Background and purposeInfarct volume is a valuable outcome measure in treatment trials of acute ischemic stroke and is strongly associated with functional outcome. Its manual volumetric assessment is, however, too demanding to be implemented in clinical practice.ObjectiveTo assess the value of convolutional neural networks (CNNs) in the automatic segmentation of infarct volume in follow-up CT images in a large population of patients with acute ischemic stroke.Materials and methodsWe included CT images of 1026 patients from a large pooling of patients with acute ischemic stroke. A reference standard for the infarct segmentation was generated by manual delineation. We introduce three CNN models for the segmentation of subtle, intermediate, and severe hypodense lesions. The fully automated infarct segmentation was defined as the combination of the results of these three CNNs. The results of the three-CNNs approach were compared with the results from a single CNN approach and with the reference standard segmentations.ResultsThe median infarct volume was 48 mL (IQR 15–125 mL). Comparison between the volumes of the three-CNNs approach and manually delineated infarct volumes showed excellent agreement, with an intraclass correlation coefficient (ICC) of 0.88. Even better agreement was found for severe and intermediate hypodense infarcts, with ICCs of 0.98 and 0.93, respectively. Although the number of patients used for training in the single CNN approach was much larger, the accuracy of the three-CNNs approach strongly outperformed the single CNN approach, which had an ICC of 0.34.ConclusionConvolutional neural networks are valuable and accurate in the quantitative assessment of infarct volumes, for both subtle and severe hypodense infarcts in follow-up CT images. Our proposed three-CNNs approach strongly outperforms a more straightforward single CNN approach.

The aim of this study was to develop a convolutional neural network (CNN) that automatically detects and segments intra-arterial thrombi on baseline non-contrast computed tomography (NCCT) scans. We retrospectively collected computed tomography (CT)-scans of patients with an anterior circulation large vessel occlusion (LVO) from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands trial, both for training (n = 86) and validation (n = 43). For testing we included patients with (n = 58) and without (n = 45) an LVO from our comprehensive stroke center. Ground truth was established by consensus between two experts using both CT angiography and NCCT. We evaluated the CNN for correct identification of a thrombus, its location and thrombus segmentation and compared these with the results of a neurologist in training and expert neuroradiologist. Sensitivity of the CNN thrombus detection was 0.86, vs. 0.95 and 0.79 for the neuroradiologists. Specificity was 0.65 for the network vs. 0.58 and 0.82 for the neuroradiologists. The CNN correctly identified the location of the thrombus in 79% of the cases, compared to 81% and 77% for the neuroradiologists. The sensitivity and specificity for thrombus identification and the rate for correct thrombus location assessment by the CNN were similar to those of expert neuroradiologists.

Thrombus imaging characteristics are associated with treatment success and functional outcomes in stroke patients. However, assessing these characteristics based on manual annotations is labor intensive and subject to observer bias. Therefore, we aimed to create an automated pipeline for consistent and fast full thrombus segmentation. We used multi-center, multi-scanner datasets of anterior circulation stroke patients with baseline NCCT and CTA for training (n = 228) and testing (n = 100). We first found the occlusion location using StrokeViewer LVO and created a bounding box around it. Subsequently, we trained dual modality U-Net based convolutional neural networks (CNNs) to segment the thrombus inside this bounding box. We experimented with: (1) U-Net with two input channels for NCCT and CTA, and U-Nets with two encoders where (2) concatenate, (3) add, and (4) weighted-sum operators were used for feature fusion. Furthermore, we proposed a dynamic bounding box algorithm to adjust the bounding box. The dynamic bounding box algorithm reduces the missed cases but does not improve Dice. The two-encoder U-Net with a weighted-sum feature fusion shows the best performance (surface Dice 0.78, Dice 0.62, and 4% missed cases). Final segmentation results have high spatial accuracies and can therefore be used to determine thrombus characteristics and potentially benefit radiologists in clinical practice.

Introduction: Rapid detection of a large vessel occlusion (LVO) in acute ischemic stroke is beneficial as it may accelerate therapeutic management. The hyperdense artery sign (HAS) on non-contrast CT (NCCT) is result of an acute thrombolytic occlusion and is known as the earliest radiological marker to identify LVO. However, HAS is not apparent for all LVOs and detection by a less-trained eye can be sub-optimal. Artificial intelligence (AI) may support the non-trained radiologist in swift and accurate LVO detection. We evaluated the LVO detection accuracy of AI, and compared this to human expert performance. Materials and Methods: We used a convolutional neural network (CNN) developed by Nico.lab (www.nico-lab.com) that automatically detects and segments thrombi on NCCT using a patch-based approach. The CNN was trained using thrombus and non-thrombus patches, combined with the contralateral side. In a retrospective analysis of thin-slice NCCTs, obtained from Amsterdam UMC patients, with (ICA-T, M1, or M2) and without occlusions (stroke mimics), the CNN and two observers (>15y and >4y of experience) assessed LVO-presence and location. Ground truth was established by consensus between two different experts with >5y of experience, using both CTA and NCCT. Thrombus segmentations were considered accurate if segmentations overlapped with the ground truth. Sensitivity and specificity of the LVO detection were assessed. Results: We included 107 patients, of which 59 had proven LVOs. Nico.lab CNN showed a thrombus segmentation accuracy of 81%, vs. 81% and 77% of human experts. Sensitivity of LVO detection by CNN was 0.86, vs. 0.95 and 0.79 for human observers. Specificity was 0.65 for the CNN vs. 0.58 and 0.82 for human experts respectively. Conclusion: AI-based LVO detection on NCCT showed comparable results to expert observers. This supportive tool could lead to earlier detection of LVO in acute stroke situations. Fig 1A. HAS; B. CNN output (red); C. clot on CTA (blue arrow)

No abstract

No abstract

No abstract

No abstract