ObjectiveTo develop and validate CT-based radiomics models incorporating intrathrombus and perithrombus features for predicting complete recanalization [modified Thrombolysis in Cerebral Infarction (mTICI)2c/3] after endovascular thrombectomy (EVT) in acute ischemic stroke (AIS), and to identify the optimal machine learning classifier.Materials and methodsThis retrospective study included 406 AIS patients with anterior circulation large-vessel occlusion from three centers (December 2018–April 2024). Patients were allocated to training (n = 178), internal testing (n = 77), and external validation (n = 151) cohorts. Complete recanalization was defined as mTICI 2c/3. A total of 428 radiomics features were extracted from non-contrast CT and CT angiography (CTA). Least absolute shrinkage and selection operator (LASSO) regression and eleven classifiers were employed.ResultsThe combined intrathrombus-perithrombus model with logistic regression achieved area under the curve (AUC) values of 0.93 (training), 0.88 (testing), and 0.86 (validation), outperforming single-region models. Decision curve analysis confirmed superior clinical utility. The perithrombus region contributed dominantly (10 of 15 features) to the combined model.ConclusionThe combined CT-based radiomics model effectively predicts complete recanalization, providing an objective tool for patient selection and treatment optimization.

Abstract Objectives Complete reperfusion is the optimal technical goal of endovascular therapy (EVT) and is closely linked to favorable outcomes in acute ischemic stroke (AIS). This study developed and validated clot- and peri-clot–based radiomics models on pre-interventional dual-energy CT angiography (DE-CTA) to predict complete reperfusion and clinical outcome after EVT. Materials and methods A total of 371 patients from three centers were retrospectively enrolled and assigned to training (n = 154), test (n = 66), and validation (n = 151) cohorts. Radiomics features from clot and peri-clot regions were extracted, and three machine learning models—clot-based, peri-clot-based, and combined—were constructed. Model performance for predicting complete reperfusion and 90-day outcome was assessed using AUC. Results Small, optimized feature subsets were selected for each model (11/11, 17/10, and 13/10 features for clot-based, peri-clot-based, and combined models for reperfusion and outcome prediction, respectively). For complete reperfusion, the peri-clot model showed the best performance with AUCs of 0.885 (95% CI: 0.834–0.937), 0.860 (95% CI: 0.771–0.948), and 0.847 (95% CI: 0.778–0.916) in the training, test, and validation cohorts, outperforming the clot-based (0.809, 0.759, 0.719) and combined models (0.867, 0.840, 0.820). A similar advantage was observed for outcome prediction, where the peri-clot model achieved the highest AUCs (0.854, 0.817, 0.850), exceeding the combined (0.839, 0.763, 0.804) and clot-based models (0.826, 0.709, 0.734). Conclusions DE-CTA peri-clot radiomics provides superior prediction of both complete reperfusion and functional outcome after EVT, underscoring the clinical relevance of peri-clot microenvironment imaging and its potential to enhance pre-EVT patient selection and individualized prognostic evaluation. Trial registration The trial registration number (Chinese Clinical Trial Registry, ChiCTR2400092800) and date of registration (2024.11.22) were retrospectively registered. Critical relevance statement Peri-clot dual-energy computed tomography angiography radiomics outperformed clot-based models in multicenter external validation for predicting complete reperfusion and good 90-day functional outcome after endovascular therapy for acute ischemic stroke, supporting improved preprocedural risk stratification and patient selection. Key Points Complete reperfusion and functional recovery after endovascular thrombectomy remain hard to predict using thrombus features alone. Peri-clot radiomics on dual-energy computed tomography angiography achieved an external validation performance of 0.847 for complete reperfusion prediction. The same peri-clot model predicted 90-day functional outcome with 0.850 performance, supporting microenvironment-informed risk stratification. Peri-clot signatures consistently outperformed clot-only and combined models across multicenter cohorts, supporting robustness. Graphical Abstract

