Radiomics assessment of bladder cancer grade using texture features from diffusion‐weighted imaging

Abstract

Purpose To 1) describe textural features from diffusion-weighted images (DWI) and apparent diffusion coefficient (ADC) maps that can distinguish low-grade bladder cancer from high-grade one, and 2) propose a radiomics-based strategy for cancer grading using texture features. Materials and Methods 61 patients with bladder cancer (29 in high- and 32 in low-grade groups) were enrolled in this retrospective study. Histogram- and gray-level co-occurrence matrix (GLCM)-based radiomics features were extracted from cancerous volumes of interest (VOIs) on DWI and corresponding ADC maps of each patient acquired from 3.0T MR. Mann-Whitney U test was applied to select features with significant difference between low-and high-grade groups (p<0.05). Then support vector machine with recursive feature elimination (SVM-RFE) and classification strategy was adopted to find an optimal feature subset and then to establish a classification model for grading. Results A total 102 features were derived from each VOI and among them, 47 candidate features were selected, which showed significant inter-group difference (p<0.05). By the SVM-RFE method, an optimal feature subset including 22 features was further selected from candidate features. The SVM classifier using the optimal feature subset achieved the best performance in bladder cancer grading, with an area under the receiver operating characteristic curve, accuracy, sensitivity and specificity of 0.861, 82.9%, 78.4% and 87.1%, respectively. Conclusion Textural features from DWI and ADC maps can reflect the difference between low- and high-grade bladder cancer, especially those GLCM features from ADC maps. The proposed radiomics strategy using these features, combined with the SVM classifier, may better facilitate image-based bladder cancer grading preoperatively.

Radiomics strategy for glioma grading using texture features from multiparametric MRI

“…Recently, the term radiomics has drawn attention in medical image research, which has two important aspects: 1) many quantitative features extracted from medical images that represent structural, physiopathologic, and genetic characteristics; and 2) machine‐learning models to deal with these image features and produce clinically significant output. So far, promising results have been obtained in many cancer studies based on the radiomics strategy, but few of these studies were conducted for glioma grading . Ryu et al and Brynolfsson et al reported that the gray‐level co‐occurrence matrix (GLCM)‐based texture features derived from DWI and corresponding apparent diffusion coefficient (ADC) maps of glioma patients could be used to represent a set of discriminative image patterns automatically, suggesting that GLCM texture features analysis may be a useful tool for glioma grading and prognosis prediction.…”

A predictive nomogram for individualized recurrence stratification of bladder cancer using multiparametric MRI and clinical risk factors

“…Based on these findings above, we hypothesized that: 1) the radiomics features extracted from preoperative mpMRI to characterize the subtle variations of tissue distribution within the lesion might be potential in predicting BCa recurrence; 2) the combination of the radiomics strategy with important clinical factors, mainly including age, gender, histological grade, and MIS of the archived tumor with the maximal size in bladder lumen, tumor size, NoT, operation choice, together with the imaging signs like stalk and SLE, might add the incremental value for TFTY BCa prediction.…”

Radiomics Strategy for Molecular Subtype Stratification of Lower‐Grade Glioma: Detecting IDH and TP53 Mutations Based on Multimodal MRI

“…The radiomics process usually requires: 1) large amounts of image features from medical images which could reflect structural, physiopathologic, and genetic information; and 2) machine‐learning models for excavating implicit clinical values of these image features. So far, many oncologic radiomics studies have yielded promising results . Specifically, radiomics features have been reported to be useful tools for glioma grading or prognosis prediction .…”