026plusmn; 11.2 years, with 51.2% males. All participants were Caucasian. Among the total population:\u003c/p\u003e\n\u003cp\u003e\u0026middot; 261 patients (42.7%) had type 2 diabetes mellitus (T2DM).\u003c/p\u003e\n\u003cp\u003e\u0026middot; The prevalence of hypertension was 81.2%, and dyslipidemia was recorded in 58.9% of participants.\u003c/p\u003e\n\u003cp\u003e\u0026middot; Current or former smokers accounted for 29.3% of the sample.\u003c/p\u003e\n\u003cp\u003eDetailed demographic and cardiovascular risk profiles for both the diabetic and non-diabetic groups are presented in\u0026nbsp;Table 1.\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.2 Comparison of Carotid Atherosclerotic Findings Between T2DM and Control Groups\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eA significant association was observed between the presence of T2DM and carotid atherosclerotic changes:\u003c/p\u003e\n\u003cp\u003e\u0026middot; Atherosclerotic plaques were identified in 82.4% of diabetic patients compared to 67.6% in non-diabetic controls (p = 0.015).\u003c/p\u003e\n\u003cp\u003e\u0026middot; The mean number of plaques per patient was significantly higher in the T2DM group (2.12 vs. 1.55, p = 0.001).\u003c/p\u003e\n\u003cp\u003e\u0026middot; Stenosis \u0026gt;50% was present in 13.7% of diabetic patients versus 4.9% in the control group (p = 0.037).\u003c/p\u003e\n\u003cp\u003e\u0026middot; No statistically significant difference was found in mean intima-media thickness (IMT) between the two groups (p = 0.101), although diabetic patients tended to present thicker IMT values.\u003c/p\u003e\n\u003cp\u003eWhen stratifying vascular alterations:\u003c/p\u003e\n\u003cp\u003e\u0026middot; Diabetic patients had higher rates of significant stenosis (\u0026ge;70%) and arterial occlusion compared to non-diabetic individuals (Table 2).\u003c/p\u003e\n\u003cp\u003e\u0026middot; The association between the presence of T2DM and more severe degrees of stenosis was statistically significant (p = 0.004).\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.3 Multivariate Computational Risk Model\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eA computational predictive model was constructed using binary logistic regression, aiming to assess independent predictors of carotid atherosclerotic alterations (presence of plaques and/or stenosis). Independent variables included:\u003c/p\u003e\n\u003cp\u003e\u0026middot; T2DM status\u003c/p\u003e\n\u003cp\u003e\u0026middot; Age\u003c/p\u003e\n\u003cp\u003e\u0026middot; Sex\u003c/p\u003e\n\u003cp\u003e\u0026middot; Hypertension\u003c/p\u003e\n\u003cp\u003e\u0026middot; Dyslipidemia\u003c/p\u003e\n\u003cp\u003e\u0026middot; Smoking status\u003c/p\u003e\n\u003cp\u003eSignificant independent predictors identified:\u003c/p\u003e\n\u003cp\u003e\u0026middot; Type 2 Diabetes Mellitus: OR = 2.56 (95% CI: 1.48\u0026ndash;4.42), p \u0026lt; 0.01\u003c/p\u003e\n\u003cp\u003e\u0026middot; Age (per year): OR = 1.04 (95% CI: 1.02\u0026ndash;1.06), p \u0026lt; 0.001\u003c/p\u003e\n\u003cp\u003e\u0026middot; Dyslipidemia: OR = 1.87 (95% CI: 1.10\u0026ndash;3.21), p = 0.02\u003c/p\u003e\n\u003cp\u003eOther variables (sex, hypertension, smoking status) did not reach statistical significance in the final model.\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.4 Model Validation: ROC Curve Analysis\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eThe predictive performance of the multivariate model was assessed using a Receiver Operating Characteristic (ROC) curve.\u003c/p\u003e\n\u003cp\u003e\u0026middot; The Area Under the Curve (AUC) was 0.836 (95% CI: 0.794\u0026ndash;0.878), indicating excellent discriminative ability of the model in classifying patients according to carotid atherosclerotic risk (Figure 1).\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.5 Ordinal Analysis of Stenosis Severity\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eTo explore predictors of stenosis severity, an ordinal logistic regression model was developed, incorporating IMT and vascular risk factors:\u003c/p\u003e\n\u003cp\u003e\u0026middot; Age, IMT, and hypertension emerged as significant predictors of increasing stenosis severity:\u003c/p\u003e\n\u003cp\u003eo Age: OR = 1.041 (95% CI: 1.017\u0026ndash;1.065), p = 0.001\u003c/p\u003e\n\u003cp\u003eo IMT: OR = 174.57 (95% CI: 44.4\u0026ndash;682.1), p \u0026lt; 0.001\u003c/p\u003e\n\u003cp\u003eo Hypertension: p = 0.008\u003c/p\u003e\n\u003cp\u003eInterestingly, T2DM status did not remain a significant predictor of stenosis severity after multivariate adjustment (p = 0.361), suggesting that its contribution may be primarily related to earlier stages of atherosclerotic progression.