Texture‐based classification of focal liver lesions on MRI at 3.0 Tesla: A feasibility study in cysts and hemangiomas

Abstract

Purpose: To determine the feasibility of texture analysis for the classification of liver cysts and hemangiomas, on nonenhanced, zero-fill interpolated T1-and T2-weighted MR images.Materials and Methods: Forty-five patients (26 women and 19 men; mean age, 58.1 6 16.9 years) with liver cysts or hemangiomas were enrolled in the study. After exclusion of images with artifacts, T1-weighted images of 42 patients, and T2-weighted images of 39 patients, obtained at 3.0 Tesla (T), were available for further analysis. Texture features derived from the gray-level histogram, co-occurrence and run-length matrix, gradient, autoregressive model, and wavelet transform were calculated. Fisher, probability of classification error and average correlation (POEþACC), and mutual information coefficients were used to extract subsets of optimized texture features. Linear discriminant analysis (LDA) in combination with k nearest neighbor (k-NN) classification, and k-means clustering, were used for lesion classification.Results: LDA/k-NN produced misclassification rates of 16-18% on T1-weighted, and 12-18% on T2-weighted images. K-means clustering yielded misclassification rates of 15-23% on T1-weighted, and 15-25% on T2-weighted images. Conclusion:Texture-based classification of liver cysts and hemangiomas is feasible on zero-fill interpolated MR images obtained at 3.0T. Further studies are warranted to investigate the value of texture-based classification of other liver lesions, such as hepatocellular and cholangiocellular carcinoma, on MRI.

Multi‐institutional validation of a novel textural analysis tool for preoperative stratification of suspected thyroid tumors on diffusion‐weighted MRI

“…Previous studies have shown that zero-fill interpolated images enhance physically distinct structures’ textural differences (29,30). However, it is routine clinical practice to use MR scanner software to interpolate images by automatic zero-filling of k-space to achieve a resolution of 256×256.…”

MALDI imaging MS reveals candidate lipid markers of polycystic kidney disease

“…Recently, nonimaging MALDI-TOF MS has been combined with linear discriminant analysis to achieve that goal ( 29 ). In microarray data sets even larger than MALDI images, Fisher discriminant analysis (FDA), a related discriminant analysis method, has successfully been applied for gene selection ( 30 ); it has previously been used for discrimination of magnetic resonance or positron emission tomography images (31)(32)(33). FDA does not make the assumptions of a t -test, such as normally distributed classes, and may therefore be suitable for analysis of MALDI-IMS data sets for which no normal distribution can be assumed.…”

MRI Texture Analysis Predicts p53 Status in Head and Neck Squamous Cell Carcinoma

“…38 An earlier study of methylation-status prediction in glioblastoma yielded accuracies of up to 93.2%. 27 Other studies have shown that normal liver tissue could be differentiated from focal liver lesions with accuracies up to 88% 39 and that normal bladder wall could be differentiated from bladder cancer with an accuracy of 87.0%. 40 We found that texture features of the Post-Gad T1WI, T2WI, and ADC map were influential in the analysis.…”