Meningioma grading using conventional MRI histogram analysis based on 3D tumor measurement

Development of a nomogram based on whole-tumor multiparametric MRI histogram analysis to predict deep myometrial invasion in stage I endometrioid endometrial carcinoma preoperatively

BACKGROUND

The depth of myometrial invasion determines whether International Federation of Gynecology and Obstetrics stage I endometrioid endometrial carcinoma (EEC) patients undergo lymph node dissection. However, subjective evaluation results relying on magnetic resonance imaging (MRI) are not always satisfactory.

PURPOSE

To develop a nomogram based on whole-volume tumor MRI histogram parameters to preoperatively predict deep myometrial invasion (DMI) in patients with stage I EEC.

MATERIAL AND METHODS

This retrospective analysis included 131 EEC patients and a training/validation cohort of 92/39 patients at a 7:3 ratio. The histogram parameters were obtained from multiple sequences (ADC mapping and T2-weighted imaging) within volumes of interest. Univariate analysis, least absolute shrinkage and selection operator (LASSO) regression, and multivariate logistic regression were used for feature selection. The performance of clinical model, histogram model, and histogram nomogram was evaluated by calculating the area under the receiver operating characteristic curve (AUC).

RESULTS

Age and two morphological features (maximum anteroposterior tumor diameter on sagittal T2-weighted images [APsag] and the tumor area ratio [TAR]) were selected to construct the clinical model. Five histogram parameters were selected for the creation of the histogram model. The nomogram, which combines the histogram parameters, age, APsag, and TAR, achieved the highest AUCs in both the training and validation cohorts (nomogram vs. histogram vs. clinical model: 0.973 vs. 0.871 vs. 0.934 [training] and 0.972 vs. 0.870 vs. 0.928 [validation]).

CONCLUSION

The MR histogram nomogram can help predict the DMI of patients with stage I EEC preoperatively, assisting physicians in the development of personalized treatment strategies.

A large scale multi institutional study for radiomics driven machine learning for meningioma grading

This study aims to develop and evaluate radiomics-based machine learning (ML) models for predicting meningioma grades using multiparametric magnetic resonance imaging (MRI). The study utilized the BraTS-MEN dataset's training split, including 698 patients (524 with grade 1 and 174 with grade 2-3 meningiomas). We extracted 4872 radiomic features from T1, T1 with contrast, T2, and FLAIR MRI sequences using PyRadiomics. LASSO regression reduced features to 176. The data was split into training (60%), validation (20%), and test (20%) sets. Five ML algorithms (TabPFN, XGBoost, LightGBM, CatBoost, and Random Forest) were employed to build models differentiating low-grade (grade 1) from high-grade (grade 2-3) meningiomas. Hyperparameter tuning was performed using Optuna, optimizing model-specific parameters and feature selection. The CatBoost model demonstrated the best performance, achieving an area under the receiver operating characteristic curve (AUROC) of 0.838 [95% confidence interval (CI): 0.689-0.935], precision of 0.492 (95% CI: 0.371-0.623), recall of 0.838 (95% CI: 0.689-0.935), F1 score of 0.620 (95% CI: 0.495-0.722), accuracy of 0.729 (95% CI: 0.650-0.800), an area under the precision-recall curve (AUPRC) of 0.620 (95% CI: 0.433-0.753), and Brier score of 0.156 (95% CI: 0.122-0.200). Other models showed comparable performance, with mean AUROCs ranging from 0.752 to 0.784. The radiomics-based ML approach presented in this study showcases the potential for non-invasive and pre-operative grading of meningiomas using multiparametric MRI. Further validation on larger and independent datasets is necessary to establish the robustness and generalizability of these findings.

T1 Pre- and Post-contrast Delta Histogram Parameters in Predicting the Grade of Meningioma and Their Relationship to Ki-67 Proliferation Index

RATIONALE AND OBJECTIVES

To explore the feasibility of delta histogram parameters (including absolute delta histogram parameters (AdHP) and relative delta histogram parameters (RdHP)) in predicting the grade of meningioma and to further investigate whether delta histogram parameters correlate with the Ki-67 proliferation index.

METHODS

92 patients with meningioma who underwent MRI examination (including T1-weighted (T1) and contrast-enhanced T1-weighted images (T1C)) were enrolled in this retrospective study. A total of 46 low-grade cases formed the low-grade group (grade 1, LGM), and a total of 46 high-grade cases formed the high-grade group (38 grade 2, 8 grade 3, HGM). Histogram parameters (HP) of T1 and T1C were extracted. Subsequently, morphological MRI features, AdHP (AdHP=T1CHP-T1HP), and RdHP (RdHP=(T1CHP-T1HP)/T1HP) were recorded and compared, respectively. Binary logistic regression analysis was used to obtain combined performance of the significant parameters. Diagnostic performance was identified by ROC. Spearman's correlation coefficients were taken to assess the relationship between delta histogram parameters and the Ki-67 proliferation index.

RESULTS

In morphological MRI features, HGM is more prone to lobulation and necrosis/cystic changes (all p < 0.05). In delta histogram parameters, HGM exhibits higher mean, Perc.01, Perc.25, Perc.50, Perc.75, Perc.99, SD, and variance of AdHP, maximum, mean, Perc.25, Perc.50, Perc.75, and Perc.99 of RdHP, compared to LGM (all p < 0.00357). The optimal predictive performance was obtained by combining morphological MRI features and delta histogram parameters with an AUC of 0.945. Significant correlations were observed between significant delta histogram parameters and the Ki-67 proliferation index (all p < 0.05).

