Machine Learning-Based Multiparametric Magnetic Resonance Imaging Radiomics for Prediction of H3K27M Mutation in Midline Gliomas

Machine learning radiomics for H3K27M mutation prediction in gliomas: A systematic review and meta-analysis

PURPOSE

Noninvasive prediction and identification of the H3K27M mutation play an important role in optimizing therapeutic strategies and improving outcomes in gliomas. In this systematic review and meta-analysis, we aimed to evaluate the performance of machine learning (ML)-based models in predicting H3K27M mutation in gliomas.

METHODS

Literature records were retrieved on September 16th, 2024, in PubMed, Embase, Scopus, and Web of Science. Records were screened according to the eligibility criteria, and the data from the included studies were extracted. The meta-analysis, sensitivity analysis, and meta-regression were conducted using R software.

RESULTS

A total of 15 studies were included in our study. Our meta-analysis demonstrated a pooled AUC, sensitivity, and specificity of 0.87 (95% CI: 0.77-0.97), 92% (95% CI: 83%-96%), and 89% (95% CI: 86%-91%)), respectively. The subgroup meta-analysis revealed that despite the higher sensitivity of the deep learning (DL) models, the sensitivity is not superior to ML (P = 0.6). In contrast, the ML-based pooled specificity was significantly higher (P < 0.01). The meta-analysis revealed a 78.1 (95% CI: 33.3 - 183.5). The SROC curve indicated an AUC of 0.921, and the estimated sensitivity is 0.898 concurrent with the false positive rate of 0.126, which indicates high sensitivity with a low false positive rate.

CONCLUSION

Our systematic review and meta-analysis demonstrated that ML-based magnetic resonance imaging (MRI) radiomics models are associated with promising diagnostic performance in predicting H3K27M mutation in gliomas.

Artificial Intelligence for Neuroimaging in Pediatric Cancer

BACKGROUND/OBJECTIVES

Artificial intelligence (AI) is transforming neuroimaging by enhancing diagnostic precision and treatment planning. However, its applications in pediatric cancer neuroimaging remain limited. This review assesses the current state, potential applications, and challenges of AI in pediatric neuroimaging for cancer, emphasizing the unique needs of the pediatric population.

METHODS

A comprehensive literature review was conducted, focusing on AI's impact on pediatric neuroimaging through accelerated image acquisition, reduced radiation, and improved tumor detection. Key methods include convolutional neural networks for tumor segmentation, radiomics for tumor characterization, and several tools for functional imaging. Challenges such as limited pediatric datasets, developmental variability, ethical concerns, and the need for explainable models were analyzed.

RESULTS

AI has shown significant potential to improve imaging quality, reduce scan times, and enhance diagnostic accuracy in pediatric neuroimaging, resulting in improved accuracy in tumor segmentation and outcome prediction for treatment. However, progress is hindered by the scarcity of pediatric datasets, issues with data sharing, and the ethical implications of applying AI in vulnerable populations.

CONCLUSIONS

To overcome current limitations, future research should focus on building robust pediatric datasets, fostering multi-institutional collaborations for data sharing, and developing interpretable AI models that align with clinical practice and ethical standards. These efforts are essential in harnessing the full potential of AI in pediatric neuroimaging and improving outcomes for children with cancer.

Advanced imaging techniques and non-invasive biomarkers in pediatric brain tumors: state of the art

In the pediatric age group, brain neoplasms are the second most common tumor category after leukemia, with an annual incidence of 6.13 per 100,000. Conventional MRI sequences, complemented by CT whenever necessary, are fundamental for the initial diagnosis and surgical planning as well as for post-operative evaluations, assessment of response to treatment, and surveillance; however, they have limitations, especially concerning histopathologic or biomolecular phenotyping and grading. In recent years, several advanced MRI sequences, including diffusion-weighted imaging, diffusion tensor imaging, arterial spin labelling (ASL) perfusion, and MR spectroscopy, have emerged as a powerful aid to diagnosis as well as prognostication; furthermore, other techniques such as diffusion kurtosis, amide proton transfer imaging, and MR elastography are being translated from the research environment to clinical practice. Molecular imaging, especially PET with amino-acid tracers, complement MRI in several aspects, including biopsy targeting and outcome prediction. Finally, radiomics with radiogenomics are opening entirely new perspectives for a quantitative approach aiming at identifying biomarkers that can be used for personalized, precision management strategies.

