HR-MRI-based nomogram network calculator to predict stroke recurrence in high-risk non-disabling ischemic cerebrovascular events patients.

Background and objective: To investigate the use of high-resolution magnetic resonance imaging (HR-MRI) to identify the characteristics of culprit plaques in intracranial arteries, and to evaluate the predictive value of the characteristics of culprit plaques combined with the modified Essen score for the recurrence risk of high-risk non-disabling ischemic cerebrovascular events (HR-NICE) patients. Methods: A retrospective analysis was conducted on 180 patients with HR-NICE at the First Affiliated Hospital of Xinxiang Medical University, including 128 patients with no recurrence (non-recurrence group) and 52 patients with recurrence (recurrence group). A total of 65 patients with HR-NICE were collected from the Sixth Affiliated Hospital of Shanghai Jiaotong University as a validation group, and their modified Essen scores, high-resolution magnetic resonance vessel wall images, and clinical data were collected. The culprit plaques were analyzed using VesselExplorer2 software. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for recurrence, and a nomogram was constructed using R software to evaluate the discrimination of the model. The area under the curve (AUC) of the receiver operating characteristic curve (ROC) was used to evaluate the model performance. Calibration curves and Decision Curve Analysis (DCA) were used to evaluate the model efficacy. Results: Intra-plaque hemorrhage (OR = 3.592, 95% CI = 1.474-9.104, p = 0.006), homocysteine (OR = 1.098, 95% CI = 1.025-1.179, p = 0.007), and normalized wall index (OR = 1.114, 95% CI = 1.027-1.222, p = 0.015) were significantly higher in the recurrent stroke group than in the non-recurrent stroke group, and were independent risk factors for recurrent stroke. The performance of the nomogram model (AUC = 0.830, 95% CI: 0.769-0.891; PR-AUC = 0.628) was better than that of the modified Essen scoring model (AUC = 0.660, 95% CI: 0.583-0.738) and the independent risk factor combination model (AUC = 0.827, 95% CI: 0.765-0.889). The nomogram model still had good model performance in the validation group (AUC = 0.785, 95% CI: 0.671-0.899), with a well-fitting calibration curve and a DCA curve indicating good net benefit efficacy for patients. Conclusion: High-resolution vessel wall imaging combined with a modified Essen score can effectively assess the recurrence risk of HR-NICE patients, and the nomogram model can provide a reference for identifying high-risk populations with good clinical application prospects.

An interpretable machine learning model for stroke recurrence in patients with symptomatic intracranial atherosclerotic arterial stenosis.

Background and objective: Symptomatic intracranial atherosclerotic stenosis (SICAS) is the most common etiology of ischemic stroke and one of the main causes of high stroke recurrence. The recurrence of stroke is closely related to the prognosis of ischemic stroke. This study aims to develop a machine learning model based on high-resolution vessel wall imaging (HR-VWI) to predict the risk of stroke recurrence in SICAS. Methods: This study retrospectively collected data from 180 SICAS stroke patients treated at the hospital between 2020.01 and 2022.01. Relevant imaging and clinical data were collected, and follow-up was conducted. The dataset was divided into a training set and a validation set in a ratio of 7:3. We employed the least absolute shrinkage and selection operator (LASSO) regression to perform a selection on the baseline data, laboratory tests, and neuroimaging data generated by HR-VWI scans collected from the training set. Finally, five machine learning techniques, including logistic regression model (LR), support vector machine (SVM), Gaussian naive Bayes (GNB), Complement naive Bayes (CNB), and k-nearest neighbors algorithm (kNN), were employed to develop a predictive model for stroke recurrence. Shapley Additive Explanation (SHAP) was used to provide visualization and interpretation for each patient. The model's effectiveness was evaluated using average accuracy, sensitivity, specificity, precision, f1 score, PR curve, calibration curve, and decision curve analysis. Results: LASSO analysis revealed that "history of hypertension," "homocysteine level," "NWI value," "stenosis rate," "intracranial hemorrhage," "positive remodeling," and "enhancement grade" were independent risk factors for stroke recurrence in SICAS patients. In 10-fold cross-validation, the area under the curve (AUC) ranged from 0.813 to 0.912 in ROC curve analysis. The area under the precision-recall curve (AUPRC) ranged from 0.655 to 0.833, with the Gaussian Naive Bayes (GNB) model exhibiting the best ability to predict stroke recurrence in SICAS. SHAP analysis provided interpretability for the machine learning model and revealed essential factors related to the risk of stroke recurrence in SICAS. Conclusion: A precise machine learning-based prediction model for stroke recurrence in SICAS has been established to assist clinical practitioners in making clinical decisions and implementing personalized treatment measures.

[Finite-element analysis of mandibular first molar with two marginal designs of endocrown for the repair of different defects].

OBJECTIVE: This study aimed to evaluate the stress distribution of the mandibular first molar with different thicknesses and heights of the axial wall restored by the endocrown with two marginal designs and thus provide a theoretical basis for selecting clinical preparation through the finite-element method. METHODS: Two marginal endocrowns of the mandibular first molar with different axial-wall thicknesses (t=1, 2, 3 mm) and heights (h=2, 3, 4 mm) were established. Group A was the butt-joint design, whereas group B was the shoulder-surrounded design. After applying vertical and oblique loads , the size and distribution of the maximum principal stress and equivalent stress of residual tooth tissue were recorded. RESULTS: The maximum principal stress and equivalent stress distribution of residual tooth tissue were similar among different models. Group A showed a lower maximum principal stress and equivalent stress than group B at the same thickness and height under vertical load. Meanwhile, under oblique load, the maximum principal stress values of groups A and B decreased with increased thickness at constant height. Group A showed lower equivalent stress than group B at the same thickness and height of 2 and 3 mm. However, when the height was 4 mm, the trend was reversed. CONCLUSIONS: In mastication, when bearing the vertical force, the retention of the butt-joint marginal endocrown preferred to the shoulder-surrounded one. Given the higher axial wall of the shoulder-surrounded marginal endocrown, it showed better ability to bear the oblique force than the butt-joint one.