To date, studies investigating radiomics-based predictive models have tended to err on the side of data-driven or exploratory analysis of many thousands of extracted features. In particular, spatial assessments of texture have proven to be especially adept at assessing for features of intratumoral heterogeneity in oncologic imaging, which likewise may correspond with tumor biology and behavior. These spatial assessments can be generally classified as spatial filters, which detect areas of rapid change within the grayscale in order to enhance edges and/or textures within an image, or neighborhood-based methods, which quantify gray-level differences of neighboring pixels/voxels within a set distance. Given the high dimensionality of radiomics datasets, data dimensionality reduction methods have been proposed in an attempt to optimize model performance in machine learning studies; however, it should be noted that these approaches should only be applied to training data in order to avoid information leakage and model overfitting. While area under the curve of the receiver operating characteristic is perhaps the most commonly reported assessment of model performance, it is prone to overestimation when output classifications are unbalanced. In such cases, confusion matrices may be additionally reported, whereby diagnostic cut points for model predicted probability may hold more clinical significance to clinical colleagues with respect to related forms of diagnostic testing.

To identify the top brain regions affected by MS-specific atrophy (i.e., atrophy in excess of normal aging) and to test whether normal aging and MS-specific atrophy increase or decrease in these regions with age. Six hundred fifty subjects (2,790 MRI time points) were analyzed: 520 subjects with relapse-onset MS from a 5-year prospective cohort with annual standardized 1-mm 3D T1-weighted images (3DT1s; 2,483 MRIs) and 130 healthy controls with longitudinal 3DT1s (307 MRIs). Rates of change in all FreeSurfer regions (v5.3) and Structural Image Evaluation Using Normalization of Atrophy (SIENA) were estimated with mixed-effects models. All FreeSurfer regions were ranked by the MS-specific atrophy slope/standard error ratio (βMS × time/SEβMS × time). In the top regions, age was added as an effect modifier to test whether MS-specific atrophy varied by age. The top-ranked regions were all gray matter structures. For SIENA, normal aging increased from 0.01%/y at age 30 years to -0.31%/y at age 60 years (-0.11% ± 0.032%/decade, p < 0.01), whereas MS-specific atrophy decreased from -0.38%/y at age 30 years to -0.12%/y at age 60 years (0.09% ± 0.035%/decade, p = 0.01). Similarly, in the thalamus, normal aging increased from -0.15%/y at age 30 years to -0.62%/y at age 60 years (-0.16% ± 0.079%/decade, p < 0.05), and MS-specific atrophy decreased from -0.59%/y at age 30 years to -0.05%/y at age 60 years (0.18% ± 0.08%/decade, p < 0.05). In the putamen and caudate, normal aging and MS-specific atrophy did not vary by age. For SIENA and thalamic atrophy, the contribution of normal aging increases with age, but does not change in the putamen and caudate. This may have substantial implications to understand the biology of brain atrophy in MS.

Objectives

To evaluate the performance of machine learning-augmented MRI-based radiomics models for predicting response to neoadjuvant chemotherapy (NAC) in soft tissue sarcomas.

Methods

Forty-four subjects were identified retrospectively from patients who received NAC at our institution for pathologically proven soft tissue sarcomas. Only subjects who had both a baseline MRI prior to initiating chemotherapy and a post-treatment scan at least 2 months after initiating chemotherapy and prior to surgical resection were included. 3D ROIs were used to delineate whole-tumor volumes on pre- and post-treatment scans, from which 1708 radiomics features were extracted. Delta-radiomics features were calculated by subtraction of baseline from post-treatment values and used to distinguish treatment response through univariate analyses as well as machine learning-augmented radiomics analyses.

Results

Though only 4.74% of variables overall reached significance at p ≤ 0.05 in univariate analyses, Laws Texture Energy (LTE)-derived metrics represented 46.04% of all such features reaching statistical significance. ROC analyses similarly failed to predict NAC response, with AUCs of 0.40 (95% CI 0.22-0.58) and 0.44 (95% CI 0.26-0.62) for RF and AdaBoost, respectively.