BACKGROUND AND PURPOSE:To develop a two-stage framework that combines deep learning-based super-resolution with subsequent image processing to generate high-contrast thin-slice CT (HCCT) from NCCT, thereby improving the detection of early ischemic changes (EICs) in acute anterior circulation stroke. MATERIALS AND METHODS:A retrospective study was conducted on patients with large vessel occlusion stroke between 2020 and 2024 across four medical centers. NCCT images were converted to HCCT using the two-stage framework. Two neuroradiologists (NRADs) independently assessed ASPECTS and ischemic volumes on both NCCT and HCCT-assisted interpretation, with diagnostic accuracy measured using Tmax > 6 seconds as a reference standard. RESULTS:The study included 303 participants (mean age, 67.2 years ± 12.7; 187 male). The mean Tmax-ASPECTS was 4.9 ± 2.7, and the median ischemic volume was 50.36 mL (IQR, 29.36-72.49 mL). The inter-rater ASPECTS correlation improved significantly from 0.72 on NCCT to 0.94 on HCCT-assisted (p < 0.001). The intra-class correlation coefficient (ICC) for HCCT-assisted ASPECTS was 0.85 (95% CI: 0.80 to 0.89; p < 0.001), and for Tmax-ASPECTS, it was 0.90 (95% CI: 0.87 to 0.93; p < 0.001), as evaluated by two NRADs. Two NRADs reported strong correlations observed of ischemic volumes between the HCCT-assisted group and Tmax (NRAD1: r = 0.83, NRAD2: r = 0.83; both p < 0.001). ASPECTS ≥6 in the HCCT-assisted group showed the strongest association with favorable outcome (NRAD1: OR 2.87, 95% CI 1.97-5.21; NRAD2: OR 2.74, 95% CI 1.78-5.55; both p <0.001), outperforming conventional NCCT. CONCLUSION:HCCT significantly improves the interpretation of EICs, improving inter-rater agreement for ASPECTS scoring in acute anterior circulation stroke.

Objective: This study aimed to investigate the utility of magnetic resonance imaging (MRI)-based habitat radiomics for preoperatively distinguishing early-stage endometrial carcinoma (EC) from submucous leiomyoma (SML). Materials and Methods: A retrospective study was conducted on uterine lesions patients who underwent MRI from three hospitals. The k-means clustering algorithm was applied to segment the MRI into distinct habitats based on T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and apparent diffusion coefficient (ADC) maps. Radiomic features were extracted from whole-tumor and these habitats and selected by the Pearson correlation coefficient and least absolute shrinkage and selection operator (LASSO) regression. A logistic regression (LR) model was constructed by these radiomics in the training set. Results: A total of 231 eligible patients were incorporated, 97 EC and 134 SML confirmed by histopathology. In the training cohort, the AUCs of the models based on features from the whole-tumor, habitat_1, habitat_2, and habitat_3 were 0.826, 0.787, 0.770, and 0.907, respectively, while in the test and external validation cohorts, the corresponding AUCs were 0.774/0.751, 0.486/0.608, 0.663/0.514, and 0.858/0.881. Compared with whole-tumor model, habitat_3 model demonstrated incrementally improved predictive performance in the external validation cohort (0.881 [95% CI: 0.799-0.934]). Conclusion: MRI-based habitat radiomics offers incremental improvement for preoperative differentiation between early-stage EC and SML.

