A simple model for glioma grading based on texture analysis applied to conventional brain MRI

MCE: Medical Cognition Embedded in 3D MRI feature extraction for advancing glioma staging

In recent years, various data-driven algorithms have been applied to the classification and staging of brain glioma MRI detection. However, the restricted availability of brain glioma MRI data in purely data-driven deep learning algorithms has presented challenges in extracting high-quality features and capturing their complex patterns. Moreover, the analysis methods designed for 2D data necessitate the selection of ideal tumor image slices, which does not align with practical clinical scenarios. Our research proposes an novel brain glioma staging model, Medical Cognition Embedded (MCE) model for 3D data. This model embeds knowledge characteristics into data-driven approaches to enhance the quality of feature extraction. Approach includes the following key components: (1) Deep feature extraction, drawing upon the imaging technical characteristics of different MRI sequences, has led to the design of two methods at both the algorithmic and strategic levels to mimic the learning process of real image interpretation by medical professionals during film reading; (2) We conduct an extensive Radiomics feature extraction, capturing relevant features such as texture, morphology, and grayscale distribution; (3) By referencing key points in radiological diagnosis, Radiomics feature experimental results, and the imaging characteristics of various MRI sequences, we manually create diagnostic features (Diag-Features). The efficacy of proposed methodology is rigorously evaluated on the publicly available BraTS2018 and BraTS2020 datasets. Comparing it to most well-known purely data-driven models, our method achieved higher accuracy, recall, and precision, reaching 96.14%, 93.4%, 97.06%, and 97.57%, 92.80%, 95.96%, respectively.

Radiomics Model for Predicting FOXP3 Expression Level and Survival in Clear Cell Renal Carcinoma

RATIONALE AND OBJECTIVES

We aimed to evaluate the predictive significance of forkhead box protein 3 (FOXP3) expression levels among individuals with clear cell renal carcinoma (ccRCC) and establish a radiomics model for predicting FOXP3 expression.

MATERIALS AND METHODS

430 patients with ccRCC were included in the gene-based prognostic analyses; 100 samples were used for radiomics feature generation, model development, and evaluation. A gradient boosting machine was employed to model the selected radiomics features. The developed model generated radiomics scores (RS) that predicted FOXP3 expression. The FOXP3 prognostic model combining imaging features was applied for survival and clinical indicator correlation analyses.

RESULTS

FOXP3 was highly expressed in patients with ccRCC and served as an independent predictive marker (hazard ratio [HR]=2.357, 95% CI [confidence interval]: 1.582-3.511, p < 0.001). The radiomics model formed by three radiomics characteristics was identified as a strong prognostic indicator of overall survival (OS). The predictive power of the model was commendable (areas under the curve: 0.835 and 0.809 for training and validation sets, respectively). Significant between-group variations in RS distribution were identified, as indicated by gene expression levels (p < 0.05). Disparities were observed in pathological stage, pharmaceutical therapy, and neoplasm status between low and high RS cohorts (p < 0.001). Kaplan-Meier curves revealed a significant correlation between increased RS and decreased OS (p = 0.001), which was also observed in the multivariate analyses (HR=3.411, 95% CI: 1.039-11.196, p = 0.043).

CONCLUSION

Prognostic outcome of ccRCC is closely linked to FOXP3 expression level. Computed tomography-based radiomics shows promise for prognostic prediction in ccRCC.

Fractal dimension: analyzing its potential as a neuroimaging biomarker for brain tumor diagnosis using machine learning