CONCLUSION

Delta histogram parameter is a promising potential biomarker, which may be helpful in noninvasive predicting the grade and proliferative activity of meningioma.

Research Progress of Artificial Intelligence in the Grading and Classification of Meningiomas

A meningioma is a common primary central nervous system tumor. The histological features of meningiomas vary significantly depending on the grade and subtype, leading to differences in treatment and prognosis. Therefore, early diagnosis, grading, and typing of meningiomas are crucial for developing comprehensive and individualized diagnosis and treatment plans. The advancement of artificial intelligence (AI) in medical imaging, particularly radiomics and deep learning (DL), has contributed to the increasing research on meningioma grading and classification. These techniques are fast and accurate, involve fully automated learning, are non-invasive and objective, enable the efficient and non-invasive prediction of meningioma grades and classifications, and provide valuable assistance in clinical treatment and prognosis. This article provides a summary and analysis of the research progress in radiomics and DL for meningioma grading and classification. It also highlights the existing research findings, limitations, and suggestions for future improvement, aiming to facilitate the future application of AI in the diagnosis and treatment of meningioma.

Deep learning and radiomics-based approach to meningioma grading: exploring the potential value of peritumoral edema regions

Objective.To address the challenge of meningioma grading, this study aims to investigate the potential value of peritumoral edema (PTE) regions and proposes a unique approach that integrates radiomics and deep learning techniques.Approach.The primary focus is on developing a transfer learning-based meningioma feature extraction model (MFEM) that leverages both vision transformer (ViT) and convolutional neural network (CNN) architectures. Additionally, the study explores the significance of the PTE region in enhancing the grading process.Main results.The proposed method demonstrates excellent grading accuracy and robustness on a dataset of 98 meningioma patients. It achieves an accuracy of 92.86%, precision of 93.44%, sensitivity of 95%, and specificity of 89.47%.Significance.This study provides valuable insights into preoperative meningioma grading by introducing an innovative method that combines radiomics and deep learning techniques. The approach not only enhances accuracy but also reduces observer subjectivity, thereby contributing to improved clinical decision-making processes.

The experience of living with malignant meningioma

OBJECTIVES

Meningiomas are the most common, primary intracranial tumor and most are benign. Little is known of the rare patient group living with a malignant meningioma, comprising 1-3% of all meningiomas. Our aim was to explore how patients perceived quality of daily life after a malignant meningioma diagnosis.

METHODS

This qualitative explorative study was composed of individual semi-structured interviews. Eligible patients (n = 12) were selected based on ability to participate in an interview, from a background population of 23 patients diagnosed with malignant meningioma at Rigshospitalet from 2000 to 2021. We performed an inductive thematic analysis following Braun and Clarke's guidelines.

RESULTS

Eight patients were interviewed. The analysis revealed 4 overarching themes: (1) perceived illness and cause of symptoms, (2) identity, roles, and interaction, (3) threat and uncertainty of the future, and (4) belief in authority. The perceived quality of daily life is negatively impacted by the disease. Patients experience a shift in self-concept and close interactions, and some struggle with accepting a new everyday life. Patients have a high risk of discordant prognostic awareness in relation to health-care professionals.

SIGNIFICANCE OF RESULTS

We provide a much-needed patient-centered perspective of living with malignant meningioma: quality of life was affected by perception of threat and an uncertainty of the future. Perception of illness and the interpretation of the cause of symptoms varied between subjects, but a common trait was that patients' identity, roles, and interactions were affected. Shared decision-making and a strengthened continuity during follow-up could aid this rare patient group.

Whole-tumor histogram analysis of postcontrast T1-weighted and apparent diffusion coefficient in predicting the grade and proliferative activity of adult intracranial ependymomas

PURPOSE

To investigate the value of histogram analysis of postcontrast T1-weighted (T1C) and apparent diffusion coefficient (ADC) images in predicting the grade and proliferative activity of adult intracranial ependymomas.

METHODS

Forty-seven adult intracranial ependymomas were enrolled and underwent histogram parameters extraction (including minimum, maximum, mean, 1st percentile (Perc.01), Perc.05, Perc.10, Perc.25, Perc.50, Perc.75, Perc.90, Perc.95, Perc.99, standard deviation (SD), variance, coefficient of variation (CV), skewness, kurtosis, and entropy of T1C and ADC) using FireVoxel software. Differences in histogram parameters between grade 2 and grade 3 adult intracranial ependymomas were compared. Receiver operating characteristic curves and logistic regression analyses were conducted to evaluate the diagnostic performance. Spearman's correlation analysis was used to evaluate the relationship between histogram parameters and Ki-67 proliferation index.

RESULTS

Grade 3 intracranial ependymomas group showed significantly higher Perc.95, Perc.99, SD, variance, CV, and entropy of T1C; lower minimum, mean, Perc.01, Perc.05, Perc.10, Perc.25, Perc.50 of ADC; and higher CV and entropy of ADC than grade 2 intracranial ependymomas group (all p < 0.05). Entropy (T1C) and Perc.10 (ADC) had a higher diagnostic performance with AUCs of 0.805 and 0.827 among the histogram parameters of T1C and ADC, respectively. The diagnostic performance was improved by combining entropy (T1C) and Perc.10 (ADC), with an AUC of 0.857. Significant correlations were observed between significant histogram parameters of T1C (r = 0.296-0.417, p = 0.001-0.044) and ADC (r = -0.428-0.395, p = 0.003-0.038).