BACKGROUND:Predicting the final infarct after an extended time window mechanical thrombectomy (MT) is beneficial for treatment planning in acute ischemic stroke (AIS). By introducing guidance from prior knowledge, this study aims to improve the accuracy of the deep learning model for post-MT infarct prediction using pre-MT brain perfusion data. METHODS:This retrospective study collected CT perfusion data at admission for AIS patients receiving MT over 6 hours after symptom onset, from January 2020 to December 2024, across three centers. Infarct on post-MT diffusion weighted imaging served as ground truth. Five Swin transformer based models were developed for post-MT infarct segmentation using pre-MT CT perfusion parameter maps: BaselineNet served as the basic model for comparative analysis, CollateralFlowNet included a collateral circulation evaluation score, InfarctProbabilityNet incorporated infarct probability mapping, ArterialTerritoryNet was guided by artery territory mapping, and UnifiedNet combined all prior knowledge sources. Model performance was evaluated using the Dice coefficient and intersection over union (IoU). RESULTS:A total of 221 patients with AIS were included (65.2% women) with a median age of 73 years. Baseline ischemic core based on CT perfusion threshold achieved a Dice coefficient of 0.50 and IoU of 0.33. BaselineNet improved to a Dice coefficient of 0.69 and IoU of 0.53. Compared with BaselineNet, models incorporating medical knowledge demonstrated higher performance: CollateralFlowNet (Dice coefficient 0.72, IoU 0.56), InfarctProbabilityNet (Dice coefficient 0.74, IoU 0.58), ArterialTerritoryNet (Dice coefficient 0.75, IoU 0.60), and UnifiedNet (Dice coefficient 0.82, IoU 0.71) (all P<0.05). CONCLUSIONS:In this study, integrating medical knowledge into deep learning models enhanced the accuracy of infarct predictions in AIS patients undergoing extended time window MT.

Acute ischemic stroke (AIS) presents significant heterogeneity in clinical and thrombus imaging characteristics, which can profoundly impact therapeutic decisions and outcomes. This study analyzed 520 AIS patients who underwent endovascular thrombectomy, integrating clinical variables and thrombus imaging features to identify potential subtypes through unsupervised clustering and principal component analysis. Three distinct subtypes emerged: Cluster 1, characterized by middle cerebral artery occlusion, shorter thrombus lengths, and favorable outcomes; Cluster 2, comprising predominantly male smokers and drinkers with no significant outcome differences; and Cluster 3, consisting of older patients with higher stroke severity, internal carotid artery occlusion, longer thrombus lengths, and poor outcomes. Key features driving subtype differentiation included atrial fibrillation, thrombus perviousness, and clot burden scores. Significant variations in recanalization and hemorrhagic transformation rates were also observed among clusters. These findings underscore the potential of integrating thrombus imaging characteristics into personalized treatment strategies, offering a more precise approach to prognosis and management for AIS patients.

OBJECTIVE:This study evaluated the utility of dual-energy computed tomography angiography (DECT) in predicting malignant middle cerebral artery infarction (MMI) in acute ischemic stroke (AIS). METHODS:A total of 153 AIS patients undergoing DECT within 12 hours of symptom onset and follow-up imaging within 48 hours were included. DECT-derived parameters-virtual non-contrast (VNC) images, virtual monoenergetic (VM) images (40/60 keV), iodine concentration (IC), effective atomic number (Zeff), and spectral Hounsfield unit curve slope (λHU)-were analyzed. Clinical and imaging parameters were compared between MMI (n = 34, 22.2%) and non-MMI groups. RESULTS:MMI patients exhibited higher National Institute of Health Stroke Scale (NIHSS) scores, more frequent internal carotid artery (ICA) occlusion, larger baseline infarct volumes, and significantly lower IC, λHU, Zeff, VNC, and VM values (40/60 keV) compared with non-MMI patients (all P < 0.05). Combined DECT parameters demonstrated superior diagnostic performance for MMI prediction (area under the curve [AUC] 0.98; sensitivity 88%, specificity 98%), outperforming individual parameters and clinical predictors (all P < 0.05). Integrating clinical features (NIHSS, admission infarct volume, ICA occlusion) with DECT parameters achieved optimal performance (AUC 1.00; sensitivity 100%, specificity 93%), comparable to final infarct volume (P = 0.08). CONCLUSIONS:DECT-derived quantitative parameters, particularly when combined with clinical data, serve as reliable early biomarkers for identifying stroke patients at high risk of MMI, offering predictive accuracy akin to final infarct outcomes. This approach may guide timely intervention in malignant cerebral edema.

