Impact of Severe Extracranial ICA Stenosis on MRI Perfusion and Diffusion Parameters in Acute Ischemic Stroke

Deep Learning in Ischemic Stroke Imaging Analysis: A Comprehensive Review

Ischemic stroke is a cerebrovascular disease with a high morbidity and mortality rate, which poses a serious challenge to human health and life. Meanwhile, the management of ischemic stroke remains highly dependent on manual visual analysis of noncontrast computed tomography (CT) or magnetic resonance imaging (MRI). However, artifacts and noise of the equipment as well as the radiologist experience play a significant role on diagnostic accuracy. To overcome these defects, the number of computer-aided diagnostic (CAD) methods for ischemic stroke is increasing substantially during the past decade. Particularly, deep learning models with massive data learning capabilities are recognized as powerful auxiliary tools for the acute intervention and guiding prognosis of ischemic stroke. To select appropriate interventions, facilitate clinical practice, and improve the clinical outcomes of patients, this review firstly surveys the current state-of-the-art deep learning technology. Then, we summarized the major applications in acute ischemic stroke imaging, particularly in exploring the potential function of stroke diagnosis and multimodal prognostication. Finally, we sketched out the current problems and prospects.

Early Diagnosis of Intracranial Internal Carotid Artery Stenosis Using Extracranial Hemodynamic Indices from Carotid Doppler Ultrasound

Atherosclerotic intracranial internal carotid artery stenosis (IICAS) is a leading cause of strokes. Due to the limitations of major cerebral imaging techniques, the early diagnosis of IICAS remains challenging. Clinical studies have revealed that arterial stenosis may have complicated effects on the blood flow’s velocity from a distance. Therefore, based on a patient-specific one-dimensional hemodynamic model, we quantitatively investigated the effects of IICAS on extracranial internal carotid artery (ICA) flow velocity waveforms to identify sensitive hemodynamic indices for IICAS diagnoses. Classical hemodynamic indices, including the peak systolic velocity (PSV), end-diastolic velocity (EDV), and resistive index (RI), were calculated on the basis of simulations with and without IICAS. In addition, the first harmonic ratio (FHR), which is defined as the ratio between the first harmonic amplitude and the sum of the amplitudes of the 1st−20th order harmonics, was proposed to evaluate flow waveform patterns. To investigate the diagnostic performance of the indices, we included 52 patients with mild-to-moderate IICAS (<70%) in a case−control study and considered 24 patients without stenosis as controls. The simulation analyses revealed that the existence of IICAS dramatically increased the FHR and decreased the PSV and EDV in the same patient. Statistical analyses showed that the average PSV, EDV, and RI were lower in the stenosis group than in the control group; however, there were no significant differences (p > 0.05) between the two groups, except for the PSV of the right ICA (p = 0.011). The FHR was significantly higher in the stenosis group than in the control group (p < 0.001), with superior diagnostic performance. Taken together, the FHR is a promising index for the early diagnosis of IICAS using carotid Doppler ultrasound methods.

Management and outcome of patients with acute ischemic stroke and tandem carotid occlusion in the ESCAPE-NA1 trial

BACKGROUND

The optimal treatment and prognosis for stroke patients with tandem cervical carotid occlusion are unclear. We analyzed outcomes and treatment strategies of tandem occlusion patients in the ESCAPE-NA1 trial.

METHODS

ESCAPE-NA1 was a multicenter international randomized trial of nerinetide versus placebo in 1105 patients with acute ischemic stroke who underwent endovascular treatment. We defined tandem occlusions as complete occlusion of the cervical internal carotid artery (ICA) on catheter angiography, in addition to a proximal ipsilateral intracranial large vessel occlusion. Baseline characteristics and outcome parameters were compared between patients with tandem occlusions versus those without, and between patients with tandem occlusion who underwent ICA stenting versus those who did not. The influence of tandem occlusions on functional outcome was analyzed using multivariable regression modeling.

RESULTS

Among 115/1105 patients (10.4%) with tandem occlusions, 62 (53.9%) received stenting for the cervical ICA occlusion. Of these, 46 (74.2%) were stented after and 16 (25.8%) before the intracranial thrombectomy. A modified Rankin Score (mRS) of 0-2 at 90 days was achieved in 82/115 patients (71.3%) with tandem occlusions compared with 579/981 (59.5%) patients without tandem occlusions. Tandem occlusion did not impact functional outcome in the adjusted analysis (OR 1.5, 95% CI 0.95 to 2.4). Among the subgroup of patients with tandem occlusion, cervical carotid stenting was not associated with different outcomes compared with no stenting (mRS 0-2: 75.8% vs 66.0%, adjusted OR 2.0, 95% CI 0.8 to 5.1).

CONCLUSIONS

Tandem cervical carotid occlusion in patients with acute large vessel stroke did not lower the odds of good functional outcome in our study. Functional outcomes were similar irrespective of the management of the cervical ICA occlusion (stenting vs not stenting).