CONCLUSION

Whole-tumor histogram analysis of T1C and ADC may be a promising approach for predicting the grade and proliferative activity of adult intracranial ependymomas.

Apparent diffusion coefficient histogram analysis for differentiating fibroblastic meningiomas from non-fibroblastic WHO grade 1 meningiomas

PURPOSE

To investigate the role of apparent diffusion coefficient (ADC) histogram analysis in differentiating fibroblastic meningiomas (FM) from non-fibroblastic WHO grade 1 meningiomas (nFM).

METHODS

This retrospective study analyzed the histopathological and diagnostic imaging data of 220 patients with histopathologically confirmed FM and nFM. The whole tumors were delineated on axial ADC images, and histogram parameters (mean, variance, skewness, kurtosis, as well as the 1st, 10th, 50th, 90th, and 99th percentile ADC [ADCp1, ADCp10, ADCp50, ADCp90, and ADCp99, respectively]) were obtained. Multivariate logistic regression analysis was used to identify the most valuable variables for discriminating FM from nFM WHO grade 1 meningiomas, and their diagnostic efficacy in differentiating FM from nFM before surgery was assessed using receiver operating characteristic (ROC) curves.

RESULTS

The mean, variance, ADCp50, ADCp90, and ADCp99 of the FM group were all lower than those of the nFM group (P < 0.05), there was significant difference in location and sex (P < 0.05). Multivariate logistic regression showed ADCp99 (P < 0.001) and location (P = 0.007) were the most valuable parameters in the discrimination of FM and nFM WHO grade 1 meningiomas. The diagnostic efficacy was achieved an AUC of 0.817(95% CI, 0.759-0.866), the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were 66.4%, 83.6%, 75.0%, 80.2%, and 71.3%, respectively.

CONCLUSION

ADC histogram analysis is helpful in noninvasive differentiation of FM and nFM WHO grade 1 meningiomas, and combined ADCp99 and location have the best diagnostic efficacy.

Differential diagnosis of atypical and anaplastic meningiomas based on conventional MRI features and ADC histogram parameters using a logistic regression model nomogram

The purpose of the study was to determine the value of a logistic regression model nomogram based on conventional magnetic resonance imaging (MRI) features and apparent diffusion coefficient (ADC) histogram parameters in differentiating atypical meningioma (AtM) from anaplastic meningioma (AnM). Clinical and imaging data of 34 AtM and 21 AnM diagnosed by histopathology were retrospectively analyzed. The whole tumor delineation along the tumor edge on ADC images and ADC histogram parameters were automatically generated and comparisons between the two groups using the independent samples t test or Mann-Whitney U test. Univariate and multivariate logistic regression analyses were used to construct the nomogram of the AtM and AnM prediction model, and the model's predictive efficacy was evaluated using calibration and decision curves. Significant differences in the mean, enhancement, perc.01%, and edema were noted between the AtM and AnM groups (P < 0.05). Age, sex, location, necrosis, shape, max-D, variance, skewness, kurtosis, perc.10%, perc.50%, perc.90%, and perc.99% exhibited no significant differences (P > 0.05). The mean and enhancement were independent risk factors for distinguishing AtM from AnM. The area under the curve, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the nomogram were 0.871 (0.753-0.946), 80.0%, 81.0%, 79.4%, 70.8%, and 87.1%, respectively. The calibration curve demonstrated that the model's probability to predict AtM and AnM was in favorable agreement with the actual probability, and the decision curve revealed that the prediction model possessed satisfactory clinical availability. A logistic regression model nomogram based on conventional MRI features and ADC histogram parameters is potentially useful as an auxiliary tool for the preoperative differential diagnosis of AtM and AnM.

Whole-tumor histogram analysis of multi-parametric MRI for differentiating brain metastases histological subtypes in lung cancers: relationship with the Ki-67 proliferation index

This study aims to investigate the predictive value of preoperative whole-tumor histogram analysis of multi-parametric MRI for histological subtypes in patients with lung cancer brain metastases (BMs) and explore the correlation between histogram parameters and Ki-67 proliferation index. The preoperative MRI data of 95 lung cancer BM lesions obtained from 73 patients (42 men and 31 women) were retrospectively analyzed. Multi-parametric MRI histogram was used to distinguish small-cell lung cancer (SCLC) from non-small cell lung cancer (NSCLC), and adenocarcinoma (AC) from squamous cell carcinoma (SCC), respectively. The T1-weighted contrast-enhanced (T1C) and apparent diffusion coefficient (ADC) histogram parameters of the volumes of interest (VOIs) in all BMs lesions were extracted using FireVoxel software. The following histogram parameters were obtained: maximum, minimum, mean, standard deviation (SD), variance, coefficient of variation (CV), skewness, kurtosis, entropy, and 1st-99th percentiles. Then investigated their relationship with the Ki-67 proliferation index. The skewness-T1C, kurtosis-T1C, minimum-ADC, mean-ADC, CV-ADC and 1st - 90th ADC percentiles were significantly different between the SCLC and NSCLC groups (all p < 0.05). When the 10th-ADC percentile was 668, the sensitivity, specificity, and accuracy (90.80%, 76.70% and 86.32%, respectively) for distinguishing SCLC from NSCLC reached their maximum values, with an AUC of 0.895 (0.824 - 0.966). Mean-T1C, CV-T1C, skewness-T1C, 1st - 50th T1C percentiles, maximum-ADC, SD-ADC, variance-ADC and 75th - 99th ADC percentiles were significantly different between the AC and SCC groups (all p < 0.05). When the CV-T1C percentiles was 3.13, the sensitivity, specificity and accuracy (75.00%, 75.60% and 75.38%, respectively) for distinguishing AC and SCC reached their maximum values, with an AUC of 0.829 (0.728-0.929). The 5th-ADC and 10th-ADC percentiles were strongly correlated with the Ki-67 proliferation index in BMs. Multi-parametric MRI histogram parameters can be used to identify the histological subtypes of lung cancer BMs and predict the Ki-67 proliferation index.

A radiomics nomogram for predicting the meningioma grade based on enhanced T1WI images

OBJECTIVES

The objective of this study was to develop a radiomics nomogram for predicting the meningioma grade based on enhanced T1 weighted imaging (T1WI) images.

METHODS

188 patients with meningioma were analyzed retrospectively. There were 94 high-grade meningioma to form high-grade group and 94 low-grade meningioma were selected randomly to form low-grade group. Clinical data and MRI features were analyzed and compared. The clinical model was built by using the significant variables. The least absolute shrinkage and selection operator regression was used to select the most valuable radiomics feature. The radiomics signature was built and the Rad-score was calculated. The radiomics nomogram was developed by the significant variables of the clinical factors and Rad-score. The calibration curve and the Hosmer-Lemeshow test were used to evaluate the radiomics nomogram. Different models were compared by Delong test and decision curve analysis curve.

RESULTS

The sex, size and surrounding invasion were used to build clinical model. The area under the receiver operator characteristic curve (AUC) of clinical model was 0.870 (95% CI: 0.782-0.959). Nine features were used to construct the radiomics signature. The AUC of the radiomics signature was 0.885 (95% CI: 0.802-0.968). The AUC of radiomics nomogram was 0.952 (95% CI: 0.904-1). The AUC of radiomics nomogram was higher than that of clinical model and radiomics signature with a significant difference (p＜0.05). The decision curve analysis curve showed that the radiomics nomogram had a larger net benefit than the clinical model and radiomics signature.

CONCLUSION

The radiomics nomogram based on enhanced T1 weighted imaging images for predicting the meningioma grade showed high predictive value and might contribute to the diagnosis and treatment of meningioma.

ADVANCES IN KNOWLEDGE

1. We first constructed a radiomic nomogram to predict the meningioma grade.2. We compared the results of the clinical model, radiomics signature and radiomics nomogram.

T1 and ADC histogram parameters may be an in vivo biomarker for predicting the grade, subtype, and proliferative activity of meningioma

OBJECTIVE

To investigate the value of histogram analysis of T1 mapping and diffusion-weighted imaging (DWI) in predicting the grade, subtype, and proliferative activity of meningioma.

METHODS

This prospective study comprised 69 meningioma patients who underwent preoperative MRI including T1 mapping and DWI. The histogram metrics, including mean, median, maximum, minimum, 10th percentiles (C10), 90th percentiles (C90), kurtosis, skewness, and variance, of T1 and apparent diffusion coefficient (ADC) values were extracted from the whole tumour and peritumoural oedema using FeAture Explorer. The Mann-Whitney U test was used for comparison between low- and high-grade tumours. Receiver operating characteristic (ROC) curve and logistic regression analyses were performed to identify the differential diagnostic performance. The Kruskal-Wallis test was used to further classify meningioma subtypes. Spearman's rank correlation coefficients were calculated to analyse the correlations between histogram parameters and Ki-67 expression.

RESULTS

High-grade meningiomas showed significantly higher mean, maximum, C90, and variance of T1 (p = 0.001-0.009), lower minimum, and C10 of ADC (p = 0.013-0.028), compared to low-grade meningiomas. For all histogram parameters, the highest individual distinctive power was T1 C90 with an AUC of 0.805. The best diagnostic accuracy was obtained by combining the T1 C90 and ADC C10 with an AUC of 0.864. The histogram parameters differentiated 4/6 pairs of subtype pairs. Significant correlations were identified between Ki-67 and histogram parameters of T1 (C90, mean) and ADC (C10, kurtosis, variance).

CONCLUSION

T1 and ADC histogram parameters may represent an in vivo biomarker for predicting the grade, subtype, and proliferative activity of meningioma.

KEY POINTS

• The histogram parameter based on T1 mapping and DWI is useful to preoperatively evaluate the grade, subtype, and proliferative activity of meningioma. • The combination of T1 C90 and ADC C10 showed the best performance for differentiating low- and high-grade meningiomas.

Deep learning-based automatic segmentation of meningioma from multiparametric MRI for preoperative meningioma differentiation using radiomic features: a multicentre study

OBJECTIVES

Develop and evaluate a deep learning-based automatic meningioma segmentation method for preoperative meningioma differentiation using radiomic features.