\u003c/p\u003e\n\u003cp\u003eThe model demonstrated acceptable explanatory power (Nagelkerke R\u0026sup2; = 0.192) and good calibration (Hosmer-Lemeshow p = 0.377).\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.6 Risk Stratification Based on the Computational Model\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eBy combining clinical data and Doppler ultrasound findings, the computational model enabled effective vascular risk stratification:\u003c/p\u003e\n\u003cp\u003e\u0026middot; Patients with T2DM, dyslipidemia, and advanced age exhibited the highest predicted probabilities of significant carotid atherosclerotic alterations.\u003c/p\u003e\n\u003cp\u003e\u0026middot; The predictive model, supported by ROC curve validation, provides a practical tool for early identification of high-risk individuals within the clinical setting.\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003e3.7 Summary of Key Findings\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003e\u0026middot; T2DM was significantly associated with the presence and severity of carotid atherosclerosis.\u003c/p\u003e\n\u003cp\u003e\u0026middot; Age and dyslipidemia contributed independently to vascular risk.\u003c/p\u003e\n\u003cp\u003e\u0026middot; A computational model integrating clinical and Doppler-derived data demonstrated excellent performance for risk stratification.\u003c/p\u003e\n\u003cp\u003e\u0026middot; IMT emerged as a key indicator of stenosis severity.\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eThe present study reinforces the strong association between type 2 diabetes mellitus (T2DM) and carotid atherosclerosis, confirming that individuals with T2DM exhibit a higher prevalence of carotid plaques, greater plaque burden, and more severe degrees of stenosis when compared to non-diabetic individuals. These findings align with previous evidence demonstrating the pro-atherogenic effects of chronic hyperglycemia, endothelial dysfunction, and systemic inflammation observed in diabetic populations (Lu et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Wu et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Notably, the identification of T2DM as an independent predictor of carotid atherosclerosis even after multivariate adjustment underlines its critical role in early vascular damage pathophysiology.\u003c/p\u003e\u003cp\u003eA key contribution of this study lies in the integration of a computational predictive model for vascular risk stratification, combining clinical risk factors and Doppler ultrasound-derived data within a multivariate logistic regression framework. The model identified T2DM, dyslipidemia, and age as independent predictors of carotid atherosclerotic alterations, in agreement with established pathophysiological mechanisms underlying vascular disease progression in diabetic individuals (Telles et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Ifrim \u0026amp; Vasilescu, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). These findings reinforce the importance of early vascular evaluation in T2DM patients, even in the absence of overt cardiovascular symptoms.\u003c/p\u003e\u003cp\u003eThe predictive performance of the developed model was further validated through ROC curve analysis, with an area under the curve (AUC) of 0.836, confirming its excellent discriminatory power. This result emphasizes the potential applicability of such computational tools in clinical practice, particularly when anchored in structured, objective imaging data like Doppler ultrasound findings (Dossabhoy et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Lareyre et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). By enabling the stratification of vascular risk in a reproducible and interpretable manner, the model can support individualized patient assessment and targeted preventive strategies.