Deep mutual learning on hybrid amino acid PET predicts H3K27M mutations in midline gliomas

Predicting H3K27M mutation status in midline gliomas non-invasively is of considerable interest, particularly using deep learning with 11C-methionine (MET) and 18F-fluoroethyltyrosine (FET) positron emission tomography (PET). To optimise prediction efficiency, we derived an assistance training (AT) scheme to allow mutual benefits between MET and FET learning to boost the predictability but still only require either PET as inputs for predictions. Our method significantly surpassed conventional convolutional neural network (CNN), radiomics-based, and MR-based methods, achieved an area under the curve (AUC) of 0.9343 for MET, and an AUC of 0.8619 for FET during internal cross-validation (n = 90). The performance remained high in hold-out testing (n = 19) and consecutive testing cohorts (n = 21), with AUCs of 0.9205 and 0.7404. The clinical feasibility of the proposed method was confirmed by the agreements to multi-departmental decisions and outcomes in pathology-uncertain cases. The findings positions our method as a promising tool for aiding treatment decisions in midline glioma.

-New frontiers in domain-inspired radiomics and radiogenomics: increasing role of molecular diagnostics in CNS tumor classification and grading following WHO CNS-5 updates

Gliomas and Glioblastomas represent a significant portion of central nervous system (CNS) tumors associated with high mortality rates and variable prognosis. In 2021, the World Health Organization (WHO) updated its Glioma classification criteria, most notably incorporating molecular markers including CDKN2A/B homozygous deletion, TERT promoter mutation, EGFR amplification, + 7/-10 chromosome copy number changes, and others into the grading and classification of adult and pediatric Gliomas. The inclusion of these markers and the corresponding introduction of new Glioma subtypes has allowed for more specific tailoring of clinical interventions and has inspired a new wave of Radiogenomic studies seeking to leverage medical imaging information to explore the diagnostic and prognostic implications of these new biomarkers. Radiomics, deep learning, and combined approaches have enabled the development of powerful computational tools for MRI analysis correlating imaging characteristics with various molecular biomarkers integrated into the updated WHO CNS-5 guidelines. Recent studies have leveraged these methods to accurately classify Gliomas in accordance with these updated molecular-based criteria based solely on non-invasive MRI, demonstrating the great promise of Radiogenomic tools. In this review, we explore the relative benefits and drawbacks of these computational frameworks and highlight the technical and clinical innovations presented by recent studies in the landscape of fast evolving molecular-based Glioma subtyping. Furthermore, the potential benefits and challenges of incorporating these tools into routine radiological workflows, aiming to enhance patient care and optimize clinical outcomes in the evolving field of CNS tumor management, have been highlighted.

The Performance of Machine Learning for Prediction of H3K27 M Mutation in Midline Gliomas: A Systematic Review and Meta-Analysis

BACKGROUND

Diffuse midline gliomas (DMGs) encompass a set of tumors, and those tumors with H3K27 M mutation carry a poor prognosis. In recent years, machine learning (ML)-based radiomics have shown promising results in predicting gene mutation status non-invasively. Therefore, this study aims to comprehensively evaluate the diagnostic performance of ML-based magnetic resonance imaging radiomics in predicting H3K27 M mutation status in DMG patients.

METHODS

A systematic search was conducted using relevant keywords in PubMed/Medline, Scopus, Embase, and Web of Science from inception to May 2023. Original studies evaluating the diagnostic performance of ML models in predicting H3K27 M mutation status in DMGs were enrolled. Quality assessment of the enrolled studies was conducted using QUADAS-2. Data were analyzed using STATA version 17.0 to calculate pooled sensitivity, specificity, positive and negative likelihood ratio, diagnostic score, and diagnostic odds ratio.