METHODS

A retrospective multicentre inclusion of MR examinations (T1/T2-weighted and contrast-enhanced T1-weighted imaging) was conducted. Data from centre 1 were allocated to training (n = 307, age = 50.94 ± 11.51) and internal testing (n = 238, age = 50.70 ± 12.72) cohorts, and data from centre 2 external testing cohort (n = 64, age = 48.45 ± 13.59). A modified attention U-Net was trained for meningioma segmentation. Segmentation accuracy was evaluated by five quantitative metrics. The agreement between radiomic features from manual and automatic segmentations was assessed using intra class correlation coefficient (ICC). After univariate and minimum-redundancy-maximum-relevance feature selection, L1-regularized logistic regression models for differentiating between low-grade (I) and high-grade (II and III) meningiomas were separately constructed using manual and automatic segmentations; their performances were evaluated using ROC analysis.

RESULTS

Dice of meningioma segmentation for the internal testing cohort were 0.94 ± 0.04 and 0.91 ± 0.05 for tumour volumes in contrast-enhanced T1-weighted and T2-weighted images, respectively; those for the external testing cohort were 0.90 ± 0.07 and 0.88 ± 0.07. Features extracted using manual and automatic segmentations agreed well, for both the internal (ICC = 0.94, interquartile range: 0.88-0.97) and external (ICC = 0.90, interquartile range: 0.78-70.96) testing cohorts. AUC of radiomic model with automatic segmentation was comparable with that of the model with manual segmentation for both the internal (0.95 vs. 0.93, p = 0.176) and external (0.88 vs. 0.91, p = 0.419) testing cohorts.

CONCLUSIONS

The developed deep learning-based segmentation method enables automatic and accurate extraction of meningioma from multiparametric MR images and can help deploy radiomics for preoperative meningioma differentiation in clinical practice.

KEY POINTS

• A deep learning-based method was developed for automatic segmentation of meningioma from multiparametric MR images. • The automatic segmentation method enabled accurate extraction of meningiomas and yielded radiomic features that were highly consistent with those that were obtained using manual segmentation. • High-grade meningiomas were preoperatively differentiated from low-grade meningiomas using a radiomic model constructed on features from automatic segmentation.

Grading of meningioma tumors based on analyzing tumor volumetric histograms obtained from conventional MRI and apparent diffusion coefficient images

Background

Our purpose was to evaluate the application of volumetric histogram parameters obtained from conventional MRI and apparent diffusion coefficient (ADC) images for grading the meningioma tumors.

Results

Tumor volumetric histograms of preoperative MRI images from 45 patients with the diagnosis of meningioma at different grades were analyzed to find the histogram parameters. Kruskal-Wallis statistical test was used for comparison between the parameters obtained from different grades. Multi-parametric regression analysis was used to find the model and parameters with high predictive value for the classification of meningioma. Mode; standard deviation on post-contrast T1WI, T2-FLAIR, and ADC images; kurtosis on post-contrast T1WI and T2-FLAIR images; mean and several percentile values on ADC; and post-contrast T1WI images showed significant differences among different tumor grades (P < 0.05). The multi-parametric linear regression showed that the ADC histogram parameters model had a higher predictive value, with cutoff values of 0.212 (sensitivity = 79.6%, specificity = 84.3%) and 0.180 (sensitivity = 70.9%, specificity = 80.8%) for differentiating the grade I from II, and grade II from III, respectively.

Conclusions

The multi-parametric model of volumetric histogram parameters in some of the conventional MRI series (i.e., post-contrast T1WI and T2-FLAIR images) along with the ADC images are appropriate for predicting the meningioma tumors’ grade.

Magnetic Resonance Imaging (MRI) Differential Diagnosis of Meningiomas Using ANOVA

This work explored the diagnostic value of different subtypes of meningiomas under T2WI low signal based on analysis of variance (ANOVA), and the expression differences of Ki67, VEGF, and P73 in different subtypes were analyzed. 67 patients with meningioma confirmed surgically and pathologically in hospital were selected as the research subjects, whose pathological classification occurs with obvious low signal on T2WI. First, the age distribution of the subjects and the distribution of different subtypes were counted. Then, ANOVA was adopted to analyze the MRI imaging signs of patients with different subtypes of meningioma. Finally, the differences of Ki67, VEGF, and P73 proteins and mRNA expression levels in different subtypes were detected via immunohistochemical assay and qPCR. The results showed that the proportion of patients with transitional meningioma was the most, which was 43.28%, while the proportion of patients with meningeal melanoma was the least, which was 7.46%. In patients with transitional meningioma, the MRI images showed mixed signals in different layers. Fibrous MRI images showed hyalinosis and calcification of collagen fibers in the tumor, with low T2WI signal. Sand-shape MRI images showed double low signals. MRI images of meningeal melanoma showed high signal on T1-weighted Imaging (T1WI) and low signal on T2WI. The protein expression and mRNA levels of Ki67 and P73 in transitional meningioma were evidently higher in contrast to those in fibrous meningioma (P < 0.05). The expression level of VEGF protein and mRNA in meningeal melanoma were notably higher in contrast to those in fibro meningioma (P < 0.05). It was revealed that the MRI images of the four subtypes of meningiomas under ANOVA-based T2WI low signal were quite different, and the expressions of Ki67, P73, and VEGF in different subtypes had significant differences. This work provided a reference basis for the preoperative diagnosis, treatment, and prognosis of meningiomas.