To explore a novel approach for the early prediction of malignant cerebral edema (MCE) in stroke patients using radiomics features extracted from dual-energy computed tomography angiography (DE-CTA) reconstructed images of infarcted brain tissue. This retrospective study enrolled 398 stroke patients who underwent DE-CTA between April 2016 and November 2022 from three medical centers. Patients were allocated into a training cohort (n = 227) and a test cohort (n = 171) based on their source institution. Radiomics features were extracted from DE-CTA reconstructions of the infarct tissue. Radiomics models were built for each reconstruction scan images and for a combined model using features from all images. Additionally, a nomogram model integrating significant radiomics features and clinical characteristics was developed. The diagnostic performance of all models was evaluated using receiver operating characteristic (ROC) curve analysis, calculating the area under the curve (AUC). Radiomics models based on each individual DE-CTA reconstruction demonstrated robust predictive performance. In the test cohort, the combined radiomics model integrating features from all DE-CTA reconstructions significantly superior predictive performance compared to most single-image models. In the test cohort, both the combined radiomics model (AUC = 0.950) and the nomogram model (AUC = 0.935) significantly outperformed the clinical model (AUC = 0.574), both p < 0.001. No significant difference in performance was observed between the nomogram and the radiomics model (p = 0.30). Radiomics models derived from DE-CTA reconstructed images of infarcted tissue provide a reliable method for the early prediction of MCE development in stroke patients.

The viability of using thrombus heterogeneity (TH) data derived from dual-energy CT (DECT) as a visual thrombotic biomarker is unclear. The first aim of this study is to develop a quantitative measure of TH on DECT and test its performance for predicting the stroke source (cardiogenic vs. non-cardiogenic) and clinical outcomes (functional status assessed by the modified Rankin Scale score at 90 days) following machine thrombectomy (MT). The second aim is to associate thrombus subregions with the thrombus composition to facilitate visualization of thrombus constituents. Radiomics data are extracted from the whole thrombus and subregions in CT/DECT to construct predictive models. The performances of all models are evaluated and compared in the validation and comparative cohorts. Histopathologic analysis is performed to correlate the subregion data with the actual thrombus composition. This study included 221 and 255 participants who underwent DECT and CT examinations, respectively. DECT outperformed CT in predicting stroke source and clinical outcomes, with the TH-related models showing the highest performance in the validation and comparative cohorts. Thrombus composition is correlated with the different CT/DECT-based subregions, with DECT-habitat_c showing the strongest association. Thrombus subregion analyses may help visualize the related constituents.

OBJECTIVE:The purpose of this research was to evaluate the effectiveness of contrast-enhanced computed tomography (CECT)-based habitat radiomics in differentiating low-risk thymomas from high-risk thymomas prior to surgery. MATERIALS AND METHODS:A retrospective study was conducted involving patients with thymomas who had undergone CECT at three medical centers. The patients were divided into two cohorts: a training cohort comprising 134 patients from Centers A and B, and a validation cohort consisting of 41 patients from Center C. The k-means clustering algorithm was employed to segment the CECT images into distinct tumor habitats. Radiomic features were extracted from the entire tumor and the specific habitats identified. After feature selection, logistic regression (LR) model was developed to distinguish between low-risk and high-risk thymomas. RESULTS:A total of 175 patients were enrolled in the study, with 106 diagnosed with low-risk thymomas and 69 with high-risk thymomas. In the validation cohort, the area under the receiver operating characteristic curve (AUC) values for the models derived from the whole tumor, habitat_1, habitat_2, and habitat_3 were 0.806 (95% CI 0.675-0.938), 0.946 (95% CI 0.861-1.000), 0.620 (95% CI 0.446-0.794), and 0.946 (95% CI 0.885-1.000), respectively. The habitats model demonstrated superior predictive performance compared to the whole tumor model. CONCLUSION:CECT-based habitat radiomics represents a promising diagnostic approach for distinguishing between low-risk and high-risk thymomas in the preoperative setting, highlighting its potential for enhanced diagnostic accuracy.

