Dataset Information

Noninvasive model for predicting future ischemic strokes in patients with silent lacunar infarction using radiomics.

ABSTRACT:

Background

This study aimed to investigate integrating radiomics with clinical factors in cranial computed tomography (CT) to predict ischemic strokes in patients with silent lacunar infarction (SLI).

Methods

Radiomic features were extracted from baseline cranial CT images of patients with SLI. A least absolute shrinkage and selection operator (LASSO)-Cox regression analysis was used to select significant prognostic factors based on Model^C with clinical factors, Model^R with radiomic features, and Model^CR with both factors. The Kaplan-Meier method was used to compare stroke-free survival probabilities. A nomogram and a calibration curve were used for further evaluation.

Results

Radiomic signature (p < 0.01), age (p = 0.09), dyslipidemia (p = 0.03), and multiple infarctions (p = 0.02) were independently associated with future ischemic strokes. Model^CR had the best accuracy with 6-, 12-, and 18-month areas under the curve of 0.84, 0.81, and 0.79 for the training cohort and 0.79, 0.88, and 0.75 for the validation cohort, respectively. Patients with a Model^CR score < 0.17 had higher probabilities of stroke-free survival. The prognostic nomogram and calibration curves of the training and validation cohorts showed acceptable discrimination and calibration capabilities (concordance index [95% confidence interval]: 0.7864 [0.70-0.86]; 0.7140 [0.59-0.83], respectively).

Conclusions

Radiomic analysis based on baseline CT images may provide a novel approach for predicting future ischemic strokes in patients with SLI. Older patients and those with dyslipidemia or multiple infarctions are at higher risk for ischemic stroke and require close monitoring and intensive intervention.

SUBMITTER: Su JH

PROVIDER: S-EPMC7346609 | biostudies-literature | 2020 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Noninvasive model for predicting future ischemic strokes in patients with silent lacunar infarction using radiomics.

Su Jie-Hua JH Meng Ling-Wei LW Dong Di D Zhuo Wen-Yan WY Wang Jian-Ming JM Liu Li-Bin LB Qin Yi Y Tian Ye Y Tian Jie J Li Zhao-Hui ZH

BMC medical imaging 20200708 1

<h4>Background</h4>This study aimed to investigate integrating radiomics with clinical factors in cranial computed tomography (CT) to predict ischemic strokes in patients with silent lacunar infarction (SLI).<h4>Methods</h4>Radiomic features were extracted from baseline cranial CT images of patients with SLI. A least absolute shrinkage and selection operator (LASSO)-Cox regression analysis was used to select significant prognostic factors based on Model<sup>C</sup> with clinical factors, Model<s ...[more]

PMID: 32641095

Similar Datasets

Project description:Neuroinflammation is believed to be involved in the pathophysiological mechanisms of silent brain infarcts (SBI). However, the immunological profile of SBI has been scarcely investigated. In the context of a national research project named SILENCE, aimed at investigating clinical, biochemical and pathogenic features of SBI, we have measured the plasma profile of some inflammatory-related molecules in SBI patients (n?=?21), patients with recent lacunar infarcts (LI, n?=?28) and healthy controls (n?=?31), consecutively enrolled in four Italian centres. A panel of chemokines (MIG, CTACK, IL16, SDF1a, MCP1), growth factors (SCF, SCGFb, HGF, IL3), immunoglobulin-type adhesion molecules (ICAM1, VCAM1), proinflammatory cytokines (IL18, INFa2, MIF, IL12p40), cell surface receptors on T-cells (IL2Ra), and inductors of apoptosis (TRAIL) was assessed in plasma samples by Luminex xMAP™ technology. Immunological parameters were compared using non-parametric statistics and performance to distinguish SBI and LI was evaluated by receiver operating characteristic (ROC) analysis. Plasma levels of ICAM1 were significantly higher in both SBI and LI patients as compared to controls (SBI?LI>Ctrl). A different trend was observed for IL16 (SBI<LI>Ctrl), SCF (LI<SBI>Ctrl) and SCGFb (SBI>LI<Ctrl). SBI subjects had significantly increased levels of MIG when compared to controls (LI?SBI>Ctrl) and IL18 when compared to LI patients (Ctrl?SBI>LI). All the other immunological markers did not significantly differ among groups. According to ROC analysis, the best predictor for SBI condition was the chemokine MIG (AUC?=?0.84, sensitivity 86%, specificity 77%), while SCF had the best performance in distinguishing LI patients (AUC?=?0.84, sensitivity 86%, specificity 68%). These results confirm the involvement of inflammatory processes in cerebrovascular disorders, particularly in SBI, a very common age-related condition. The differences in plasma profile of inflammatory molecules may underlie different pathological mechanisms in SBI and LI patients.