Conclusion

Overall, while our result was not able to separate NAC responders from non-responders, our analyses did identify a subset of LTE-derived metrics that show promise for further investigations. Future studies will likely benefit from larger sample size constructions so as to avoid the need for data filtering and feature selection techniques, which have the potential to significantly bias the machine learning procedures.

OBJECTIVE:Thalamic volume is a candidate magnetic resonance imaging (MRI)-based marker associated with neurodegeneration to hasten development of neuroprotective treatments. Our objective is to describe the longitudinal evolution of thalamic atrophy in MS and normal aging, and to estimate sample sizes for study design. METHODS:Six hundred one subjects (2,632 MRI scans) were analyzed. Five hundred twenty subjects with relapse-onset MS (clinically isolated syndrome, n = 90; relapsing-remitting MS, n = 392; secondary progressive MS, n = 38) underwent annual standardized 3T MRI scans for an average of 4.1 years, including a 1mm3 3-dimensional T1-weighted sequence (3DT1; 2,485 MRI scans). Eighty-one healthy controls (HC) were scanned longitudinally on the same scanner using the same protocol (147 MRI scans). 3DT1s were processed using FreeSurfer's longitudinal pipeline after lesion inpainting. Rates of normalized thalamic volume loss in MS and HC were compared in linear mixed effects models. Simulation-based sample size calculations were performed incorporating the rate of atrophy in HC. RESULTS:Thalamic volume declined significantly faster in MS subjects compared to HC, with an estimated decline of -0.71% per year (95% confidence interval [CI] = -0.77% to -0.64%) in MS subjects and -0.28% per year (95% CI = -0.58% to 0.02%) in HC (p for difference = 0.007). The rate of decline was consistent throughout the MS disease duration and across MS clinical subtypes. Eighty or 100 subjects per arm (α = 0.1 or 0.05, respectively) would be needed to detect the maximal effect size with 80% power in a 24-month study. INTERPRETATION:Thalamic atrophy occurs early and consistently throughout MS. Preliminary sample size calculations appear feasible, adding to its appeal as an MRI marker associated with neurodegeneration. Ann Neurol 2018;83:223-234.

Introduction

1.5 Tesla (1.5T) remain a significant field strength for brain imaging worldwide. Recent computer simulations and clinical studies at 3T MRI have suggested that dynamic susceptibility contrast (DSC) MRI using a 30° flip angle ("low-FA") with model-based leakage correction and no gadolinium-based contrast agent (GBCA) preload provides equivalent relative cerebral blood volume (rCBV) measurements to the reference-standard acquisition using a single-dose GBCA preload with a 60° flip angle ("intermediate-FA") and model-based leakage correction. However, it remains unclear whether this holds true at 1.5T. The purpose of this study was to test this at 1.5T in human high-grade glioma (HGG) patients.

Methods

This was a single-institution cross-sectional study of patients who had undergone 1.5T MRI for HGG. DSC-MRI consisted of gradient-echo echo-planar imaging (GRE-EPI) with a low-FA without preload (30°/P-); this then subsequently served as a preload for the standard intermediate-FA acquisition (60°/P+). Both normalized (nrCBV) and standardized relative cerebral blood volumes (srCBV) were calculated using model-based leakage correction (C+) with IBNeuro™ software. Whole-enhancing lesion mean and median nrCBV and srCBV from the low- and intermediate-FA methods were compared using the Pearson's, Spearman's and intraclass correlation coefficients (ICC).

Results

Twenty-three HGG patients composing a total of 31 scans were analyzed. The Pearson and Spearman correlations and ICCs between the 30°/P-/C+ and 60°/P+/C+ acquisitions demonstrated high correlations for both mean and median nrCBV and srCBV.

Conclusion

Our study provides preliminary evidence that for HGG patients at 1.5T MRI, a low FA, no preload DSC-MRI acquisition can be an appealing alternative to the reference standard higher FA acquisition that utilizes a preload.