Value of MRI Radiomics Based on Enhanced T1WI Images in Prediction of Meningiomas Grade

OBJECTIVE

Different grades of meningiomas require different treatment strategies and have a different prognosis; thus, the noninvasive classification of meningiomas before surgery is of great importance. The purpose of this study was to explore the application value of magnetic resonance imaging (MRI) radiomics based on enhanced-T1-weighted (T1WI) images in the prediction of meningiomas grade.

MATERIALS AND METHODS

A total of 98 patients with meningiomas who were confirmed by surgical pathology and underwent preoperative routine MRI between January 2017 and December 2019 were analyzed. There were 82 cases of low-grade meningiomas (WHO grade I) and 16 cases of high-grade meningiomas (7 cases of WHO grade II and 9 cases of WHO grade III). These patients were randomly divided into a training group and test group according to 7:3 ratio. The lesions were manually delineated using ITK-SNAP software, and radiomics analysis were performed using the Analysis Kit (AK) software. A total of 396 tumor texture features were extracted. Subsequently, the LASSO algorithm was used to reduce the feature dimensions. Next, a prediction model was constructed using the Logistic Regression method and receiver operator characteristic was used to evaluate the prediction performance of the model.

RESULTS

A radiomics prediction model was constructed based on the selected nine characteristic parameters, which performed well in predicting the meningiomas grade. The accuracy rates in the training group and the test group were respectively 94.3% and 92.9%, the sensitivities were respectively 94.8%, and 91.7%, the specificities were respectively 91.7% and 100%, and the area under the curve values were respectively 0.958 and 0.948.

CONCLUSION

The MRI radiomics method based on enhanced-T1WI images has a good predictive effect on the classification of meningiomas and can provide a basis for planning clinical treatment protocols.

Diffusion-weighted imaging with histogram analysis of the apparent diffusion coefficient maps in the diagnosis of parotid tumours

The aim of this study was to investigate the role of diffusion-weighted imaging (DWI) with histogram analysis of apparent diffusion coefficient (ADC) maps in the characterization of parotid tumours. This prospective study included 39 patients with parotid tumours. All patients underwent magnetic resonance imaging with DWI, and ADC maps were generated. The whole lesion was selected to obtain histogram-related parameters, including the mean (ADC_mean), minimum (ADC_min), maximum (ADC_max), skewness, and kurtosis of the ADC. The final diagnosis included pleomorphic adenoma (PA; n=18), Warthin tumour (WT; n=12), and salivary gland malignancy (SGM; n=9). ADC_mean (×10^-3mm²/s) was 1.93±0.34 for PA, 1.01±0.11 for WT, and 1.26±0.54 for SGM. There was a significant difference in whole lesion ADC_mean among the three study groups. Skewness had the best diagnostic performance in differentiating PA from WT (P=0.001; best detected cut-off 0.41, area under the curve (AUC) 0.990) and in discriminating WT from SGM (P=0.03; best detected cut-off 0.74, AUC 0.806). The whole lesion ADC_mean value had best diagnostic performance in differentiating PA from SGM (P=0.007; best detected cut-off 1.16×10^-3mm²/s, AUC 0.948). In conclusion, histogram analysis of ADC maps may offer added value in the differentiation of parotid tumours.

Association of apparent diffusion coefficient with Ki-67 proliferation index, progesterone-receptor status and various histopathological parameters, and its utility in predicting the high grade in meningiomas

BACKGROUND

Accurate preoperative determination of the histological grade and cellular proliferative potential of meningioma by non-invasive imaging is of paramount importance.

PURPOSE

To evaluate the utility of apparent diffusion coefficient (ADC) in determining the histological grade of meningioma, and to investigate the correlation of ADC with Ki-67 proliferation index (PI), progesterone receptor (PR) status, and a number of other histopathological parameters.

MATERIAL AND METHODS

Histopathologically confirmed 94 meningioma patients (72 low-grade, 22 high-grade) who had undergone preoperative diffusion-weighted imaging were retrospectively evaluated. ADC values were obtained by manually drawing the regions of interest (ROIs) within the solid components of the tumor. The relationship between ADC and Ki-67 values, PR status, and multiple histopathological parameters were investigated, and the ADC values of high-grade and low-grade meningiomas were compared. Independent sample t-test, Mann-Whitney U test, receiver operating characteristic, Pearson correlation, and multiple logistic regression analysis were used for statistical assessment.

RESULTS

All ADC and rADC values were significantly lower in high-grade meningiomas than in low-grade meningiomas (all P < 0.05). ADC values showed significantly negative correlations with Ki-67 and mitotic index (P < 0.001 for each). Numerous ADC parameters were significantly lower in meningiomas demonstrating hypercellularity and necrosis features (P < 0.05). ADC values did not show a significant correlation with PR score (all P > 0.05).

CONCLUSION

ADC can be utilized as a reliable imaging biomarker for predicting the proliferative potential and histological grade in meningiomas.

Differential Diagnosis of Solitary Fibrous Tumor/Hemangiopericytoma and Angiomatous Meningioma Using Three-Dimensional Magnetic Resonance Imaging Texture Feature Model

Background

Intracranial solitary fibrous tumor(SFT)/hemangiopericytoma (HPC) is an aggressive malignant tumor originating from the intracranial vasculature. Angiomatous meningioma (AM) is a benign tumor with a good prognosis. The imaging manifestations of the two are very similar. Thus, novel noninvasive diagnostic method is urgently needed in clinical practice. Texture analysis and model building through machine learning may have good prospects.

