Limited One-time Sampling Irregularity Map (LOTS-IM) for Automatic Unsupervised Assessment of White Matter Hyperintensities and Multiple Sclerosis Lesions in Structural Brain Magnetic Resonance Images

Abstract

We present the application of limited one-time sampling irregularity map (LOTS-IM): a fully automatic unsupervised approach to extract brain tissue irregularities in magnetic resonance images (MRI), for quantitatively assessing white matter hyperintensities (WMH) of presumed vascular origin, and multiple sclerosis (MS) lesions and their progression. LOTS-IM generates an irregularity map (IM) that represents all voxels as irregularity values with respect to the ones considered "normal". Unlike probability values, IM represents both regular and irregular regions in the brain based on the original MRI's texture information. We evaluated and compared the use of IM for WMH and MS lesions segmentation on T2-FLAIR MRI with the stateof-the-art unsupervised lesions' segmentation method, Lesion Growth Algorithm from the public toolbox Lesion Segmentation Toolbox (LST-LGA), with several well established conventional supervised machine learning schemes and with state-of-the-art supervised deep learning methods for WMH segmentation. In our experiments, LOTS-IM outperformed unsupervised method LST-LGA on WMH segmentation, both in performance and processing speed, thanks to the limited one-time sampling scheme and its implementation on GPU. Our method also outperformed supervised conventional machine learning algorithms (i.e., support vector machine (SVM) and random forest (RF)) and deep learning algorithms (i.e., deep Boltzmann machine (DBM) and convolutional encoder network (CEN)), while yielding comparable results to the convolutional neural network schemes that rank top of the algorithms developed up to date for this purpose (i.e., UResNet and UNet). LOTS-IM also performed well on MS lesions segmentation, performing similar to LST-LGA. On the other hand, the high sensitivity of IM on depicting signal change deems suitable for assessing MS progression, although care must be taken with signal changes not reflective of a true pathology.

Automatic Spatial Estimation of White Matter Hyperintensities Evolution in Brain MRI using Disease Evolution Predictor Deep Neural Networks

“…The IM is unique as it retains some of the original MRI textures (e.g., from the T2-FLAIR image intensi-ties), including gradients of WMH. IM is also independent from a human rater or training data, as it is produced using an unsupervised method (i.e., LOTS-IM) (Rachmadi et al, 2019b). Furthermore, previous studies have shown that IM can also be used for WMH segmentation (Rachmadi et al, 2018b), data augmentation of supervised WMH segmentation (Jeong et al, 2019), and simulation of WMH progression and regression (Rachmadi et al, 2018c).…”

“…The WMH segmentation of x 1 and x 1 is estimated by either thresholding IM values (i.e., irregularity values) to be above 0.178 (Rachmadi et al, 2019b) or PM values (i.e., probability values) to be above 0.5. Furthermore, each term in Equation 4 is weighted by λ 1 , λ 2 , and λ 3 which equals to 100 (Baumgartner et al, 2017), 1 and 100 respectively.…”

“…As previously explained, IM and PM are needed for DEP-GAN (i.e., the non-supervised learning approach of DEP model). We used LOTS-IM with 128 target patches (Rachmadi et al, 2019b) to generate IM from each MRI data. To generate PM, we trained a 2D UResNet (Guerrero et al, 2018) with gold standard WMH and SL masks for WMH and SL segmentation.…”

“…However, good details of the generated DEM from IM/PM did not necessarily translate to good three-class DEM label (i.e., three labels of growing, shrinking, and stable WMH) as depicted in Figure 11. Some reasons that might have caused this are; 1) the generated DEM from IM/PM is result of a regression process from the baseline IM/PM using DEP-GAN and 2) the threeclass DEM label itself is generated from the resulted regression, where WMH is defined by having ir- regularity/probability values greater than or equal to 0.178 for IM and 0.5 for PM (Rachmadi et al, 2019b). Note that regression of the whole brain using IM/PM is harder than direct segmentation of three regions of WMH (i.e., stable, shrinking, and growing WMH).…”

Predicting the Evolution of White Matter Hyperintensities in Brain MRI using Generative Adversarial Networks and Irregularity Map

“…From the co-registered scans [11], we generated IM using LOTS-IM [7] with 128 target patches and removed the extracranial tissues and skull from the baseline and follow-up T2-FLAIR images. Then, T2-FLAIR values were normalised between 0 and 1, similar to IM's values.…”