Purpose: The main purpose of this study was to comprehensively investigate the potential of fractal dimension (FD) measures in discriminating brain gliomas into low-grade glioma (LGG) and high-grade glioma (HGG) by examining tumor constituents and non-tumorous gray matter (GM) and white matter (WM) regions. Methods: Retrospective magnetic resonance imaging (MRI) data of 42 glioma patients (LGG, n = 27 and HGG, n = 15) were used in this study. Using MRI, we calculated different FD measures based on the general structure, boundary, and skeleton aspects of the tumorous and non-tumorous brain GM and WM regions. Texture features, namely, angular second moment, contrast, inverse difference moment, correlation, and entropy, were also measured in the tumorous and non-tumorous regions. The efficacy of FD features was assessed by comparing them with texture features. Statistical inference and machine learning approaches were used on the aforementioned measures to distinguish LGG and HGG patients. Results: FD measures from tumorous and non-tumorous regions were able to distinguish LGG and HGG patients. Among the 15 different FD measures, the general structure FD values of enhanced tumor regions yielded high accuracy (93%), sensitivity (97%), specificity (98%), and area under the receiver operating characteristic curve (AUC) score (98%). Non-tumorous GM skeleton FD values also yielded good accuracy (83.3%), sensitivity (100%), specificity (60%), and AUC score (80%) in classifying the tumor grades. These measures were also found to be significantly (p < 0.05) different between LGG and HGG patients. On the other hand, among the 25 texture features, enhanced tumor region features, namely, contrast, correlation, and entropy, revealed significant differences between LGG and HGG. In machine learning, the enhanced tumor region texture features yielded high accuracy, sensitivity, specificity, and AUC score. Conclusion: A comparison between texture and FD features revealed that FD analysis on different aspects of the tumorous and non-tumorous components not only distinguished LGG and HGG patients with high statistical significance and classification accuracy but also provided better insights into glioma grade classification. Therefore, FD features can serve as potential neuroimaging biomarkers for glioma.

Machine learning predicts histologic type and grade of canine gliomas based on MRI texture analysis

Conventional MRI features of canine gliomas subtypes and grades significantly overlap. Texture analysis (TA) quantifies image texture based on spatial arrangement of pixel intensities. Machine learning (ML) models based on MRI-TA demonstrate high accuracy in predicting brain tumor types and grades in human medicine. The aim of this retrospective, diagnostic accuracy study was to investigate the accuracy of ML-based MRI-TA in predicting canine gliomas histologic types and grades. Dogs with histopathological diagnosis of intracranial glioma and available brain MRI were included. Tumors were manually segmented across their entire volume in enhancing part, non-enhancing part, and peri-tumoral vasogenic edema in T2-weighted (T2w), T1-weighted (T1w), FLAIR, and T1w postcontrast sequences. Texture features were extracted and fed into three ML classifiers. Classifiers' performance was assessed using a leave-one-out cross-validation approach. Multiclass and binary models were built to predict histologic types (oligodendroglioma vs. astrocytoma vs. oligoastrocytoma) and grades (high vs. low), respectively. Thirty-eight dogs with a total of 40 masses were included. Machine learning classifiers had an average accuracy of 77% for discriminating tumor types and of 75.6% for predicting high-grade gliomas. The support vector machine classifier had an accuracy of up to 94% for predicting tumor types and up to 87% for predicting high-grade gliomas. The most discriminative texture features of tumor types and grades appeared related to the peri-tumoral edema in T1w images and to the non-enhancing part of the tumor in T2w images, respectively. In conclusion, ML-based MRI-TA has the potential to discriminate intracranial canine gliomas types and grades.

Bias and Class Imbalance in Oncologic Data-Towards Inclusive and Transferrable AI in Large Scale Oncology Data Sets

Recent technological developments have led to an increase in the size and types of data in the medical field derived from multiple platforms such as proteomic, genomic, imaging, and clinical data. Many machine learning models have been developed to support precision/personalized medicine initiatives such as computer-aided detection, diagnosis, prognosis, and treatment planning by using large-scale medical data. Bias and class imbalance represent two of the most pressing challenges for machine learning-based problems, particularly in medical (e.g., oncologic) data sets, due to the limitations in patient numbers, cost, privacy, and security of data sharing, and the complexity of generated data. Depending on the data set and the research question, the methods applied to address class imbalance problems can provide more effective, successful, and meaningful results. This review discusses the essential strategies for addressing and mitigating the class imbalance problems for different medical data types in the oncologic domain.

A Data-Driven Approach to Predicting 5-Aminolevulinic Acid-Induced Fluorescence and World Health Organization Grade in Newly Diagnosed Diffuse Gliomas

BACKGROUND

A growing body of evidence has revealed the potential utility of 5-aminolevulinic acid (5-ALA) as a surgical adjunct in selected lower-grade gliomas. However, a reliable means of identifying which lower-grade gliomas will fluoresce has not been established.