\u003c/p\u003e\u003cp\u003eFrom a computational standpoint, the use of logistic regression offers several practical advantages in clinical settings: it generates transparent, statistically robust, and easily interpretable models, thus facilitating clinical adoption without reliance on complex machine learning infrastructures (Cheng et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Wu et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Furthermore, its probabilistic output allows the generation of personalized risk estimates, enhancing clinical decision-making and resource allocation (Taslimitehrani et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Lv et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In this context, the incorporation of Doppler-derived morphological and hemodynamic parameters into computational models represents a significant advance beyond traditional risk factor-based assessments.\u003c/p\u003e\u003cp\u003eThe analysis using ordinal logistic regression provided additional insights, revealing that variables such as age, intima-media thickness (IMT), and hypertension were strongly associated with stenosis severity. Interestingly, T2DM did not independently predict more severe stenosis after multivariate adjustment, suggesting that its role may be more relevant in the initiation and early progression of carotid atherosclerosis. This highlights the complex interplay of factors in the progression of vascular disease and underlines the value of integrating imaging data, such as IMT measurements, into risk models for a more precise stratification.\u003c/p\u003e\u003cp\u003eThe early detection of subclinical atherosclerosis in high-risk populations, such as patients with T2DM, is fundamental for timely therapeutic intervention and the reduction of cardiovascular events (Robinson et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Fuster et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The computational model developed in this study could serve as a complementary clinical decision-support tool, assisting healthcare professionals in identifying high-risk patients and initiating personalized preventive measures.\u003c/p\u003e\u003cp\u003eNevertheless, the study presents some limitations. Its retrospective, single-center design may limit the generalizability of findings. While the computational model demonstrates good performance, it remains based on a conventional statistical approach. Future research should explore the integration of advanced machine learning algorithms, capable of managing larger datasets and capturing nonlinear interactions, which could enhance predictive accuracy and external applicability. Moreover, external validation in independent cohorts will be essential before clinical implementation of the proposed model.\u003c/p\u003e\u003cp\u003eIn conclusion, this study confirms the significant impact of T2DM on carotid atherosclerosis progression and introduces a clinically applicable, computationally validated risk stratification model integrating Doppler-derived data. These findings support the incorporation of vascular imaging and computational tools into cardiovascular risk assessment pathways for diabetic patients.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eThis study confirms the significant association between type 2 diabetes mellitus (T2DM) and the development of carotid atherosclerosis, with diabetic individuals presenting a higher prevalence of plaques, increased plaque burden, and more severe degrees of carotid stenosis compared to non-diabetic controls. The integration of Doppler ultrasound-derived data and clinical variables into a computational predictive model allowed accurate risk stratification, identifying T2DM, dyslipidemia, and age as independent predictors of carotid vascular alterations.\u003c/p\u003e\u003cp\u003eThe logistic regression-based model, validated through ROC curve analysis (AUC\u0026thinsp;=\u0026thinsp;0.836), demonstrated strong discriminatory power, highlighting its potential clinical applicability as a decision-support tool for early identification of high-risk individuals. Additionally, the incorporation of morphological markers, such as intima-media thickness (IMT), proved essential in assessing the severity of carotid stenosis.\u003c/p\u003e\u003cp\u003eThese findings underscore the value of integrating vascular imaging and computational risk models in cardiovascular risk assessment strategies for diabetic populations, supporting personalized and preventive approaches in clinical practice. Future studies should focus on external validation and the potential refinement of predictive models using advanced machine learning techniques to optimize risk prediction and improve clinical outcomes.