FLAIR and ADC Image-Based Radiomics Features as Predictive Biomarkers of Unfavorable Outcome in Patients With Acute Ischemic Stroke.

Guanmin Quan,Jia-Liang Ren,Ranran Ban,Tao Yuan,Weiwei Wang,Yawu Liu,Shipeng Dai

doi:10.3389/fnins.2021.730879

Abstract

At present, it is still challenging to predict the clinical outcome of acute ischemic stroke (AIS). In this retrospective study, we explored whether radiomics features extracted from fluid-attenuated inversion recovery (FLAIR) and apparent diffusion coefficient (ADC) images can predict clinical outcome of patients with AIS. Patients with AIS were divided into a training (n = 110) and an external validation (n = 80) sets. A total of 753 radiomics features were extracted from each FLAIR and ADC image of the 190 patients. Interquartile range (IQR), Wilcoxon rank sum test, and least absolute shrinkage and selection operator (LASSO) were used to reduce the feature dimension. The six strongest radiomics features were related to an unfavorable outcome of AIS. A logistic regression analysis was employed for selection of potential predominating clinical and conventional magnetic resonance imaging (MRI) factors. Subsequently, we developed several models based on clinical and conventional MRI factors and radiomics features to predict the outcome of AIS patients. For predicting unfavorable outcome [modified Rankin scale (mRS) > 2] in the training set, the area under the receiver operating characteristic curve (AUC) of ADC radiomics model was 0.772, FLAIR radiomics model 0.731, ADC and FLAIR radiomics model 0.815, clinical model 0.791, and clinical and conventional MRI model 0.782. In the external validation set, the AUCs for the prediction with ADC radiomics model was 0.792, FLAIR radiomics model 0.707, ADC and FLAIR radiomics model 0.825, clinical model 0.763, and clinical and conventional MRI model 0.751. When adding radiomics features to the combined model, the AUCs for predicting unfavorable outcome in the training and external validation sets were 0.926 and 0.864, respectively. Our results indicate that the radiomics features extracted from FLAIR and ADC can be instrumental biomarkers to predict unfavorable clinical outcome of AIS and would additionally improve predictive performance when adding to combined model.

Highlights

Acute ischemic stroke (AIS) is a critical cerebrovascular disorder worldwide with high morbidity and disability and accounts for 60–80% of all strokes (Darwish et al, 2020)
A univariate analysis showed that the following variables were significantly associated with the unfavorable outcome: age (P < 0.001), gender (P = 0.003), orthogonal diameters (ODs) (P = 0.006), admission National Institute of Health Stroke Scale (NIHSS) (P < 0.001), and diffusion-weighted imaging (DWI)-ASPECTS (P < 0.001)
We found that the radiomics signatures, especially those extracted from apparent diffusion coefficient (ADC) image, were associated with unfavorable outcome and was a value risk factor

Summary

Introduction

Acute ischemic stroke (AIS) is a critical cerebrovascular disorder worldwide with high morbidity and disability and accounts for 60–80% of all strokes (Darwish et al, 2020). Most of the clinical trials on AIS are based on computed tomography (CT), CT angiography (CTA), and CT perfusion (CTP), which provide several and fast information about cerebral ischemic tissue. Contrastenhanced CT techniques are not universally accepted methods in the routine workup in AIS patients in some institutions due to the possible risk of intravenous injection of iodinated contrast agent and technical complex. Non-contrast CT and multimodality magnetic resonance imaging (MRI) are used as substitute imaging modalities for clinical evaluation in AIS patients. Conventional MR stroke protocol, even without contrast injection, could be an alternative tool for providing both anatomic and functional information, including the lesion location and size, occluded vessels, diffusion characteristics, and cerebral blood perfusion obtained by arterial spin label (ASL) technique. The objective MRI markers would be useful to assist in predicting prognosis for an individual AIS patient

Objectives

Methods

Results

Discussion

Conclusion