ObjectivesTo accurately assess the predictive ability of radiomics and deep learning (DL) features in intrathrombus and perithrombus regions for the risk of malignant cerebral edema (MCE) after acute ischemic stroke (AIS).Materials and methodsA retrospective study was conducted, enrolling 406 AIS patients who underwent admission CT before endovascular thrombectomy (EVT). Center A patients were randomly divided (7:3) into training/testing sets; Centers B and C formed the external validation cohort. Regions of interest (ROIs) of thrombus and perithrombus were manually delineated and automatically expanded in margin by one pixel. Four hundred twenty-eight radiomic features were extracted from CT images of intrathrombus and perithrombus regions, and 128 DL features were obtained by inputting these images into a VGG16 architecture. Following features fusion, least absolute shrinkage and selection operator (LASSO) regression was employed for dimensionality reduction. Eleven machine learning classifiers were used for model development. Models’ performance was evaluated using Matthews correlation coefficient (MCC) and area under the receiver operating characteristic curve (AUC), with AUC differences tested using DeLong’s method.ResultsMCE occurred in 49 patients (12.1%). In the validation cohort, the logistic regression (LR) models demonstrated discriminative performance with perithrombus (LR-peri: MCC = 0.857, AUC = 0.891), intrathrombus, (LR-intra: MCC = 0.328, AUC = 0.626), and combined (LR-combined: MCC = 0.41, AUC = 0.869) models. The LR-combined model exhibited a significantly superior predictive capacity to that of LR-intra (p < 0.05).ConclusionPerithrombus features enhance MCE prediction after AIS, enabling optimized medical resource allocation.Clinical relevance statementEmphasis is placed on the critical significance of radiomics extracted from the area in and around the thrombus in predicting MCE after AIS, which has far-reaching significance for improving patient prognosis.

BACKGROUND AND PURPOSE:Complications from endovascular thrombectomy (EVT) can negatively affect clinical outcomes, making the development of a more precise and objective prediction model essential. This research aimed to assess the effectiveness of radiomics features derived from presurgical CT scans in predicting the prognosis post-EVT in patients with acute ischemic stroke. MATERIALS AND METHODS:This investigation included 336 patients with acute ischemic stroke from 2 medical centers from March 2018 to March 2024. The participants were split into a training cohort of 161 patients and a validation cohort of 175 patients. Patient outcomes were rated with the mRS: 0-2 for good, 3-6 for poor. A total of 428 radiomics features were derived from intrathrombus and perithrombus regions in noncontrast CT and CTA images. Feature selection was conducted using a least absolute shrinkage and selection operator regression model. The efficacy of 8 different supervised learning models was assessed using the area under the curve (AUC) of the receiver operating characteristic curve. RESULTS:Among all models tested in the validation cohort, the logistic regression algorithm for the combined model achieved the highest AUC (0.87; 95% CI, 0.81-0.92), outperforming other algorithms. The combined use of radiomics features from both the intrathrombus and perithrombus regions significantly enhanced diagnostic accuracy over models using features from a single region (0.81 versus 0.70, 0.77), highlighting the benefit of integrating data from both regions for improved prediction. CONCLUSIONS:The findings suggest that a combined radiomics model based on CT serves as a potent approach to assessing the prognosis following EVT. The logistic regression model, in particular, proved to be both effective and stable, offering critical insights for the management of stroke.