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e: Conceptualization, S.M. and P.C.; methodology, S.M; software F.R.; validation, F.R., P.C. and S.M.; formal analysis, F.R; investigation, F.R.; data curation, P.C. and S.M.; writing—preparation of the original draft, S.M. and P.C.; writing—review and editing F.R.; visualization, P.C. and F.R; supervision, S.M.; project coordination, S.M. All authors have read and agreed to the published version of the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e: This research received no external funding.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInstitutional Review Board Statement\u003c/strong\u003e: This work was approved by the Ethics Committee from the Local Health Unit of Central Alentejo, and all ethical precepts were scrupulously respected by the researchers. Informed consent was waived, given the retrospective nature and the fact that no user-identifying data were used.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent Statement\u003c/strong\u003e: Not applicable, as this was a data review.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e: Data may be shared, in case of a duly justified request, considering the necessary protection.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e: The authors declare no conflicts of interest\u003c/p\u003e\n\u003cp dir=\"LTR\"\u003e\u003cstrong\u003eDisclaimer/Publisher’s Note:\u003c/strong\u003e The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eBranco, M. C., Coelho, P., \u0026amp; Rodrigues, F. (2024). Urinary tract infections in a single hospital in central Portugal, a 5-year analysis. \u003cem\u003eMicrobiology Research\u003c/em\u003e, 15(2), 850\u0026ndash;863. https://doi.org/10.3390/microbiolres15020055\u003c/li\u003e\n \u003cli\u003eCheng, X., Han, W., Liang, Y., Lin, X., Luo, J., Zhong, W., \u0026amp; Chen, D. (2022). Risk Prediction of Coronary Artery Stenosis in Patients with Coronary Heart Disease Based on Logistic Regression and Artificial Neural Network. \u003cem\u003eComputational and mathematical methods in medicine\u003c/em\u003e, \u003cem\u003e2022\u003c/em\u003e, 3684700. https://doi.org/10.1155/2022/3684700\u003c/li\u003e\n \u003cli\u003eDossabhoy, S. S., Ho, V. T., Ross, E. G., Rodriguez, F., \u0026amp; Arya, S. (2023). Artificial intelligence in clinical workflow processes in vascular surgery and beyond. \u003cem\u003eSeminars in vascular surgery\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e(3), 401\u0026ndash;412. https://doi.org/10.1053/j.semvascsurg.2023.07.002\u003c/li\u003e\n \u003cli\u003eDuarte-Mendes, P., Paulo, R., Coelho, P., Rodrigues, F., Marques, V., \u0026amp; Mateus, S. (2020). Variability of Lower Limb Artery Systolic-Diastolic Velocities in Futsal Athletes and Non-Athletes: Evaluation by Arterial Doppler Ultrasound. International journal of environmental research and public health, 17(2), 570. https://doi.org/10.3390/ijerph17020570\u003c/li\u003e\n \u003cli\u003eFedera\u0026ccedil;\u0026atilde;o Internacional de Diabetes (IDF). (2021). \u003cem\u003eIDF Diabetes Atlas\u003c/em\u003e (10th ed.). https://diabetesatlas.org/\u003c/li\u003e\n \u003cli\u003eFreitas, D. P. (2002). \u003cem\u003eDiabetes Mellitus tipo 2\u003c/em\u003e [Internet]. ALERT. http://www.alert-online.com/pt/medical-guide/diabetes-mellitus-tipo-2\u003c/li\u003e\n \u003cli\u003eEsteva, A., Robicquet, A., Ramsundar, B., et al. (2019). A guide to deep learning in healthcare. \u003cem\u003eNature Medicine, 25\u003c/em\u003e(1), 24-29. https://doi.org/10.1038/s41591-018-0316-z\u003c/li\u003e\n \u003cli\u003eFawcett, T. (2006). An introduction to ROC analysis. \u003cem\u003ePattern Recognition Letters, 27\u003c/em\u003e(8), 861-874. https://doi.org/10.1016/j.patrec.2005.10.010\u003c/li\u003e\n \u003cli\u003eFreitas, P. (2008). Aterosclerose carot\u0026iacute;dea avaliada pelo eco-Doppler: associa\u0026ccedil;\u0026atilde;o com fatores de risco e doen\u0026ccedil;as arteriais sist\u0026ecirc;micas. \u003cem\u003eJornal Vascular Brasileiro, 7\u003c/em\u003e(4), 298\u0026ndash;307.\u003c/li\u003e\n \u003cli\u003eFuster, V., Lois, F., \u0026amp; Franco, M. (2010). Early identification of atherosclerotic disease by noninvasive imaging. \u003cem\u003eNature reviews. Cardiology\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(6), 327\u0026ndash;333. https://doi.org/10.1038/nrcardio.2010.54\u003c/li\u003e\n \u003cli\u003eIfrim, S., \u0026amp; Vasilescu, R. (2004). Early detection of atherosclerosis in type 2 diabetic patients by endothelial dysfunction and intima-media thickness. \u003cem\u003eRomanian journal of internal medicine = Revue roumaine de medecine interne\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(2), 343\u0026ndash;354.