Brain lesion segmentation through image synthesis and outlier detection

• Published in: NeuroImage clinical Vol. 16; pp. 643 - 658
• Main Authors: Bowles, Christopher, Qin, Chen, Guerrero, Ricardo, Gunn, Roger, Hammers, Alexander, Dickie, David Alexander, Valdés Hernández, Maria, Wardlaw, Joanna, Rueckert, Daniel
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.01.2017; Elsevier
• Subjects: Adult; Algorithms; Brain - diagnostic imaging; Cerebral Small Vessel Diseases - diagnostic imaging; Female; Humans; Image Processing, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Male; Radiology; Regular; Support Vector Machine
• ISSN: 2213-1582; 2213-1582
• DOI: 10.1016/j.nicl.2017.09.003

Cerebral small vessel disease (SVD) can manifest in a number of ways. Many of these result in hyperintense regions visible on T2-weighted magnetic resonance (MR) images. The automatic segmentation of these lesions has been the focus of many studies. However, previous methods tended to be limited to certain types of pathology, as a consequence of either restricting the search to the white matter, or by training on an individual pathology. Here we present an unsupervised abnormality detection method which is able to detect abnormally hyperintense regions on FLAIR regardless of the underlying pathology or location. The method uses a combination of image synthesis, Gaussian mixture models and one class support vector machines, and needs only be trained on healthy tissue. We evaluate our method by comparing segmentation results from 127 subjects with SVD with three established methods and report significantly superior performance across a number of metrics. •A novel method for image synthesis is presented•The method is particularly suited to synthesising non-pathological brain images•Synthetic subject specific non-pathological images are used to segment hyperintense lesions on 127 magnetic resonance images•Resulting segmentations compare favourably to existing methods•A particular improvement is observed in the segmentation of smaller lesions, stroke lesions and lesions closer to the cortex