Objectives: This study aimed to assess the predictive performance of radiomics derived from computed tomography (CT) images of thrombus regions in predicting the risk of intracranial hemorrhage (ICH) following endovascular thrombectomy (EVT). Materials and Methods: This retrospective multicenter study included 336 patients who underwent admission CT and EVT for acute anterior-circulation large vessel occlusion between December 2018 and December 2023. Follow-up imaging was performed 24 h post-procedure to evaluate the occurrence of ICH. 230 patients from centers A and B were randomly allocated into training and test groups in a 7:3 ratio, while the remaining 106 patients from center C comprised the validation cohort. Radiologists manually segmenting the thrombus on CT images, and the perithrombus region was defined by expanding the initial region of interest (ROI). A total of 428 radiomics features were extracted from both intrathrombus and perithrombus regions on CT images. The Mann-Whitney U test was used for feature selection, and least absolute shrinkage and selection operator (LASSO) regression was employed for model development, followed by validation using a 5-fold cross-validation approach. Model performance was assessed using the area under the curve (AUC) of the receiver operating characteristic (ROC). Results: Among the eligible patients, 128 (38.1 %) experienced ICH after EVT. The combined model exhibited superior performance in the training cohort (AUC: 0.913, 95 % CI: 0.861-0.965), test cohort (AUC: 0.868, 95 % CI: 0.775-0.962), and validation cohort (AUC: 0.850, 95 % CI: 0.768-0.912). Notably, in the validation group, both the perithrombus and combined models demonstrated higher predictive accuracy compared to the intrathrombus model (0.837 vs. 0.684, p = 0.02; AUC: 0.850 vs. 0.684, p = 0.01). Conclusions: Radiomics features derived from the perithrombus region significantly enhance the prediction of ICH after EVT, providing valuable insights for optimizing post-procedural clinical decisions. Clinical relevance statement: This study highlights the importance of radiomics extracted from intrathrombus and perithrombus region in predicting intracranial hemorrhage following endovascular thrombectomy, which can aid in improving patient outcomes.

We aimed to develop and validate a radiomics nomogram based on dual-energy computed tomography (DECT) images and clinical features to classify the time since stroke (TSS), which could facilitate stroke decision-making. This retrospective three-center study consecutively included 488 stroke patients who underwent DECT between August 2016 and August 2022. The eligible patients were divided into training, test, and validation cohorts according to the center. The patients were classified into two groups based on an estimated TSS threshold of ≤ 4.5 h. Virtual images optimized the visibility of early ischemic lesions with more CT attenuation. A total of 535 radiomics features were extracted from polyenergetic, iodine concentration, virtual monoenergetic, and non-contrast images reconstructed using DECT. Demographic factors were assessed to build a clinical model. A radiomics nomogram was a tool that the Rad score and clinical factors to classify the TSS using multivariate logistic regression analysis. Predictive performance was evaluated using receiver operating characteristic (ROC) analysis, and decision curve analysis (DCA) was used to compare the clinical utility and benefits of different models. Twelve features were used to build the radiomics model. The nomogram incorporating both clinical and radiomics features showed favorable predictive value for TSS. In the validation cohort, the nomogram showed a higher AUC than the radiomics-only and clinical-only models (AUC: 0.936 vs 0.905 vs 0.824). DCA demonstrated the clinical utility of the radiomics nomogram model. The DECT-based radiomics nomogram provides a promising approach to predicting the TSS of patients. The findings support the potential clinical use of DECT-based radiomics nomograms for predicting the TSS.

Rationale and Objectives: The purpose of this study was to determine the association between hemispheric synchrony in venous outflow at baseline and tissue fate after mechanical thrombectomy (MT) for acute ischemic stroke (AIS). Materials and Methods: A two-center retrospective analysis involving AIS patients who underwent MT was performed. The four cortical veins of interest include the superficial middle cerebral vein (SMCV), sphenoparietal sinus (SS), vein of Labb & eacute; (VOL), and vein of Trolard (VOT). Baseline computed tomography perfusion data were used to compare the following outflow parameters between the hemispheres: first filling time ( triangle FFT), time to peak ( triangle TTP) and total filling time ( triangle TFT). Synchronous venous outflow was defined as triangle FFT = 0. Multivariable regression analyses were performed to evaluate the association of venous outflow synchrony with penumbral salvage, infarct growth, and intracranial hemorrhage (ICH) after MT. Results: A total of 151 patients (71.4 +/- 13.2 years, 65.6% women) were evaluated. Patients with synchronous SMCV outflow demonstrated significantly greater penumbral salvage (41.3 mL vs. 33.1 mL, P = 0.005) and lower infarct growth (9.0 mL vs. 14.4 mL, P = 0.015) compared to those with delayed SMCV outflow. Higher triangle FFT SMCV ( beta = -1.44, P = 0.013) and triangle TTP SMCV ( beta = -0.996, P = 0.003) significantly associated with lower penumbral salvage, while higher triangle FFT SMCV significantly associated with larger infarct growth ( beta = 1.09, P = 0.005) and increased risk of ICH (odds ratio [OR] = 1.519, P = 0.047). Conclusion: Synchronous SMCV outflow is an independent predictor of favorable tissue outcome and low ICH risk, and thereby carries the potential as an auxiliary radiological marker aiding the treatment planning of AIS patients.

