http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Delineation of ischemic lesion from brain MRI using attention gated fully convolutional network
R. Karthik,Menaka Radhakrishnan,R. Rajalakshmi,Joel Raymann 대한의용생체공학회 2021 Biomedical Engineering Letters (BMEL) Vol.11 No.1
Precise delineation of the ischemic lesion from unimodal Magnetic Resonance Imaging (MRI) is a challenging task due tothe subtle intensity difference between the lesion and normal tissues. Hence, multispectral MRI modalities are used for characterizingthe properties of brain tissues. Traditional lesion detection methods rely on extracting significant hand-engineeredfeatures to differentiate normal and abnormal brain tissues. But the identification of those discriminating features is quitecomplex, as the degree of differentiation varies according to each modality. This can be addressed well by ConvolutionalNeural Networks (CNN) which supports automatic feature extraction. It is capable of learning the global features from imageseffectively for image classification. But it loses the context of local information among the pixels that need to be retained forsegmentation. Also, it must provide more emphasis on the features of the lesion region for precise reconstruction. The majorcontribution of this work is the integration of attention mechanism with a Fully Convolutional Network (FCN) to segmentischemic lesion. This attention model is applied to learn and concentrate only on salient features of the lesion region bysuppressing the details of other regions. Hence the proposed FCN with attention mechanism was able to segment ischemiclesion of varying size and shape. To study the effectiveness of attention mechanism, various experiments were carried outon ISLES 2015 dataset and a mean dice coefficient of 0.7535 was obtained. Experimental results indicate that there is animprovement of 5% compared to the existing works.