RESULTS

A total of 13 studies, including 12 retrospectives and 1 both retrospective and prospective study, enrolled 1510 (male = 777) DMG patients. Six studies underwent meta-analysis which showed a pooled sensitivity, specificity, positive and negative likelihood ratio, diagnostic score, and diagnostic odds ratio of 0.91 (95% CI 0.77-0.97), 0.81 (95% CI 0.73-0.88), 4.86 (95% CI 3.25-7.24), 0.11 (95% CI 0.04-0.29), 3.75 (95% CI 2.62-4.88), and 42.61 (95% CI 13.77-131.87), respectively.

CONCLUSIONS

Non-invasive prediction of H3K27 M mutation status in patients with DMGs using magnetic resonance imaging radiomics is a promising tool with good diagnostic performance. However, the pooled metrics had a wide confidence interval, which required further studies to enhance ML algorithms' accuracy and facilitate their integration into daily clinical practice.

Diffusion MRI is valuable in brainstem glioma genotyping with quantitative measurements of white matter tracts

OBJECTIVES

To investigate the value of diffusion MRI (dMRI) in H3K27M genotyping of brainstem glioma (BSG).

METHODS

A primary cohort of BSG patients with dMRI data (b = 0, 1000 and 2000 s/mm2) and H3K27M mutation information were included. A total of 13 diffusion tensor and kurtosis imaging (DTI; DKI) metrics were calculated, then 17 whole-tumor histogram features and 29 along-tract white matter (WM) microstructural measurements were extracted from each metric and assessed within genotypes. After feature selection through univariate analysis and the least absolute shrinkage and selection operator method, multivariate logistic regression was used to build dMRI-derived genotyping models based on retained tumor and WM features separately and jointly. Model performances were tested using ROC curves and compared by the DeLong approach. A nomogram incorporating the best-performing dMRI model and clinical variables was generated by multivariate logistic regression and validated in an independent cohort of 27 BSG patients.

RESULTS

At total of 117 patients (80 H3K27M-mutant) were included in the primary cohort. In total, 29 tumor histogram features and 41 WM tract measurements were selected for subsequent genotyping model construction. Incorporating WM tract measurements significantly improved diagnostic performances (p < 0.05). The model incorporating tumor and WM features from both DKI and DTI metrics showed the best performance (AUC = 0.9311). The nomogram combining this dMRI model and clinical variables achieved AUCs of 0.9321 and 0.8951 in the primary and validation cohort respectively.

CONCLUSIONS

dMRI is valuable in BSG genotyping. Tumor diffusion histogram features are useful in genotyping, and WM tract measurements are more valuable in improving genotyping performance.

CLINICAL RELEVANCE STATEMENT

This study found that diffusion MRI is valuable in predicting H3K27M mutation in brainstem gliomas, which is helpful to realize the noninvasive detection of brainstem glioma genotypes and improve the diagnosis of brainstem glioma.

KEY POINTS

• Diffusion MRI has significant value in brainstem glioma H3K27M genotyping, and models with satisfactory performances were built. • Whole-tumor diffusion histogram features are useful in H3K27M genotyping, and quantitative measurements of white matter tracts are valuable as they have the potential to improve model performance. • The model combining the most discriminative diffusion MRI model and clinical variables can help make clinical decision.

Prediction of H3K27M alteration in midline gliomas of the brain using radiomics: A multi-institute study

Background

Noninvasive prediction of H3K27M-altered Diffuse midline gliomas is important because of the involvement of deep locations and proximity to eloquent structures. We aim to predict H3K27M alteration in midline gliomas using radiomics features of T2W images.