Project description:The outcome prediction of acute anterior circulation non-lacunar infarction (AACNLI) is important for the precise clinical treatment of this disease. However, the accuracy of prognosis prediction is still limited. This study aims to develop and compare machine learning models based on MRI radiomics of multiple ischaemic-related areas for prognostic prediction in AACNLI. This retrospective multicentre study consecutively included 372 AACNLI patients receiving MRI examinations and conventional therapy between October 2020 and February 2023. These were grouped into training set, internal test set and external test set. MRI radiomics features were extracted from the mask diffusion-weighted imaging, mask apparent diffusion coefficient (ADC) and mask ADC620 by AACNLI segmentations. Grid search parameter tuning was performed on 12 feature selection and 9 machine learning algorithms, and algorithm combinations with the smallest rank-sum of area under the curve (AUC) was selected for model construction. The performances of all models were evaluated in the internal and external test sets. The AUC of radiomics model was larger than that of non-radiomics model with the same machine learning algorithm in the three mask types. The radiomics model using least absolute shrinkage and selection operator-random forest algorithm combination gained the smallest AUC rank-sum among all the algorithm combinations. The AUC of the model with ADC620 was 0.98 in the internal test set and 0.91 in the external test set, and the weighted average AUC in the three sets was 0.96, the largest among three mask types. The Shapley additive explanations values of the maximum of National Institute of Health Stroke Scale score within 7 days from onset (7-d NIHSSmax), stroke-associated pneumonia and admission Glasgow coma scale score ranked top three among the features in AACNLI outcome prediction. In conclusion, the random forest model with mask ADC620 can accurately predict the AACNLI outcome and reveal the risk factors leading to the poor prognosis.

Project description:Although free radicals have been reported to play a role in the expansion of ischemic brain lesions, the effect of free radical scavengers is still under debate. In this study, the temporal profile of ischemic stroke lesion sizes was assessed for more than one year to evaluate the effect of edaravone which might reduce ischemic damage.We sequentially enrolled acute ischemic stroke patients, who admitted between April 2003 and March 2004, into the edaravone(-) group (n = 83) and, who admitted between April 2004 and March 2005, into the edaravone(+) group (n = 93). Because, edaravone has been used as the standard treatment after April 2004 in our hospital. To assess the temporal profile of the stroke lesion size, the ratio of the area [T2-weighted magnetic resonance images (T2WI)/iffusion-weighted magnetic resonance images (DWI)] were calculated. Observations on T2WI were continued beyond one year, and observational times were classified into subacute (1-2 months after the onset), early chronic (3-6 month), late chronic (7-12 months) and old (?13 months) stages. Neurological deficits were assessed by the National Institutes of Health Stroke Scale upon admission and at discharge and by the modified Rankin Scale at 1 year following stroke onset.Stroke lesion size was significantly attenuated in the edaravone(+) group compared with the edaravone(-) group in the period of early and late chronic observational stages. However, this reduction in lesion size was significant within a year and only for the small-vessel occlusion stroke patients treated with edaravone. Moreover, patients with small-vessel occlusion strokes that were treated with edaravone showed significant neurological improvement during their hospital stay, although there were no significant differences in outcome one year after the stroke.Edaravone treatment reduced the volume of the infarct and improved neurological deficits during the subacute period, especially in the small-vessel occlusion strokes.

Project description:BackgroundWidespread use of brain imaging in China has resulted in an increased prevalence of silent lacunar infarct (LACI) in addition to symptomatic LACI, but their clinical relevance is not fully understood.MethodsWe compared the 5-year risks of recurrent stroke and all-cause mortality for silent LACI vs symptomatic LACI in a prospective study of 489,597 Chinese adults with no history of stroke or ischemic heart disease at baseline. Data on recurrent stroke and all-cause mortality were obtained by linkage with local stroke and mortality registries and health insurance records for all hospital admissions.FindingsAmong 12,150 cases with an adjudicated diagnosis of first-ever LACI, 4,134 (34%) had silent LACI and 8,016 (66%) had symptomatic LACI. All cases had brain imaging, but only 33% of silent LACI and 40% of symptomatic LACI cases had brain magnetic resonance imaging (MRI). The standardized event rates for silent LACI were 2-fold greater in urban than rural areas, but the ratios of silent LACI vs symptomatic LACI were similar in all areas. Cases with silent LACI vs symptomatic LACI had comparable 5-year risks of recurrent stroke (38% vs 43%) and all-cause mortality (11% vs 14%), respectively. For both silent and symptomatic LACI cases, most cases of recurrent stroke had non-LACI (70% vs 72%). While the relative risks of recurrent stroke did not differ by age, sex and area, the absolute risks of all-cause mortality varied by sex, age and area.InterpretationThe prognosis of cases with silent LACI was comparable with symptomatic LACI, and the results highlight the need for further randomized trials assessing the efficacy and safety of established treatments for ischemic stroke in cases with silent LACI.FundingWellcome Trust (212946/Z/18/Z, 202922/Z/16/Z, 104085/Z/14/Z, 088158/Z/09/Z) and National Key Research and Development Program of China (2016YFC0900500, 2016YFC0900501, 2016YFC0900504, 2016YFC1303904) and National Natural Science Foundation of China (91843302); UK Medical Research Council (MC_UU_00017/1,MC_UU_12026/2 MC_U137686851), Cancer Research UK (C16077/A29186; C500/A16896) and British Heart Foundation (CH/1996001/9454). ZH was supported West China Hospital, Sichuan University (ZYGD18009 and 2016YFC1300505) for a visiting scholarship to the University of Oxford, UK, during 2018-19.

Dataset Information

Noninvasive model for predicting future ischemic strokes in patients with silent lacunar infarction using radiomics.

Background

Methods

Results

Conclusions

Publications

Noninvasive model for predicting future ischemic strokes in patients with silent lacunar infarction using radiomics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets