Limited One-time Sampling Irregularity Age Map (LOTS-IAM): Automatic Unsupervised Detection of Brain White Matter Abnormalities in Structural Magnetic Resonance Images [article]

Muhammad Febrian Rachmadi, Maria Valdés Hernández, Hongwei Li, Ricardo Guerrero, Jianguo Zhang, Daniel Rueckert, Taku Komura
2018 bioRxiv   pre-print
We propose a novel unsupervised approach of detecting and segmenting white matter abnormalities, using limited one-time sampling irregularity age map (LOTS-IAM). LOTS-IAM is a fully automatic unsupervised approach to extract brain tissue irregularities in magnetic resonance images (MRI) (e.g. T2-FLAIR white matter hyperintensities (WMH)). In this study, the limited one-time sampling scheme is proposed and implemented on GPU. We compared the performance of LOTS-IAM in detecting and segmenting
more » ... , with three unsupervised methods; the original IAM, one-time sampling IAM (OTS-IAM) and Lesion Growth Algorithm from public toolbox Lesion Segmentation Toolbox (LST-LGA), and two conventional supervised machine learning algorithms; support vector machine (SVM) and random forest (RF). Furthermore, we also compared LOTS-IAM's performance with five supervised deep neural networks algorithms; deep Boltzmann machine (DBM), convolutional encoder network (CEN), and three convolutional neural network (CNN) schemes: the 2D implementation of DeepMedic with the addition of global spatial information (2D-CNN-GSI), patch-uResNet and patch-uNet. Based on our experiments, LOTS-IAM outperformed LST-LGA, the state-of-the-art of unsupervised WMH segmentation method, both in performance and processing speed. Our method also outperformed supervised conventional machine learning algorithms SVM and RF, and supervised deep neural networks algorithms DBM and CEN.
doi:10.1101/334292 fatcat:ouw5mtiobjeghi6zqs56v7payq