Methods

Radiomics features extracted from 124 subjects (69 H3K27M-altered/55 H3K27M-wild type). T2W images were resampled to 1 × 1 × 1mm3 voxel size, preprocessed, and normalized for artifact correction, intensity variations. The feature set was normalized and subjected to reduction by variance thresholding, correlation coefficient thresholding, and sequential feature selector. Adaptive synthesis oversampling technique was used to oversample the training data. Random forest classifier (RFC), Decision tree classifier (DTC), and K-nearest neighbors classifier (KNN) were trained over the training dataset and the performance was assessed over the internal test dataset and external test data set (52 subjects: 33 H3K27M-altered/19-H3K27M-wild type).

Results

DTC achieved a validation score of 77.33% (5-fold cross-validation) and an accuracy of 80.64%, 75% on internal and external test datasets. RFC achieved a validation score of 80.7% (5-fold cross-validation) an accuracy of 80.6%, and 73% on internal and external test datasets. DTC achieved a validation score of 78.67% (5-fold cross-validation) an accuracy of 80.64%, and 61.53% on internal and external test datasets. The accuracy score of DTC, RFC, and KNN on the internal test dataset was approximately 80% while on the external test dataset, DTC achieved 75% accuracy, RFC achieved 73% accuracy and KNN achieved 65.1% accuracy.

Conclusions

H3K27M alteration is a potential immunotherapeutic marker and is associated with poor prognosis and radiomics features extracted from conventional T2W-images can help in identifying H3K27M-altered cases non-invasively with high precision.

Automated Determination of the H3 K27-Altered Status in Spinal Cord Diffuse Midline Glioma by Radiomics Based on T2-Weighted MR Images

BACKGROUND AND PURPOSE

Conventional MR imaging is not sufficient to discern the H3 K27-altered status of spinal cord diffuse midline glioma. This study aimed to develop a radiomics-based model based on preoperative T2WI to determine the H3 K27-altered status of spinal cord diffuse midline glioma.

MATERIALS AND METHODS

Ninety-seven patients with confirmed spinal cord diffuse midline gliomas were retrospectively recruited and randomly assigned to the training (n = 67) and test (n = 30) sets. One hundred seven radiomics features were initially extracted from automatically-segmented tumors on T2WI, then 11 features selected by the Pearson correlation coefficient and the Kruskal-Wallis test were used to train and test a logistic regression model for predicting the H3 K27-altered status. Sensitivity analysis was performed using additional random splits of the training and test sets, as well as applying other classifiers for comparison. The performance of the model was evaluated through its accuracy, sensitivity, specificity, and area under the curve. Finally, a prospective set including 28 patients with spinal cord diffuse midline gliomas was used to validate the logistic regression model independently.

RESULTS

The logistic regression model accurately predicted the H3 K27-altered status with accuracies of 0.833 and 0.786, sensitivities of 0.813 and 0.750, specificities of 0.857 and 0.833, and areas under the curve of 0.839 and 0.818 in the test and prospective sets, respectively. Sensitivity analysis confirmed the robustness of the model, with predictive accuracies of 0.767-0.833.

CONCLUSIONS

Radiomics signatures based on preoperative T2WI could accurately predict the H3 K27-altered status of spinal cord diffuse midline glioma, providing potential benefits for clinical management.

Deep Learning for Noninvasive Assessment of H3 K27M Mutation Status in Diffuse Midline Gliomas Using MR Imaging

BACKGROUND

Determination of H3 K27M mutation in diffuse midline glioma (DMG) is key for prognostic assessment and stratifying patient subgroups for clinical trials. MRI can noninvasively depict morphological and metabolic characteristics of H3 K27M mutant DMG.

PURPOSE

This study aimed to develop a deep learning (DL) approach to noninvasively predict H3 K27M mutation in DMG using T2-weighted images.

STUDY TYPE

Retrospective and prospective.

POPULATION

For diffuse midline brain gliomas, 341 patients from Center-1 (27 ± 19 years, 184 males), 42 patients from Center-2 (33 ± 19 years, 27 males) and 35 patients (37 ± 18 years, 24 males). For diffuse spinal cord gliomas, 133 patients from Center-1 (30 ± 15 years, 80 males).