\u003c/li\u003e\n \u003cli\u003eJordan, M. I., \u0026amp; Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. \u003cem\u003eScience, 349\u003c/em\u003e(6245), 255-260. https://doi.org/10.1126/science.aaa8415\u003c/li\u003e\n \u003cli\u003eLareyre, F., Behrendt, C. A., Chaudhuri, A., Lee, R., Carrier, M., Adam, C., L\u0026ecirc;, C. D., \u0026amp; Raffort, J. (2023). Applications of artificial intelligence for patients with peripheral artery disease. \u003cem\u003eJournal of vascular surgery\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(2), 650\u0026ndash;658.e1. https://doi.org/10.1016/j.jvs.2022.07.160\u003c/li\u003e\n \u003cli\u003eLu, S. X., Wu, T. W., Chou, C. L., Cheng, C. F., \u0026amp; Wang, L. Y. (2023). Combined effects of hypertension, hyperlipidemia, and diabetes mellitus on the presence and severity of carotid atherosclerosis in community-dwelling elders: A community-based study. \u003cem\u003eJournal of the Chinese Medical Association : JCMA\u003c/em\u003e, \u003cem\u003e86\u003c/em\u003e(2), 220\u0026ndash;226. https://doi.org/10.1097/JCMA.0000000000000839\u003c/li\u003e\n \u003cli\u003eLv, H., Yang, X., Wang, B., Wang, S., Du, X., Tan, Q., Hao, Z., Liu, Y., Yan, J., \u0026amp; Xia, Y. (2021). Machine Learning-Driven Models to Predict Prognostic Outcomes in Patients Hospitalized With Heart Failure Using Electronic Health Records: Retrospective Study. \u003cem\u003eJournal of medical Internet research\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(4), e24996. https://doi.org/10.2196/24996\u003c/li\u003e\n \u003cli\u003eOrganiza\u0026ccedil;\u0026atilde;o Mundial da Sa\u0026uacute;de (OMS). (2022). \u003cem\u003eDiabetes\u003c/em\u003e. https://www.who.int/news-room/fact-sheets/detail/diabetes\u003c/li\u003e\n \u003cli\u003eTaslimitehrani, V., Dong, G., Pereira, N. L., Panahiazar, M., \u0026amp; Pathak, J. (2016). Developing EHR-driven heart failure risk prediction models using CPXR(Log) with the probabilistic loss function. \u003cem\u003eJournal of biomedical informatics\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e, 260\u0026ndash;269. https://doi.org/10.1016/j.jbi.2016.01.009\u003c/li\u003e\n \u003cli\u003eRyden, L., Standl, E., Bartnik, M., Van den Berghe, G., Betteridge, J., de Boer, M.-J., et al. (2007). Guidelines on diabetes, pre-diabetes, and cardiovascular diseases: full text. \u003cem\u003eEuropean Heart Journal Supplements, 9\u003c/em\u003e(Suppl C), C3\u0026ndash;C74. https://doi.org/10.1093/eurheartj/ehl261\u003c/li\u003e\n \u003cli\u003eRobinson, J. G., Fox, K. M., Bullano, M. F., Grandy, S., \u0026amp; SHIELD Study Group (2009). Atherosclerosis profile and incidence of cardiovascular events: a population-based survey. \u003cem\u003eBMC cardiovascular disorders\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e, 46. https://doi.org/10.1186/1471-2261-9-46\u003c/li\u003e\n \u003cli\u003eRodrigues, J. B., Barroso, F. P. D., \u0026amp; A. (2011). Etiology and bacterial susceptibility to urinary tract infections | Etiologia e sensibilidade bacteriana em infec\u0026ccedil;\u0026otilde;es do trato urin\u0026aacute;rio. \u003cem\u003eRevista Portuguesa de Sa\u0026uacute;de P\u0026uacute;blica\u003c/em\u003e, 29(2), 123\u0026ndash;131.\u003c/li\u003e\n \u003cli\u003eSociedade Portuguesa de Aterosclerose. (2023). \u003cem\u003eHighlights do Congresso Europeu de Aterosclerose 2023\u003c/em\u003e. https://spaterosclerose.org/noticias/highlights-do-congresso-europeu-de-aterosclerose-2023/\u003c/li\u003e\n \u003cli\u003eSun, H., Saeedi, P., Karuranga, S., Pinkepank, M., Ogurtsova, K., Duncan, B. B., ... \u0026amp; Williams, R. (2022). IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. \u003cem\u003eDiabetes Research and Clinical Practice, 183\u003c/em\u003e, 109119. https://doi.org/10.1016/j.diabres.2021.109119\u003c/li\u003e\n \u003cli\u003eTelles, R. W., Lanna, C. C., Ferreira, G. A., Souza, A. J., Navarro, T. P., \u0026amp; Ribeiro, A. L. (2008). Carotid atherosclerotic alterations in systemic lupus erythematosus patients treated at a Brazilian university setting. \u003cem\u003eLupus\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(2), 