A radiomics-based study for differentiating parasellar cavernous hemangiomas from meningiomas

Abstract

To investigate the value of the radiomic models for differentiating parasellar cavernous hemangiomas from meningiomas and to compare the classification performance with different MR sequences and classifiers. A total of 96 patients with parasellar tumors (40 cavernous hemangiomas and 56 meningiomas) were enrolled in this retrospective multiple-center study. Univariate and multivariate analyses were performed to identify the clinical factors and semantic features of MRI scans. Radiomics features were extracted from five MRI sequences using radiomics software. Three feature selection methods and six classifiers were evaluated in the training cohort to construct favorable radiomic machine-learning classifiers. The performance of different classifiers was evaluated using the AUC and compared to neuroradiologists. The detection rates of T1WI, T2WI, and CE-T1WI for parasellar cavernous hemangiomas and meningiomas were approximately 100%. In contrast, the ADC maps had the detection rate of 18/22 and 19/25, respectively, (AUC, 0.881) with 2.25 cm as the critical value diameter. Radiomics models with the SVM and KNN classifiers based on T2WI and ADC maps had favorable predictive performances (AUC > 0.90 and F-score value > 0.80). These models outperformed MRI model (AUC 0.805) and neuroradiologists (AUC, 0.756 and 0.545, respectively). Radiomic models based on T2WI and ADC and combined with SVM and KNN classifiers have the potential to be a viable method for differentiating parasellar hemangiomas from meningiomas. T2WI is more universally applicable than ADC values due to its higher detection rate for parasellar tumors.

Rare case of tentorial cavernous malformation mimicking a meningioma: illustrative case

“…However, Wang et al recently reported that the application of a radiomics-based algorithm using diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) facilitates the distinction of meningiomas from cavernous malformations. 19 Future automated radiomics may assist in the preoperative distinction, which may have implications for treatment evaluation, such as additional angiographic imaging for larger lesions. In single-photon emission computed tomography (SPECT), cavernous malformations are usually low in uptake, whereas in other tumors such as meningiomas, the metabolic uptake may be higher.…”

Dual-Level Augmentation Radiomics Analysis for Multisequence MRI Meningioma Grading

“…From prior meningioma grading studies, it is known that multilevel feature selection can give better results. Fifteen feature selection methods were selected based on previous related research [13,31,[33][34][35][36], including the filter methods Chi-square (CHSQ), t-test (TSQ), Kruskal-Wallis H-test tests (KWH), variance (VAR), relief (RELF), mutual information (MI), minimum redundancy maximum relevance ensemble (mRMRe) and the embedded methods L1-based logistic regression (L1-LR), elastic net (EN), least absolute shrinkage and selection operator (LASSO), L1-based linear support vector machine (L1-SVM), random forest (RF), extra tree ensemble (ETE), gradient boosting decision tree (GBDT), and xgboost (XGB). In this study, the filter methods were used as the first level of screening to reduce the number of features, and the embedded methods were used as the second level of screening to obtain the final features.…”

A Comprehensive Review of the Impact of Machine Learning and Omics on Rare Neurological Diseases

“…The field of radiomics is another area where AI is making significant strides. By extracting quantitative features from radiographic images, machine learning algorithms are enhancing the diagnosis of rare neurological diseases and tumors [61][62][63]. These models not only compete with but, in some cases, outperform human experts in diagnosing conditions like high-grade gliomas [64,65].…”