FIELD STRENGTH/SEQUENCE

5T and 3T, T2-weighted turbo spin echo imaging.

ASSESSMENT

Conventional radiological features were independently reviewed by two neuroradiologists. H3 K27M status was determined by histopathological examination. The Dice coefficient was used to evaluate segmentation performance. Classification performance was evaluated using accuracy, sensitivity, specificity, and area under the curve.

STATISTICAL TESTS

Pearson's Chi-squared test, Fisher's exact test, two-sample Student's t-test and Mann-Whitney U test. A two-sided P value <0.05 was considered statistically significant.

RESULTS

In the testing cohort, Dice coefficients of tumor segmentation using DL were 0.87 for diffuse midline brain and 0.81 for spinal cord gliomas. In the internal prospective testing dataset, the predictive accuracies, sensitivities, and specificities of H3 K27M mutation status were 92.1%, 98.2%, 82.9% in diffuse midline brain gliomas and 85.4%, 88.9%, 82.6% in spinal cord gliomas. Furthermore, this study showed that the performance generalizes to external institutions, with predictive accuracies of 85.7%-90.5%, sensitivities of 90.9%-96.0%, and specificities of 82.4%-83.3%.

DATA CONCLUSION

In this study, an automatic DL framework was developed and validated for accurately predicting H3 K27M mutation using T2-weighted images, which could contribute to the noninvasive determination of H3 K27M status for clinical decision-making.

EVIDENCE LEVEL

2 Technical Efficacy: Stage 2.

Identification of Radiomic Signatures in Brain MRI Sequences T1 and T2 That Differentiate Tumor Regions of Midline Gliomas with H3.3K27M Mutation

BACKGROUND

Radiomics refers to the acquisition of traces of quantitative features that are usually non-perceptible to human vision and are obtained from different imaging techniques and subsequently transformed into high-dimensional data. Diffuse midline gliomas (DMG) represent approximately 20% of pediatric CNS tumors, with a median survival of less than one year after diagnosis. We aimed to identify which radiomics can discriminate DMG tumor regions (viable tumor and peritumoral edema) from equivalent midline normal tissue (EMNT) in patients with the positive H3.F3K27M mutation, which is associated with a worse prognosis.

PATIENTS AND METHODS

This was a retrospective study. From a database of 126 DMG patients (children, adolescents, and young adults), only 12 had H3.3K27M mutation and available brain magnetic resonance DICOM file. The MRI T1 post-gadolinium and T2 sequences were uploaded to LIFEx software to post-process and extract radiomic features. Statistical analysis included normal distribution tests and the Mann-Whitney U test performed using IBM SPSS® (Version 27.0.0.1, International Business Machines Corp., Armonk, NY, USA), considering a significant statistical p-value ≤ 0.05.

RESULTS

EMNT vs. Tumor: From the T1 sequence 10 radiomics were identified, and 14 radiomics from the T2 sequence, but only one radiomic identified viable tumors in both sequences (p < 0.05) (DISCRETIZED_Q1). Peritumoral edema vs. EMNT: From the T1 sequence, five radiomics were identified, and four radiomics from the T2 sequence. However, four radiomics could discriminate peritumoral edema in both sequences (p < 0.05) (CONVENTIONAL_Kurtosis, CONVENTIONAL_ExcessKurtosis, DISCRETIZED_Kurtosis, and DISCRETIZED_ExcessKurtosis). There were no radiomics useful for distinguishing tumor tissue from peritumoral edema in both sequences.

CONCLUSIONS

Less than 5% of the radiomic characteristics identified tumor regions of medical-clinical interest in T1 and T2 sequences of conventional magnetic resonance imaging. The first-order and second-order radiomic features suggest support to investigators and clinicians for careful evaluation for diagnosis, patient classification, and multimodality cancer treatment planning.

