Multiparametric Tissue Characterization of Brain Neoplasms and Their Recurrence Using Pattern Classification of MR Images

Abstract

Rationale and Objectives: Treatment of brain neoplasms can greatly benefit from better delineation of bulk neoplasm boundary and the extent and degree of more subtle neoplastic infiltration. MRI is the primary imaging modality for evaluation before and after therapy, typically combining conventional sequences with more advanced techniques like perfusion-weighted imaging and diffusion tensor imaging (DTI). The purpose of this study is to quantify the multi-parametric imaging profile of neoplasms by integrating structural MRI and DTI via statistical image analysis methods, in order to potentially capture complex and subtle tissue characteristics that are not obvious from any individual image or parameter. Materials and Methods:Five structural MR sequences, namely, B0, Diffusion Weighted Images, FLAIR, T1-weighted, and gadolinium-enhanced T1-weighted, and two scalar maps computed from DTI, i.e., fractional anisotropy and apparent diffusion coefficient, are used to create an intensitybased tissue profile. This is incorporated into a non-linear pattern classification technique to create a multi-parametric probabilistic tissue characterization, which is applied to data from 14 patients with newly diagnosed primary high grade neoplasms who have not received any therapy prior to imaging.Results: Preliminary results demonstrate that this multi-parametric tissue characterization helps to better differentiate between neoplasm, edema and healthy tissue, and to identify tissue that is likely progress to neoplasm in the future. This has been validated on expert assessed tissue. Conclusion:This approach has potential applications in treatment, aiding computer-assisted surgery by determining the spatial distributions of healthy and neoplastic tissue, as well as in identifying tissue that is relatively more prone to tumor recurrence.

Decision Forests for Tissue-Specific Segmentation of High-Grade Gliomas in Multi-channel MR

“…A large group of discriminative methods applies learning techniques to the problem [2,4,7,13,15,16]. Our method belongs to this group.…”

“…by modeling the boundary [8,11], or by applying a variant of a random field spatial prior (MRF/CRF) [4,7,16]. Works which classify multiple labels [2,15] often use SVMs, which are inherently binary classifiers. In order to classify different tissues, they are applied hierarchically [2], or in the one-versus-all manner [15].…”

“…low-grade [6], high-grade [4,13], pediatric tumors [16]), the type of anomalous tissues being detected (e.g. GT [7,10,16]; GT and E [4,6,12,15]; AC, NC, E [2,15]), input data, and the evaluation procedure.…”

“…Works which classify multiple labels [2,15] often use SVMs, which are inherently binary classifiers. In order to classify different tissues, they are applied hierarchically [2], or in the one-versus-all manner [15]. For these approaches, several classes have to be grouped together, a step which can make the distribution inside the aggregate group more complex than the distribution of each individual class.…”

Deep patch-based priors under a fully convolutional encoder-decoder architecture for interstitial lung disease segmentation

“…However, patch-based classification of lung tissue is not very appropriate for disease segmentation since it fails to capture accurate and smooth boundaries between different tissue types. Intra-patient classifiers [7] could be used to estimate the disease in the remaining part (out of patch), but such patient-specific models are sensitive to outliers due to their limited view. ILD segmentation, i.e.…”