Open AccessSFCN: Symmetric feature comparison network for detecting ischemic stroke lesions on CT images
Author(s) -
Zhang Long,
Zhu Chuang,
Wu YueWei,
Yang Yang,
Luo Yihao,
Song Ruoning,
Liu Lian,
Yang Jie
Publication year - 2021
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/ipr2.12267
Subject(s) - stroke (engine) , ischemic stroke , feature (linguistics) , lesion , artificial intelligence , medicine , computer science , feature extraction , radiology , pattern recognition (psychology) , computer vision , pathology , ischemia , mechanical engineering , linguistics , philosophy , engineering
Abstract Ischemic stroke is the most common stroke and the leading cause ofdisability and death in the world. Computed tomography (CT) is a popular andeconomical diagnostic device for the stroke, However the ischemic strokelesions are not evident on CT images and the diagnostic result relies on thevisual observation of neurologists, which may vary from doctor to doctor. Tofacilitate the treatment, a computer‐aided detection algorithm on CT images isproposed to help clinician for the ischemic stroke screening. In order toobtain accurate lesion annotation on CT images, novel automatic algorithms aredeveloped to achieve image pairing, calibration, and registration. Then, a newframework with the symmetric feature extraction and comparison is proposed toidentify and locate the ischemic stroke lesion. Experimental results show thatthis method achieves 75% of DICE in the detection of ischemic stroke lesions,which is higher than other methods by 4%. Its competitive results compared withseven latest methods is shown in terms of extensive qualitative andquantitative evaluation. This method can accurately detect the lesion in the CTimages through the comparison of symmetric regional features, which hascontributed to the clinical diagnosis of ischemic stroke.