REliability of consensus-based segMentatIoN in raDiomic feature reproducibility (REMIND): A word of caution

OBJECTIVE

To evaluate the reliability of consensus-based segmentation in terms of reproducibility of radiomic features.

METHODS

In this retrospective study, three tumor data sets were investigated: breast cancer (n = 30), renal cell carcinoma (n = 30), and pituitary macroadenoma (n = 30). MRI was utilized for breast and pituitary data sets, while CT was used for renal data set. 12 readers participated in the segmentation process. Consensus segmentation was created by making corrections on a previous region or volume of interest. Four experiments were designed to evaluate the reproducibility of radiomic features. Reliability was assessed with intraclass correlation coefficient (ICC) with two cut-off values: 0.75 and 0.9.

RESULTS

Considering the lower bound of the 95% confidence interval and the ICC threshold of 0.90, at least 61% of the radiomic features were not reproducible in the inter-consensus analysis. In the susceptibility experiment, at least half (54%) became non-reproducible when the first reader is replaced with a different reader. In the intra-consensus analysis, at least about one-third (32%) were non-reproducible when the same second reader segmented the image over the same first reader two weeks later. Compared to inter-reader analysis based on independent single readers, the inter-consensus analysis did not statistically significantly improve the rates of reproducible features in all data sets and analyses.

CONCLUSIONS

Despite the positive connotation of the word "consensus", it is essential to REMIND that consensus-based segmentation has significant reproducibility issues. Therefore, the usage of consensus-based segmentation alone should be avoided unless a reliability analysis is performed, even if it is not practical in clinical settings.

Deep learning-based prediction of H3K27M alteration in diffuse midline gliomas based on whole-brain MRI

BACKGROUND

H3K27M mutation status significantly affects the prognosis of patients with diffuse midline gliomas (DMGs), but this tumor presents a high risk of pathological acquisition. We aimed to construct a fully automated model for predicting the H3K27M alteration status of DMGs based on deep learning using whole-brain MRI.

METHODS

DMG patients from West China Hospital of Sichuan University (WCHSU; n = 200) and Chengdu Shangjin Nanfu Hospital (CSNH; n = 35) who met the inclusion and exclusion criteria from February 2016 to April 2022 were enrolled as the training and external test sets, respectively. To adapt the model to the human head MRI scene, we use normal human head MR images to pretrain the model. The classification and tumor segmentation tasks are naturally related, so we conducted cotraining for the two tasks to enable information interaction between them and improve the accuracy of the classification task.

RESULTS

The average classification accuracies of our model on the training and external test sets was 90.5% and 85.1%, respectively. Ablation experiments showed that pretraining and cotraining could improve the prediction accuracy and generalization performance of the model. In the training and external test sets, the average areas under the receiver operating characteristic curve (AUROCs) were 94.18% and 87.64%, and the average areas under the precision-recall curve (AUPRC) were 93.26% and 85.4%.

CONCLUSIONS

The developed model achieved excellent performance in predicting the H3K27M alteration status in DMGs, and its good reproducibility and generalization were verified in the external dataset.

Diagnostic Performance of Selected MRI-Derived Radiomics Able to Discriminate Progression-Free and Overall Survival in Patients with Midline Glioma and the H3F3AK27M Mutation

BACKGROUND

Radiomics refers to a recent area of knowledge that studies features extracted from different imaging techniques and subsequently transformed into high-dimensional data that can be associated with biological events. Diffuse midline gliomas (DMG) are one of the most devastating types of cancer, with a median survival of approximately 11 months after diagnosis and 4-5 months after radiological and clinical progression.

METHODS

A retrospective study. From a database of 91 patients with DMG, only 12 had the H3.3K27M mutation and brain MRI DICOM files available. Radiomic features were extracted from MRI T1 and T2 sequences using LIFEx software. Statistical analysis included normal distribution tests and the Mann-Whitney U test, ROC analysis, and calculation of cut-off values.