OBJECTIVE

To identify clinical and radiological factors predictive of intraoperative fluorescence in intermediate-grade gliomas. In addition, given that higher-grade gliomas are more likely to fluoresce than lower-grade gliomas, we also sought to develop a means of predicting glioma grade.

METHODS

We investigated a cohort of patients with grade II and grade III gliomas who received 5-ALA before resection at a single institution. Using a logistic regression-based model, we evaluated 14 clinical and molecular variables considered plausible determinants of fluorescence. We then distilled the most predictive features to develop a model for predicting both fluorescence and tumor grade. We also explored the relationship between intraoperative fluorescence and diagnostic molecular markers.

RESULTS

One hundered seventy-nine subjects were eligible for inclusion. Our logistic regression classifier accurately predicted intraoperative fluorescence in our cohort with 91.9% accuracy and revealed enhancement as the singular variable in determining intraoperative fluorescence. There was a direct relationship between enhancement on MRI and the likelihood of observed fluorescence. Observed fluorescence correlated with MIB-1 index but not with isocitrate dehydrogenase (IDH) status, 1p19q codeletion, or methylguanine DNA methyltransferase promoter methylation.

CONCLUSION

We demonstrate a strong correlation between enhancement on preoperative MRI and the likelihood of visible fluorescence during surgery in patients with intermediate-grade glioma. Our analysis provides a robust method for predicting 5-ALA-induced fluorescence in patients with grade II and grade III gliomas.

Non-invasive assessment of heterogeneity of gliomas using diffusion and perfusion MRI: correlation with spatially co-registered PET

BACKGROUND

Heterogeneity of gliomas challenges the neuronavigated biopsy and oncological therapy. Diffusion and perfusion magnetic resonance imaging (MRI) can reveal the cellular and hemodynamic heterogeneity of tumors. Integrated positron emission tomography (PET)/MRI is expected to be a non-invasive imaging approach to characterizing glioma.

PURPOSE

To evaluate the value of apparent diffusion coefficient (ADC), cerebral blood volume (CBV), and spatially co-registered maximal standard uptake value (SUV_max) for tissue characterization and glioma grading.

MATERIAL AND METHODS

Thirty-seven consecutive patients with pathologically confirmed gliomas were retrospectively investigated. The relative minimum ADC (rADC_min), relative maximal ADC (rADC_max), relative maximal rCBV (rCBV_max), the relative minimum rCBV (rCBV_min), and the corresponding relative SUVmax (rSUV_max) were measured. The paired t-test was used to compare the quantitative parameters between different regions to clarify tumor heterogeneity. Imaging parameters between WHO grade IV and grade II/III gliomas were compared by t-test. The diagnostic efficiency of multiparametric PET/MRI was analyzed by receiver operating characteristic (ROC) curve.

RESULTS

The values of rSUV_max were significantly different between maximal diffusion/perfusion area and minimum diffusion/perfusion area (P < 0.001/P < 0.001) within tumor. The values of rADC_min (P < 0.001), rCBV_max (P = 0.002), and corresponding rSUV_max (P = 0.001/P < 0.001) could be used for grading gliomas. The areas under the ROC curves of rSUV_max defined by rADC_min and rCBV_max were 0.89 and 0.91, respectively.

CONCLUSION

Diffusion and perfusion MRI can detect glioma heterogeneity with excellent molecular imaging correlations. Regions with rCBV_max suggest tissues with the highest metabolism and malignancy for guiding glioma grading and tissue sampling.

Radiomics features based on T2-weighted fluid-attenuated inversion recovery MRI predict the expression levels of CD44 and CD133 in lower-grade gliomas

Objective: To verify the association between CD44 and CD133 expression levels and the prognosis of patients with lower-grade gliomas (LGGs) and constructing radiomic models to predict those two genes' expression levels before surgery. Materials & methods: Genomic data of patients with LGG and the corresponding T2-weighted fluid-attenuated inversion recovery images were downloaded from the Cancer Genome Atlas and the Cancer Imaging Archive, which were utilized for prognosis analysis, radiomic feature extraction and model construction, respectively. Results & conclusion: CD44 and CD133 expression levels in LGG can significantly affect the prognosis of patients with LGG. Based on the T2-weighted fluid-attenuated inversion recovery images, the radiomic features can effectively predict the expression levels of CD44 and CD133 before surgery.