Stroke Lesion Segmentation in FLAIR MRI Datasets Using Customized Markov Random Fields

Robust and reliable stroke lesion segmentation is a crucial step toward employing lesion volume as an independent endpoint for randomized trials. The aim of this work was to develop and evaluate a novel method to segment sub-acute ischemic stroke lesions from fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) datasets. After preprocessing of the datasets, a Bayesian technique based on Gabor textures extracted from the FLAIR signal intensities is utilized to generate a first estimate of the lesion segmentation. Using this initial segmentation, a customized voxel-level Markov random field model based on intensity as well as Gabor texture features is employed to refine the stroke lesion segmentation. The proposed method was developed and evaluated based on 151 multi-center datasets from three different databases using a leave-one-patient-out validation approach. The comparison of the automatically segmented stroke lesions with manual ground truth segmentation revealed an average Dice coefficient of 0.582, which is in the upper range of previously presented lesion segmentation methods using multi-modal MRI datasets. Furthermore, the results obtained by the proposed technique are superior compared to the results obtained by two methods based on convolutional neural networks and three phase level-sets, respectively, which performed best in the ISLES 2015 challenge using multi-modal imaging datasets. The results of the quantitative evaluation suggest that the proposed method leads to robust lesion segmentation results using FLAIR MRI datasets only as a follow-up sequence.

A Comparison of Relative Time to Peak and Tmax for Mismatch-Based Patient Selection

BACKGROUND AND PURPOSE

The perfusion-weighted imaging (PWI)/diffusion-weighted imaging (DWI) mismatch profile is used to select patients for endovascular treatment. A PWI map of Tmax is commonly used to identify tissue with critical hypoperfusion. A time to peak (TTP) map reflects similar hemodynamic properties with the added benefit that it does not require arterial input function (AIF) selection and deconvolution. We aimed to determine if TTP could substitute Tmax for mismatch categorization.

METHODS

Imaging data of the DEFUSE 2 trial were reprocessed to generate relative TTP (rTTP) maps. We identified the rTTP threshold that yielded lesion volumes comparable to Tmax > 6 s and assessed the effect of reperfusion according to mismatch status, determined based on Tmax and rTTP volumes.

RESULTS

Among 102 included cases, the Tmax > 6 s lesion volumes corresponded most closely with rTTP > 4.5 s lesion volumes: median absolute difference 6.9 mL (IQR: 2.3-13.0). There was 94% agreement in mismatch classification between Tmax and rTTP-based criteria. When mismatch was assessed by Tmax criteria, the odds ratio (OR) for favorable clinical response associated with reperfusion was 7.4 (95% CI 2.3-24.1) in patients with mismatch vs. 0.4 (95% CI 0.1-2.6) in patients without mismatch. When mismatch was assessed with rTTP criteria, these ORs were 7.2 (95% CI 2.3-22.2) and 0.3 (95% CI 0.1-2.2), respectively.

CONCLUSION

rTTP yields lesion volumes that are comparable to Tmax and reliably identifies the PWI/DWI mismatch profile. Since rTTP is void of the problems associated with AIF selection, it is a suitable substitute for Tmax that could improve the robustness and reproducibility of mismatch classification in acute stroke.

Deep into the Brain: Artificial Intelligence in Stroke Imaging

Artificial intelligence (AI), a computer system aiming to mimic human intelligence, is gaining increasing interest and is being incorporated into many fields, including medicine. Stroke medicine is one such area of application of AI, for improving the accuracy of diagnosis and the quality of patient care. For stroke management, adequate analysis of stroke imaging is crucial. Recently, AI techniques have been applied to decipher the data from stroke imaging and have demonstrated some promising results. In the very near future, such AI techniques may play a pivotal role in determining the therapeutic methods and predicting the prognosis for stroke patients in an individualized manner. In this review, we offer a glimpse at the use of AI in stroke imaging, specifically focusing on its technical principles, clinical application, and future perspectives.

Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study

MOTIVATION

Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various factors, such as the high variance in lesion shape, location and appearance, render it a difficult task.

METHODS

In this article, nine classification methods (e.g. Generalized Linear Models, Random Decision Forests and Convolutional Neural Networks) are evaluated and compared with each other using 37 multiparametric MRI datasets of ischemic stroke patients in the sub-acute phase in terms of their accuracy and reliability for ischemic stroke lesion segmentation. Within this context, a multi-spectral classification approach is compared against mono-spectral classification performance using only FLAIR MRI datasets and two sets of expert segmentations are used for inter-observer agreement evaluation.

RESULTS AND CONCLUSION

The results of this study reveal that high-level machine learning methods lead to significantly better segmentation results compared to the rather simple classification methods, pointing towards a difficult non-linear problem. The overall best segmentation results were achieved by a Random Decision Forest and a Convolutional Neural Networks classification approach, even outperforming all previously published results. However, none of the methods tested in this work are capable of achieving results in the range of the human observer agreement and the automatic ischemic stroke lesion segmentation remains a complicated problem that needs to be explored in more detail to improve the segmentation results.