In the field of stroke imaging, deep learning (DL) has enormousuntapped potential.When clinically significant symptoms of a cerebral strokeare detected, it is crucial to make an urgent diagnosis using available imagin...In the field of stroke imaging, deep learning (DL) has enormousuntapped potential.When clinically significant symptoms of a cerebral strokeare detected, it is crucial to make an urgent diagnosis using available imagingtechniques such as computed tomography (CT) scans. The purpose of thiswork is to classify brain CT images as normal, surviving ischemia or cerebralhemorrhage based on the convolutional neural network (CNN) model. In thisstudy, we propose a computer-aided diagnostic system (CAD) for categorizingcerebral strokes using computed tomography images. Horizontal flip datamagnification techniques were used to obtain more accurate categorization.Image Data Generator to magnify the image in real time and apply anyrandom transformations to each training image. An early stopping method toavoid overtraining. As a result, the proposed methods improved several estimationparameters such as accuracy and recall, compared to other machinelearning methods. A python web application was created to demonstrate theresults of CNN model classification using cloud development techniques. Inour case, the model correctly identified the drawing class as normal with 79%accuracy. Based on the collected results, it was determined that the presentedautomated diagnostic system could be used to assist medical professionals indetecting and classifying brain strokes.展开更多
The task of segmentation of brain regions affected by ischemic stroke is help to tackle important challenges of modern stroke imaging analysis.Unfortunately,at the moment,the models for solving this problem using mach...The task of segmentation of brain regions affected by ischemic stroke is help to tackle important challenges of modern stroke imaging analysis.Unfortunately,at the moment,the models for solving this problem using machine learning methods are far from ideal.In this paper,we consider a modified 3D UNet architecture to improve the quality of stroke segmentation based on 3Dcomputed tomography images.We use the ISLES 2018(Ischemic Stroke Lesion Segmentation Challenge 2018)open dataset to train and test the proposed model.Interpretation of the obtained results,as well as the ideas for further experiments are included in the paper.Our evaluation is performed using the Dice or f1 score coefficient and the Jaccard index.Our architecture may simply be extended to ischemia segmentation and computed tomography image identification by selecting relevant hyperparameters.The Dice/f1 score similarity coefficient of our model shown58%and results close to ground truth which is higher than the standard 3D UNet model,demonstrating that our model can accurately segment ischemic stroke.The modified 3D UNet model proposed by us uses an efficient averaging method inside a neural network.Since this set of ISLES is limited in number,using the data augmentation method and neural network regularization methods to prevent overfitting gave the best result.In addition,one of the advantages is the use of the Intersection over Union loss function,which is based on the assessment of the coincidence of the shapes of the recognized zones.展开更多
文摘In the field of stroke imaging, deep learning (DL) has enormousuntapped potential.When clinically significant symptoms of a cerebral strokeare detected, it is crucial to make an urgent diagnosis using available imagingtechniques such as computed tomography (CT) scans. The purpose of thiswork is to classify brain CT images as normal, surviving ischemia or cerebralhemorrhage based on the convolutional neural network (CNN) model. In thisstudy, we propose a computer-aided diagnostic system (CAD) for categorizingcerebral strokes using computed tomography images. Horizontal flip datamagnification techniques were used to obtain more accurate categorization.Image Data Generator to magnify the image in real time and apply anyrandom transformations to each training image. An early stopping method toavoid overtraining. As a result, the proposed methods improved several estimationparameters such as accuracy and recall, compared to other machinelearning methods. A python web application was created to demonstrate theresults of CNN model classification using cloud development techniques. Inour case, the model correctly identified the drawing class as normal with 79%accuracy. Based on the collected results, it was determined that the presentedautomated diagnostic system could be used to assist medical professionals indetecting and classifying brain strokes.
文摘The task of segmentation of brain regions affected by ischemic stroke is help to tackle important challenges of modern stroke imaging analysis.Unfortunately,at the moment,the models for solving this problem using machine learning methods are far from ideal.In this paper,we consider a modified 3D UNet architecture to improve the quality of stroke segmentation based on 3Dcomputed tomography images.We use the ISLES 2018(Ischemic Stroke Lesion Segmentation Challenge 2018)open dataset to train and test the proposed model.Interpretation of the obtained results,as well as the ideas for further experiments are included in the paper.Our evaluation is performed using the Dice or f1 score coefficient and the Jaccard index.Our architecture may simply be extended to ischemia segmentation and computed tomography image identification by selecting relevant hyperparameters.The Dice/f1 score similarity coefficient of our model shown58%and results close to ground truth which is higher than the standard 3D UNet model,demonstrating that our model can accurately segment ischemic stroke.The modified 3D UNet model proposed by us uses an efficient averaging method inside a neural network.Since this set of ISLES is limited in number,using the data augmentation method and neural network regularization methods to prevent overfitting gave the best result.In addition,one of the advantages is the use of the Intersection over Union loss function,which is based on the assessment of the coincidence of the shapes of the recognized zones.