Predictive Ki-67 Proliferation Index of Cervical Squamous Cell Carcinoma Based on IVIM-DWI Combined with Texture Features

Purpose

This study aims to determine whether IVIM-DWI combined with texture features based on preoperative IVIM-DWI could be used to predict the Ki-67 PI, which is a widely used cell proliferation biomarker in CSCC.

Methods

A total of 70 patients were included. Among these patients, 16 patients were divided into the Ki-67 PI <50% group and 54 patients were divided into the Ki-67 PI ≥50% group based on the retrospective surgical evaluation. All patients were examined using a 3.0T MRI unit with one standard protocol, including an IVIM-DWI sequence with 10 b values (0–1,500 sec/mm²). The maximum level of CSCC with a b value of 800 sec/mm² was selected. The parameters (diffusion coefficient (D), microvascular volume fraction (f), and pseudodiffusion coefficient (D^∗)) were calculated with the ADW 4.6 workstation, and the texture features based on IVIM-DWI were measured using GE AK quantitative texture analysis software. The texture features included the first order, GLCM, GLSZM, GLRLM, and wavelet transform features. The differences in IVIM-DWI parameters and texture features between the two groups were compared, and the ROC curve was performed for parameters with group differences, and in combination.

Results

The D value in the Ki-67 PI ≥50% group was lower than that in the Ki-67 PI <50% group (P < 0.05). A total of 1,050 texture features were obtained using AK software. Through univariate logistic regression, mPMR feature selection, and multivariate logistic regression, three texture features were obtained: wavelet_HHL_GLRLM_ LRHGLE, lbp_3D_k_ firstorder_IR, and wavelet_HLH_GLCM_IMC1. The AUC of the prediction model based on the three texture features was 0.816, and the combined D value and three texture features was 0.834.

Conclusions

Texture analysis on IVIM-DWI and its parameters was helpful for predicting Ki-67 PI and may provide a noninvasive method to investigate important imaging biomarkers for CSCC.

Differentiating glioblastoma multiforme from cerebral lymphoma: application of advanced texture analysis of quantitative apparent diffusion coefficients

PURPOSE

The purpose of this study was to differentiate glioblastoma multiforme from primary central nervous system lymphoma using the customised first and second-order histogram features derived from apparent diffusion coefficients.Methods and materials: A total of 82 patients (57 with glioblastoma multiforme and 25 with primary central nervous system lymphoma) were included in this study. The axial T1 post-contrast and fluid-attenuated inversion recovery magnetic resonance images were used to delineate regions of interest for the tumour and peritumoral oedema. The regions of interest were then co-registered with the apparent diffusion coefficient maps, and the first and second-order histogram features were extracted and compared between glioblastoma multiforme and primary central nervous system lymphoma groups. Receiver operating characteristic curve analysis was performed to calculate a cut-off value and its sensitivity and specificity to differentiate glioblastoma multiforme from primary central nervous system lymphoma.

RESULTS

Based on the tumour regions of interest, apparent diffusion coefficient mean, maximum, median, uniformity and entropy were higher in the glioblastoma multiforme group than the primary central nervous system lymphoma group (P ≤ 0.001). The most sensitive first and second-order histogram feature to differentiate glioblastoma multiforme from primary central nervous system lymphoma was the maximum of 2.026 or less (95% confidence interval (CI) 75.1-99.9%), and the most specific first and second-order histogram feature was smoothness of 1.28 or greater (84.0% CI 70.9-92.8%). Based on the oedema regions of interest, most of the first and second-order histogram features were higher in the glioblastoma multiforme group compared to the primary central nervous system lymphoma group (P ≤ 0.015). The most sensitive first and second-order histogram feature to differentiate glioblastoma multiforme from primary central nervous system lymphoma was the 25th percentile of 0.675 or less (100% CI 83.2-100%) and the most specific first and second-order histogram feature was the median of 1.28 or less (85.9% CI 66.3-95.8%).

CONCLUSIONS

Texture analysis using first and second-order histogram features derived from apparent diffusion coefficient maps may be helpful in differentiating glioblastoma multiforme from primary central nervous system lymphoma.