Purpose: To guide the attention of a deep learning (DL) model toward MRI characteristics of brain lesions by incorporating radiology report-derived textual features to achieve interpretable lesion detection. Materials and Methods: In this retrospective study, 35 282 brain MRI scans (January 2018 to June 2023) and corresponding radiology reports from center 1 were used for training, validation, and internal testing. A total of 2655 brain MRI scans (January 2022 to December 2022) from centers 2-5 were reserved for external testing. Textual features were extracted from radiology reports to guide a DL model (ReportGuidedNet) focusing on lesion characteristics. Another DL model (PlainNet) without textual features was developed for comparative analysis. Both models identified 15 conditions, including 14 diseases and normal brains. Performance of each model was assessed by calculating macro-averaged area under the receiver operating characteristic curve (ma-AUC) and micro-averaged AUC (mi-AUC). Attention maps, which visualized model attention, were assessed with a five-point Likert scale. Results: ReportGuidedNet outperformed PlainNet for all diagnoses on both internal (ma-AUC, 0.93 [95% CI: 0.91, 0.95] vs 0.85 [95% CI: 0.81, 0.88]; mi-AUC, 0.93 [95% CI: 0.90, 0.95] vs 0.89 [95% CI: 0.83, 0.92]) and external (ma-AUC, 0.91 [95% CI: 0.88, 0.93] vs 0.75 [95% CI: 0.72, 0.79]; mi-AUC, 0.90 [95% CI: 0.87, 0.92] vs 0.76 [95% CI: 0.72, 0.80]) testing sets. The performance difference between internal and external testing sets was smaller for ReportGuidedNet than for PlainNet (Delta ma-AUC, 0.03 vs 0.10; Delta mi-AUC, 0.02 vs 0.13). The Likert scale score of ReportGuidedNet was higher than that of PlainNet (mean +/- SD: 2.50 +/- 1.09 vs 1.32 +/- 1.20; P < .001). Conclusion: The integration of radiology report textual features improved the ability of the DL model to detect brain lesions, thereby enhancing interpretability and generalizability.

目的 分析颈内动脉颅内闭塞(IICAO)患者CTA图像,判断经Willis环的初级侧支循环,评估血管内机械取栓术(EVMT)预后良好的相关因素.方法 对162例行EVMT治疗的IICAO患者进行分析.搜集基线资料包括人口统计学特征、血管危险因素、ASPECTS评分、NIHSS初始评分、术后再出血例数.根据CTA图像将Willis环的形态分为4种,并结合颈动脉闭塞位置及Willis环变异情况,将初级侧支循环分为4组.通过90天后改良Rankin量表(mRS)评分来衡量术后临床结局,判断取栓术预后良好的相关因素.结果 在162例IICAO患者中,出院38.3％(62例)取栓治疗后mRS 0～2分;预后良好29.0％(47例)mRS 3～5分;死亡32.7％(53例)mRS 6分.三组间基线资料中年龄、入院NIHSS评分及房颤有统计学差异(P＜0.05),Willis发育变异比例无统计学差异(P＞0.05),有无初级侧支循环有统计学差异(P＜0.05),其中无初级侧支循环的患者死亡率57.1％.在多变量Logistic回归中,房颤及血流循环分组为预后良好的相关因素.结论 基于颈动脉闭塞位置,结合Willis环变异联合判断是否存在初级侧支循环可能是IICAO患者EVMT后疗效的预测因子.