105\u0026ndash;113. https://doi.org/10.1177/0961203307085312\u003c/li\u003e\n \u003cli\u003eWu, T. W., Chou, C. L., Cheng, C. F., Lu, S. X., \u0026amp; Wang, L. Y. (2022). Prevalences of diabetes mellitus and carotid atherosclerosis and their relationships in middle-aged adults and elders: a community-based study. \u003cem\u003eJournal of the Formosan Medical Association = Taiwan yi zhi\u003c/em\u003e, \u003cem\u003e121\u003c/em\u003e(6), 1133\u0026ndash;1140. https://doi.org/10.1016/j.jfma.2021.10.005\u003c/li\u003e\n \u003cli\u003eWu, Y., Xin, B., Wan, Q., Ren, Y., \u0026amp; Jiang, W. (2024). Risk factors and prediction models for cardiovascular complications of hypertension in older adults with machine learning: A cross-sectional study. \u003cem\u003eHeliyon\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(6), e27941. https://doi.org/10.1016/j.heliyon.2024.e27941\u003c/li\u003e\n \u003cli\u003eZhou, Z., et al. (2023). Arterial stiffness and obesity as predictors of diabetes. \u003cem\u003eJMIR Public Health and Surveillance, 10\u003c/em\u003e, e46088. https://doi.org/10.2196/46088\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Type 2 diabetes mellitus, carotid atherosclerosis, Doppler ultrasound, computational model, vascular risk stratification, predictive analytics, cardiovascular prevention, intima-media thickness","lastPublishedDoi":"10.21203/rs.3.rs-7160100/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7160100/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eType 2 diabetes mellitus (T2DM) is a major risk factor for the development of atherosclerotic disease. Carotid Doppler ultrasound allows non-invasive detection of subclinical vascular alterations, offering a potential tool for early risk stratification. This study aimed to evaluate the association between T2DM and carotid atherosclerosis and to develop a computational predictive model integrating Doppler-derived data and clinical risk factors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e\u003cbr\u003e\nA cross-sectional study involving 611 patients undergoing cervical Doppler ultrasound was conducted. Demographic data, cardiovascular risk factors, intima-media thickness (IMT), and carotid plaque presence were analyzed. Associations between T2DM and vascular alterations were assessed using chi-square and Mann-Whitney U tests. A multivariate logistic regression model was constructed to predict the presence of atherosclerotic changes, followed by ROC curve analysis to evaluate predictive performance. An ordinal regression analysis assessed predictors of stenosis severity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e\u003cbr\u003e\nT2DM was significantly associated with carotid plaque presence (82.4% vs. 67.6%; p=0.015), greater plaque burden (2.12 vs. 1.55 plaques/patient; p=0.001), and higher prevalence of stenosis \u0026gt;50% (13.7% vs. 4.9%; p=0.037). The multivariate predictive model identified T2DM (OR=2.56; 95% CI: 1.48–4.42; p\u0026lt;0.01), dyslipidemia (OR=1.87; p=0.02), and age (OR=1.04 per year; p\u0026lt;0.001) as independent predictors of carotid atherosclerotic alterations. The ROC curve analysis yielded an AUC of 0.836, demonstrating excellent discriminative ability. IMT, age, and hypertension were independently associated with increasing degrees of stenosis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e\u003cbr\u003e\nT2DM significantly increases the risk of carotid atherosclerosis. The developed computational model, integrating clinical and Doppler-derived data, demonstrated robust predictive performance for vascular risk stratification in diabetic patients. These findings support the incorporation of vascular imaging and computational tools into preventive cardiovascular strategies.\u003c/p\u003e","manuscriptTitle":"Computational Risk Modeling of Carotid Atherosclerosis in Type 2 Diabetes: Insights from Doppler-Based Datat","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-28 07:57:48","doi":"10.21203/rs.3.rs-7160100/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d64d4514-165d-468c-93fd-c6d2847590ef","owner":[],"postedDate":"July 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-30T10:09:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-28 07:57:48","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7160100","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7160100","identity":"rs-7160100","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}