Multiparametric Iterative Self-Organizing Data Analysis of Ischemic Lesions Using Pre- or Post-Gd T1 MRI

Abstract

Background: The purpose of this work was to evaluate effects of Gd-diethylenetriaminepentacetic acid (DTPA) injection on T₁-weighted images of stroke and lesion segmentation and characterization results generated by our multiparametric iterative self-organizing data (ISODATA) method. The post-Gd image incorporates vasculature information into the analysis. Methods: Either a pre-Gd T₁-weighted image (T1WI) or a post-Gd T1WI was used along with diffusion-, T₂- and proton-density-weighted images in the analysis. ISODATA is a data-driven method that segments and characterizes tissue damage in stroke using multiparametric MRI. Results: Experimental results in both animal and human studies showed that the use of post-Gd T1WI modified the segmentation and characterization results on the periphery of the lesion. The peripheral region that changes with Gd-DTPA has a higher permeability compared to the rest of the lesion. Either of the data sets (including pre- or post-Gd T1WI) was used to estimate the tissue recovery and generated consistent results. Conclusions: This study shows that our multiparametric ISODATA approach consistently identifies and characterizes the core of the ischemic lesion. It also shows that the inclusion of post-Gd T1WI results in the segmentation and characterization of the lesion periphery if it has a higher permeability compared to the rest of the lesion. Finally, it confirms that the multiparametric ISODATA MRI characterizes tissue damage and recovery in stroke.

Medical image analysis methods in MR/CT-imaged acute-subacute ischemic stroke lesion: Segmentation, prediction and insights into dynamic evolution simulation models. A critical appraisal

“…ISODATA was used to identify ischemic tissue in similar studies (Soltanian-Zadeh et al, 2003, 2007): in Soltanian-Zadeh et al (2003) the identification process was extended to using an abnormality scoring system between 0 and 1 in rats, and in Soltanian-Zadeh et al (2007), an improved version was proposed, applied to rats and extended to human data including the role of reinforced reperfusion in the recovery process. In Hevia-Montiel et al (2007) a nonparametric clustering strategy using mean-shift algorithm and edge confidence map was developed to identify the lesion core in DW images.…”

“…Within this category, only one paper (Soltanian-Zadeh et al, 2007) addressed the phenomenon of reperfusion, only one paper (Martel et al, 1999) required a training data set and three papers (Jacobs et al, 2000, 2001a,b) required more than one acquisition timepoint. Although these methods applied more sophisticated techniques, the phenomenon of swelling in acute/subacute ischemic stroke was not addressed.…”

Fully automatic acute ischemic lesion segmentation in DWI using convolutional neural networks

“…However, manual segmentation of acute ischemic lesions is expensive in terms of time and human expertise. Several automatic and semi-automatic methods have been proposed to assist clinicians to address this problem (Charoensuk et al, 2015, Dwyer et al, 2008, Jacobs et al, 2000, Li et al, 2009, Mah et al, 2014, Martel et al, 1999, Soltanian-Zadeh et al, 2006). A common limitation of these models is that they were developed on small datasets which only contain tens of subjects.…”

Classification of lesion area in stroke patients during the subacute phase: A multiparametric MRI study

“…Despite evidence for the activation of compensation mechanisms during the subacute phase, mainly from animal models, including rats , rodents , and dogs , to date the vast majority of human stroke imaging studies have been performed during the acute phase (<12 h). MRI and/or computed tomography (CT) are usually performed during the acute phase, with the aim of identifying the salvageable tissue, i.e., the penumbra.…”