The aging population challenges the health-care system with chronic diseases. Cerebrovascular diseases are important components of these chronic conditions. Stroke is the acute cessation of blood in the brain, which can lead to rapid tissue loss. Therefore, fast, accurate, and reliable automatic methods are required to facilitate stroke management. The performance of artificial intelligence (AI) methods is increasing in all domains. Vision tasks, including natural images and medical images, are particularly benefiting from the skills of AI models. The AI methods that can be applied to stroke imaging have a broad range, including classical machine learning tools such as support vector machines, random forests, logistic regression, and linear discriminant analysis, as well as deep learning models, such as convolutional neural networks, recurrent neural networks, autoencoders, and U-Net. Both tools can be applied to various aspects of stroke management, including time-to-event onset determination, stroke confirmation, large vessel occlusion detection, difusion restriction, perfusion deficit, core and penumbra identification, afected region segmentation, and functional outcome prediction. While building these AI models, maximum care should be exercised in order to reduce bias and build generalizable models. One of the most important prerequisites for building unbiased models is collecting large, diverse, and quality data that reflects the underlying population well and splitting the training and testing parts in a way that both represent a similar distribution. Explainability and trustworthiness are other important properties of machine learning models that could be widely adopted in clinical practices. Cite this article as: Koska IO, Selver A. Artificial intelligence in stroke imaging: A comprehensive review. Eurasian J Med., 2023;55(Suppl. Issue: 1): S91-S97.

We aimed to build a robust classifier for the MGMT methylation status of glioblastoma in multiparametric MRI. We focused on multi-habitat deep image descriptors as our basic focus. A subset of the BRATS 2021 MGMT methylation dataset containing both MGMT class labels and segmentation masks was used. A comprehensive mask fusion approach was developed to select relevant image crops of diseased tissue. These fusion masks, which were guided by multiple sequences, helped collect information from the regions that seem disease-free to radiologists in standard MRI sequences while harboring pathology. Integrating the information in different MRI sequences and leveraging the high entropic capacity of deep neural networks, we built a 3D ROI-based custom CNN classifier for the automatic prediction of MGMT methylation status of glioblastoma in multi-parametric MRI. Single sequence-based classifiers reached intermediate predictive performance with 0.65, 0.71, 0.77, and 0.82 accuracy for T1W, T2W, T1 contrast-enhanced, and FLAIR sequences, respectively. The multiparametric classifier using T1 contrast-enhanced and FLAIR images reached 0.88 accuracy. The accuracy of the four-input model that used all sequences was 0.81. The best model reached 0.90 ROC AUC value. Integrating human knowledge in the form of relevant target selection was a useful approach in MGMT methylation status prediction in MRI. Exploration of means to integrate radiology knowledge into the models and achieve human-machine collaboration may help to develop better models. MGMT methylation status of glioblastoma is an important prognostic marker and is also important for treatment decisions. The preoperative non-invasive predictive ability and the explanation tools of the developed model may help clinicians to better understand imaging phenotypes of MGMT methylation status of glial tumors.

Objectives: Machine learning methods can be applied successfully to various medical imaging tasks. Our aim with this study was to build a robust classifier using radiomics and clinical data for preoperative diagnosis of Wilms tumor (WT) or neuroblastoma (NB) in pediatric abdominal CT. Material and methods: This is a single-center retrospective study approved by the Institutional Ethical Board. CT scans of consecutive patients diagnosed with WT or NB admitted to our hospital from January 2005 to December 2021 were evaluated. Three distinct datasets based on clinical centers and CT machines were curated. Robust, non-redundant, high variance, and relevant radiomics features were selected using data science methods. Clinically relevant variables were integrated into the final model. Dice score for similarity of tumor ROI, Cohen’s kappa for interobserver agreement among observers, and AUC for model selection were used. Results: A total of 147 patients, including 90 WT (mean age 34.78 SD: 22.06 months; 43 male) and 57 NB (mean age 23.77 SD:22.56 months; 31 male), were analyzed. After binarization at 24 months cut-off, there was no statistically significant difference between the two groups for age (p = .07) and gender (p = .54). CT clinic radiomics combined model achieved an F1 score of 0.94, 0.93 accuracy, and an AUC 0.96. Conclusion: In conclusion, the CT-based clinic-radiologic-radiomics combined model could noninvasively predict WT or NB preoperatively. Notably, that model correctly predicted two patients, which none of the radiologists could correctly predict. This model may serve as a noninvasive preoperative predictor of NB/WT differentiation in CT, which should be further validated in large prospective models. Clinical relevance statement: CT-based clinic-radiologic-radiomics combined model could noninvasively predict Wilms tumor or neuroblastoma preoperatively. Key Points: • CT radiomics features can predict Wilms tumor or neuroblastoma from abdominal CT preoperatively. • Integrating clinic variables may further improve the performance of the model. • The performance of the combined model is equal to or greater than human readers, depending on the lesion size.

Background: Radial head fractures are often evaluated in emergency departments and can easily be missed. Automated or semi-automated detection methods that help physicians may be valuable regarding the high miss rate. Purpose: To evaluate the accuracy of combined deep, transfer, and classical machine learning approaches on a small dataset for determination of radial head fractures. Material and Methods: A total of 48 patients with radial head fracture and 56 patients without fracture on elbow radiographs were retrospectively evaluated. The input images were obtained by cropping anteroposterior elbow radiographs around a center-point on the radial head. For fracture determination, an algorithm based on feature extraction using distinct prototypes of pretrained networks (VGG16, ResNet50, InceptionV3, MobileNetV2) representing four different approaches was developed. Reduction of feature space dimensions, feeding the most relevant features, and development of ensemble of classifiers were utilized. Results: The algorithm with the best performance consisted of preprocessing the input, computation of global maximum and global mean outputs of four distinct pretrained networks, dimensionality reduction by applying univariate and ensemble feature selectors, and applying Support Vector Machines and Random Forest classifiers to the transformed and reduced dataset. A maximum accuracy of 90% with MobileNetV2 pretrained features was reached for fracture determination with a small sample size. Conclusion: Radial head fractures can be determined with a combined approach and limitations of the small sample size can be overcome by utilizing pretrained deep networks with classical machine learning methods.