目的 探讨基于动态增强磁共振成像(DCE-MRI)定量参数图影像组学模型对术前子宫内膜癌Ki-67表达水平的预测价值.方法 回顾性搜集术前行DCE-MRI检查子宫内膜癌患者99例.根据Ki-67表达水平分为高表达组(42例)和低表达组(57例),按照样本量7:3随机分为训练集(69例)和验证集(30例).使用Omni-Ki-netics软件在DCE-MRI定量参数图上共提取201个影像组学特征,同时获取患者的独立特征.用Lasso回归对影像组学特征进行降维筛选和影像组学标签建立,采用K-S检验、t检验或卡方检验筛选有统计学差异的独立特征.Logistic回归用建立基于影像组学特征、独立特征和两者联合的诊断模型,采用受试者工作特征曲线下面积评估各模型预测效能,Delong检验用于比较各模型的预测效能.结果 最后筛选出7个影像组学特征和5个独立特征来构建模型.在训练集和验证集,联合模型和影像组学模型的预测效能均显著优于独立特征模型(P值均＜0.05).结论 基于DCE-MRI定量参数图构建的影像组学模型对术前预测子宫内膜癌Ki-67表达情况具有较高效能.

To develop a clot-based radiomics model using CT imaging radiomic features and machine learning to identify cardioembolic (CE) stroke before mechanical thrombectomy (MTB) in patients with acute ischemic stroke (AIS). This retrospective four-center study consecutively included 403 patients with AIS who sequentially underwent CT and MTB between April 2016 and July 2021. These were grouped into training, testing, and external validation cohorts. Thrombus-extracted radiomic features and basic information were gathered to construct a machine learning model to predict CE stroke. The radiological characteristics and basic information were used to build a routine radiological model. A combined radiomics and radiological features model was also developed. The performances of all models were evaluated and compared in the validation cohort. A histological analysis helped further assess the proposed model in all patients. The radiomics model yielded an area under the curve (AUC) of 0.838 (95 • Admission CT imaging could offer valuable information to identify the acute ischemic stroke source by radiomics analysis. • The proposed CT imaging–based radiomics model yielded a higher area under the curve (0.838) than the routine radiological method (0.713; p = 0.007). • Several radiomic features showed significantly stronger correlations with two main thrombus constituents (red blood cells, |r max |, 0.74; fibrin and platelet, |r max |, 0.68) than routine radiological characteristics.

RATIONALE AND OBJECTIVES:The measurement of the time since stroke onset (TSS) is crucial for decision-making in the treatment of acute ischemic stroke (AIS). This study assessed the utility of computed tomography angiography (CTA) radiomics features (RFs) to estimate TSS. MATERIALS AND METHODS:A total of 221 patients with AIS were enrolled in this retrospective study and were divided into a training group (n = 154) and a test group (n = 67). Thrombi in CTA images were manually outlined using ITK-SNAP. Images were aligned, normalized, and pre-processed to extract RFs. The TSS was calculated as the time from stroke onset to CTA completion. The patients were classified into two groups according to estimated TSS: ≤4.5 and >4.5 hours. A total of 944 RFs were extracted from CTA images. Clinical factors associated with TSS were identified using multivariate logistic regression, and a combined model (clinical data and RFs) was constructed. The predictive value of the models was assessed by the area under the receiver operating characteristic curve (AUC). The performance of the models was compared using the DeLong test, and clinical utility was evaluated by decision curve analysis. RESULTS:The AUC of the radiomics model was 0.803 (95% confidence interval [CI]: 0.733-0.873) and 0.803 (95% CI: 0.698-0.908) in the training and test cohorts, respectively. The AUC of the combined model (containing data on age, diabetes, and atrial fibrillation) in the training and test sets was 0.813 (95% CI: 0.750-0.889) and 0.803 (95% CI: 0.699-0.907), respectively. The DeLong test showed no significant difference between the radiomics and combined models. Decision curve analysis showed that both models had clinical utility. CONCLUSION:CTA-based thrombus radiomics can estimate TSS in patients with AIS. The addition of clinical data to the model does not improve predictive performance.