Aim

To evaluate whether a 3D-MRI texture feature model could be used to differentiate malignant intracranial SFT/HPC from AM.

Method

A total of 97 patients with SFT/HPC and 95 with AM were included in this study. Patients from each group were randomly divided into the train (70%) and test (30%) sets. ROIs were drawn along the edge of the tumor on each section of T1WI, T2WI, and contrasted T1WI using ITK-SNAP software. The segmented image was imported into the AK software for texture feature extraction, and the 3D ROI signal intensity histograms of T1WI, T2WI, and contrasted T1WI were automatically obtained along with all the parameters. Modeling was performed using the language R. Confusion matrix was used to analyze the accuracy of the model. ROC curve was constructed to assess the grading ability of the logistic regression model.

Results

After Lasso dimension reduction, 5, 9, and 7 texture features were extracted from T1WI, T2WI, and contrasted T1WI, respectively; additional 8 texture features were extracted from the combined sequence for modeling. The ROC analyses on four models resulted in an area under the curve (AUC) of 0.885 (sensitivity 76.1%, specificity 87.9%) for T1WI model, 0.918 (73.1%, 95.5%) for T2WI model, 0.815 (55.2%, 93.9%) for contrasted T1WI model, and 0.959 (92.5%, 84.8%) for the combined sequence model and were enough to correctly distinguish the two groups in 71.2%, 81.4%, 69.5%, and 83.1% of cases in test set, respectively.

Conclusions

The radiological model based on texture features could be used to differentiate SFT/HPC from AM.

Training Deep Neural Networks for Small and Highly Heterogeneous MRI Datasets for Cancer Grading

Using medical images recorded in clinical practice has the potential to be a game-changer in the application of machine learning for medical decision support. Thousands of medical images are produced in daily clinical activity. The diagnosis of medical doctors on these images represents a source of knowledge to train machine learning algorithms for scientific research or computer-aided diagnosis. However, the requirement of manual data annotations and the heterogeneity of images and annotations make it difficult to develop algorithms that are effective on images from different centers or sources (scanner manufacturers, protocols, etc.). The objective of this article is to explore the opportunities and the limits of highly heterogeneous biomedical data, since many medical data sets are small and entail a challenge for machine learning techniques. Particularly, we focus on a small data set targeting meningioma grading. Meningioma grading is crucial for patient treatment and prognosis. It is normally performed by histological examination but recent articles showed that it is possible to do it also on magnetic resonance images (MRI), so non-invasive. Our data set consists of 174 T1-weighted MRI images of patients with meningioma, divided into 126 benign and 48 atypical/anaplastic cases, acquired using 26 different MRI scanners and 125 acquisition protocols, which shows the enormous variability in the data set. The performed preprocessing steps include tumor segmentation, spatial image normalization and data augmentation based on color and affine transformations. The preprocessed cases are passed to a carefully trained 2-D convolutional neural network. Accuracy above 74% was obtained, with the high-grade tumor recall above 74%. The results are encouraging considering the limited size and high heterogeneity of the data set. The proposed methodology can be useful for other problems involving classification of small and highly heterogeneous data sets.

Machine Learning in Meningioma MRI: Past to Present. A Narrative Review

Meningioma is one of the most frequent primary central nervous system tumors. While magnetic resonance imaging (MRI), is the standard radiologic technique for provisional diagnosis and surveillance of meningioma, it nevertheless lacks the prima facie capacity in determining meningioma biological aggressiveness, growth, and recurrence potential. An increasing body of evidence highlights the potential of machine learning and radiomics in improving the consistency and productivity and in providing novel diagnostic, treatment, and prognostic modalities in neuroncology imaging. The aim of the present article is to review the evolution and progress of approaches utilizing machine learning in meningioma MRI-based sementation, diagnosis, grading, and prognosis. We provide a historical perspective on original research on meningioma spanning over two decades and highlight recent studies indicating the feasibility of pertinent approaches, including deep learning in addressing several clinically challenging aspects. We indicate the limitations of previous research designs and resources and propose future directions by highlighting areas of research that remain largely unexplored. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY STAGE: 2.

Histogram analysis in predicting the grade and histological subtype of meningiomas based on diffusion kurtosis imaging

BACKGROUND

Presurgical grading is particularly important for selecting the best therapeutic strategy for meningioma patients.

PURPOSE

To investigate the value of histogram analysis of diffusion kurtosis imaging (DKI) maps in the differentiation of grades and histological subtypes of meningiomas.

MATERIAL AND METHODS

A total of 172 patients with histopathologically proven meningiomas underwent preoperative magnetic resonance imaging (MRI) and were classified into low-grade and high-grade groups. Mean kurtosis (MK), fractional anisotropy (FA), and mean diffusivity (MD) histograms were generated based on solid components of the whole tumor. The following parameters of each histogram were obtained: 10th, 25th, 75th, and 90th percentiles, mean, median, maximum, minimum, and kurtosis, skewness, and variance. Comparisons of different grades and subtypes were made by Mann-Whitney U test, Kruskal-Wallis test, ROC curves analysis, and multiple logistic regression. Pearson correlation was used to evaluate correlations between histogram parameters and the Ki-67 labeling index.

