Texture quantification of medical images using a novel complex space-frequency transform

Single-timepoint low-dimensional characterization and classification of acute versus chronic multiple sclerosis lesions using machine learning

Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disease characterized by the appearance of focal lesions across the central nervous system. The discrimination of acute from chronic MS lesions may yield novel biomarkers of inflammatory disease activity which may support patient management in the clinical setting and provide endpoints in clinical trials. On a single timepoint and in the absence of a prior reference scan, existing methods for acute lesion detection rely on the segmentation of hyperintense foci on post-gadolinium T1-weighted magnetic resonance imaging (MRI), which may underestimate recent acute lesion activity. In this paper, we aim to improve the sensitivity of acute MS lesion detection in the single-timepoint setting, by developing a novel machine learning approach for the automatic detection of acute MS lesions, using single-timepoint conventional non-contrast T1- and T2-weighted brain MRI. The MRI input data are supplemented via the use of a convolutional neural network generating "lesion-free" reconstructions from original "lesion-present" scans using image inpainting. A multi-objective statistical ranking module evaluates the relevance of textural radiomic features from the core and periphery of lesion sites, compared within "lesion-free" versus "lesion-present" image pairs. Then, an ensemble classifier is optimized through a recursive loop seeking consensus both in the feature space (via a greedy feature-pruning approach) and in the classifier space (via model selection repeated after each pruning operation). This leads to the identification of a compact textural signature characterizing lesion phenotype. On the patch-level task of acute versus chronic MS lesion classification, our method achieves a balanced accuracy in the range of 74.3-74.6% on fully external validation cohorts.

Texture Analysis of T2-Weighted MR Images to Assess Acute Inflammation in Brain MS Lesions

Brain blood barrier breakdown as assessed by contrast-enhanced (CE) T1-weighted MR imaging is currently the standard radiological marker of inflammatory activity in multiple sclerosis (MS) patients. Our objective was to evaluate the performance of an alternative model assessing the inflammatory activity of MS lesions by texture analysis of T2-weighted MR images. Twenty-one patients with definite MS were examined on the same 3.0T MR system by T2-weighted, FLAIR, diffusion-weighted and CE-T1 sequences. Lesions and mirrored contralateral areas within the normal appearing white matter (NAWM) were characterized by texture parameters computed from the gray level co-occurrence and run length matrices, and by the apparent diffusion coefficient (ADC). Statistical differences between MS lesions and NAWM were analyzed. ROC analysis and leave-one-out cross-validation were performed to evaluate the performance of individual parameters, and multi-parametric models using linear discriminant analysis (LDA), partial least squares (PLS) and logistic regression (LR) in the identification of CE lesions. ADC and all but one texture parameter were significantly different within white matter lesions compared to within NAWM (p < 0.0167). Using LDA, an 8-texture parameter model identified CE lesions with a sensitivity Se = 70% and a specificity Sp = 76%. Using LR, a 10-texture parameter model performed better with Se = 86% / Sp = 84%. Using PLS, a 6-texture parameter model achieved the highest accuracy with Se = 88% / Sp = 81%. Texture parameter from T2-weighted images can assess brain inflammatory activity with sufficient accuracy to be considered as a potential alternative to enhancement on CE T1-weighted images.

Texture analysis on MR images helps predicting non-response to NAC in breast cancer

BACKGROUND

To assess the performance of a predictive model of non-response to neoadjuvant chemotherapy (NAC) in patients with breast cancer based on texture, kinetic, and BI-RADS parameters measured from dynamic MRI.

METHODS

Sixty-nine patients with invasive ductal carcinoma of the breast who underwent pre-treatment MRI were studied. Morphological parameters and biological markers were measured. Pathological complete response was defined as the absence of invasive and in situ cancer in breast and nodes. Pathological non-responders, partial and complete responders were identified. Dynamic imaging was performed at 1.5 T with a 3D axial T1W GRE fat-suppressed sequence. Visual texture, kinetic and BI-RADS parameters were measured in each lesion. ROC analysis and leave-one-out cross-validation were used to assess the performance of individual parameters, then the performance of multi-parametric models in predicting non-response to NAC.

RESULTS

A model based on four pre-NAC parameters (inverse difference moment, GLN, LRHGE, wash-in) and k-means clustering as statistical classifier identified non-responders with 84 % sensitivity. BI-RADS mass/non-mass enhancement, biological markers and histological grade did not contribute significantly to the prediction.

CONCLUSION

Pre-NAC texture and kinetic parameters help predicting non-benefit to NAC. Further testing including larger groups of patients with different tumor subtypes is needed to improve the generalization properties and validate the performance of the predictive model.

Texture analysis of images using a two-dimensional fast time-frequency transform

The two-dimensional S-transform (ST-2D) is a time-frequency representation that is widely used in medical image processing but prohibitive in both storage and computation time. The high computation time required for generating local spectrum discourages the use of ST-2D for analyzing textures in medical images. A two-dimensional fast time-frequency transform (FTFT-2D) for computing the local spectrum instantaneously and accurately is proposed. It can also be used to compute the complete redundant discrete ST-2D output, if needed. It reduces the storage requirement by generating a compressed form of the ST-2D. In addition, the memory efficient and adaptive nature of the FTFT-2D make it suitable for user-specific requirements.

An analysis of image texture, tumor location, and MGMT promoter methylation in glioblastoma using magnetic resonance imaging

In glioblastoma (GBM), promoter methylation of the DNA repair gene O(6)-methylguanine-DNA methyltransferase (MGMT) is associated with benefit from chemotherapy. Correlations between MGMT promoter methylation and visually assessed imaging features on magnetic resonance (MR) have been reported suggesting that noninvasive detection of MGMT methylation status might be possible. Our study assessed whether MGMT methylation status in GBM could be predicted using MR imaging. We conducted a retrospective analysis of MR images in patients with newly diagnosed GBM. Tumor texture was assessed by two methods. First, we analyzed texture by expert consensus describing the tumor borders, presence or absence of cysts, pattern of enhancement, and appearance of tumor signal in T2-weighted images. Then, we applied space-frequency texture analysis based on the S-transform. Tumor location within the brain was determined using automatized image registration and segmentation techniques. Their association with MGMT methylation was analyzed. We confirmed that ring enhancement assessed visually is significantly associated with unmethylated MGMT promoter status (P=0.006). Texture features on T2-weighted images assessed by the space-frequency analysis were significantly different between methylated and unmethylated cases (P<0.05). However, blinded classification of MGMT promoter methylation status reached an accuracy of only 71%. There were no significant differences in the locations of methylated and unmethylated GBM tumors. Our results provide further evidence that individual MR features are associated with MGMT methylation but better algorithms for predicting methylation status are needed. The relevance of this study lies on the application of novel techniques for the analysis of anatomical MR images of patients with GBM allowing the evaluation of subtleties not seen by an observer and facilitating the standardization of the methods, decreasing the potential for interobserver bias.