RESULTS

Significantly higher maximum, skewness, and variance of MD, mean, median, maximum, variance, 10th, 25th, 75th, and 90th percentiles of MK were found in high-grade than low-grade meningiomas (all P < 0.05). DKI histogram parameters differentiated 7/10 pairs of subtype pairs. The 90th percentile of MK yielded the highest AUC of 0.870 and was the only independent indicator for grading meningiomas. Various DKI histogram parameters were correlated with the Ki-67 labeling index (P < 0.05).

CONCLUSION

The histogram analysis of DKI is useful for differentiating meningioma grades and subtypes. The 90th percentile of MK may serve as an optimal parameter for predicting the grade of meningiomas.

Gamma Knife radiosurgery for intracranial benign meningiomas: follow-up outcome in 130 patients

OBJECTIVEThe authors retrospectively analyzed the follow-up data in 130 patients with intracranial benign meningiomas after Gamma Knife radiosurgery (GKRS), evaluated the tumor progression-free survival (PFS) rate and neurological function preservation rate, and determined the predictors by univariate and multivariate survival analysis.METHODSThis cohort of 130 patients with intracranial benign meningiomas underwent GKRS between May 2012 and May 2015 at the Second Hospital of Tianjin Medical University. The median age was 54.5 years (range 25-81 years), and women outnumbered men at a ratio of 4.65:1. All clinical and radiological data were obtained for analysis. No patient had undergone prior traditional radiotherapy or chemotherapy. The median tumor volume was 3.68 cm3 (range 0.23-45.78 cm3). A median margin dose of 12.0 Gy (range 10.0-16.0 Gy) was delivered to the tumor with a median isodose line of 50% (range 50%-60%).RESULTSDuring a median follow-up of 36.5 months (range 12-80 months), tumor volume regressed in 37 patients (28.5%), was unchanged in 86 patients (66.2%), and increased in 7 patients (5.4%). The actuarial tumor progression-free survival (PFS) rate was 98%, 94%, and 87% at 1, 3, and 5 years, respectively, after GKRS. Tumor recurred in 7 patients at a median follow-up of 32 months (range 12-56 months). Tumor volume ≥ 10 cm3 (p = 0.012, hazard ratio [HR] 8.25, 95% CI 1.60-42.65) and pre-GKRS Karnofsky Performance Scale score < 90 (p = 0.006, HR 9.31, 95% CI 1.88-46.22) were independent unfavorable predictors of PFS rate after GKRS. Of the 130 patients, 101 (77.7%) presented with one or more neurological symptoms or signs before GKRS. Neurological symptoms or signs improved in 40 (30.8%) patients, remained stable in 83 (63.8%), and deteriorated in 7 (5.4%) after GKRS. Two (1.5%) patients developed new cranial nerve (CN) deficit. Tumor volume ≥ 10 cm3 (p = 0.042, HR = 4.73, 95% CI 1.06-21.17) and pre-GKRS CN deficit (p = 0.045, HR = 4.35, 95% CI 0.84-22.48) were independent unfavorable predictors for improvement in neurological symptoms or signs. Six (4.6%) patients developed new or worsening peritumoral edema with a median follow-up of 4.5 months (range 2-7 months).CONCLUSIONSGKRS provided good local tumor control and high neurological function preservation in patients with intracranial benign meningiomas. Patients with tumor volume < 10 cm3, pre-GKRS Karnofsky Performance Scale score ≥ 90, and no pre-GKRS CN deficit (I-VIII) can benefit from stereotactic radiosurgery. It can be considered as the primary or adjuvant management of intracranial benign meningiomas.

Preoperative vascular heterogeneity and aggressiveness assessment of pituitary macroadenoma based on dynamic contrast-enhanced MRI texture analysis

PURPOSE

To assess the vascular heterogeneity and aggressiveness of pituitary macroadenomas (PM) using texture analysis based on Dynamic Contrast-Enhanced MRI (DCE-MRI).

METHOD

Fifty patients with pathologically confirmed PM, including 32 patients with aggressive PM (aggressive group) and 18 patients with non-aggressive PM (non-aggressive group), were included in this study. The preoperative DCE-MRI and clinical data were collected from all patients. The features based on Ktrans, Ve, and Kep were generated using Omni-Kinetics software. Independent-samples t-test and Mann-Whitney U test were used for comparison between two groups. Logistic regression analysis was used to determine the optimal model for distinguishing aggressive and non-aggressive PM.

RESULTS

Six features related to tumor morphology, 24 features in Ktrans, 20 features in Ve, and 3 features in Kep were significantly different between the aggressive and non-aggressive groups. Volume count, gray-level non-uniformity in Ktrans, voxel value sum in Ve and run-length non-uniformity in Kep (AUC = 0.816, 0.903, 0.785, 0.813) were considered the best feature for tumor diagnosis. After modeling, the diagnosis efficiency of mean model and total model was desirable (AUC = 0.859 and 0.957), and the diagnostic efficiency of morphological, Ktrans, Ve and Kep features model was improved (AUC = 0.845, 0.951, 0.847, 0.804).

CONCLUSIONS

Texture analysis based on DCE-MRI elucidates the vascular heterogeneity and aggressiveness of pituitary adenoma. The total model could be used as a new noninvasive method for predicting the aggressiveness of pituitary macroadenoma.