RESULTS

A total of 5760 radiomic values were included in the analyses. AUROC demonstrated 13 radiomics with statistical significance for progression-free survival (PFS) and overall survival (OS). Diagnostic performance tests showed nine radiomics with specificity for PFS above 90% and one with a sensitivity of 97.2%. For OS, 3 out of 4 radiomics demonstrated between 80 and 90% sensitivity.

CONCLUSIONS

Several radiomic features demonstrated statistical significance and have the potential to further aid DMG diagnostic assessment non-invasively. The most significant radiomics were first- and second-order features with GLCM texture profile, GLZLM_GLNU, and NGLDM_Contrast.

Imaging features associated with H3 K27-altered and H3 G34-mutant gliomas: a narrative systematic review

Background

Advances in molecular diagnostics accomplished the discovery of two malignant glioma entities harboring alterations in the H3 histone: diffuse midline glioma, H3 K27-altered and diffuse hemispheric glioma, H3 G34-mutant. Radiogenomics research, which aims to correlate tumor imaging features with genotypes, has not comprehensively examined histone-altered gliomas (HAG). The aim of this research was to synthesize the current published data on imaging features associated with HAG.

Methods

A systematic search was performed in March 2022 using PubMed and the Cochrane Library, identifying studies on the imaging features associated with H3 K27-altered and/or H3 G34-mutant gliomas.

Results

Forty-seven studies fulfilled the inclusion criteria, the majority on H3 K27-altered gliomas. Just under half (21/47) were case reports or short series, the remainder being diagnostic accuracy studies. Despite heterogeneous methodology, some themes emerged. In particular, enhancement of H3 K27M-altered gliomas is variable and can be less than expected given their highly malignant behavior. Low apparent diffusion coefficient values have been suggested as a biomarker of H3 K27-alteration, but high values do not exclude this genotype. Promising correlations between high relative cerebral blood volume values and H3 K27-alteration require further validation. Limited data on H3 G34-mutant gliomas suggest some morphologic overlap with 1p/19q-codeleted oligodendrogliomas.

Conclusions

The existing data are limited, especially for H3 G34-mutant gliomas and artificial intelligence techniques. Current evidence indicates that imaging-based predictions of HAG are insufficient to replace histological assessment. In particular, H3 K27-altered gliomas should be considered when occurring in typical midline locations irrespective of enhancement characteristics.

Non-Invasive Prediction of Survival Time of Midline Glioma Patients Using Machine Learning on Multiparametric MRI Radiomics Features

OBJECTIVES

To explore the feasibility of predicting overall survival (OS) of patients with midline glioma using multi-parameter magnetic resonance imaging (MRI) features.

METHODS

Data of 84 patients with midline gliomas were retrospectively collected, including 40 patients with OS > 12 months (28 cases were adults, 14 cases were H3 K27M-mutation) and 44 patients with OS < 12 months (29 cases were adults, 31 cases were H3 K27M-mutation). Features were extracted from the largest slice of tumors, which were manually segmented on T2-weighted (T2w), T2 fluid-attenuated inversion recovery (T2 FLAIR), and contrast-enhanced T1-weighted (T1c) images. Data were randomly divided into training (70%) and test cohorts (30%) and normalized and standardized using Z-scores. Feature dimensionality reduction was performed using the variance method and maximum relevance and minimum redundancy (mRMR) algorithm. We used the logistic regression algorithm to construct three models for T2w, T2 FLAIR, and T1c images as well as one combined model. The test cohort was used to evaluate the models, and receiver operating characteristic (ROC) curves, areas under the curve (AUCs), sensitivity, specificity, and accuracy were calculated. The nomogram of the combined model was built and evaluated using a calibration curve. Decision curve analysis (DCA) was used to evaluate the clinical application value of the four models.

RESULTS

A total of 1,316 features were extracted from T2w, T2 FLAIR, and T1c images, respectively. And then the best non-redundant features were selected from the extracted features using the variance method and mRMR. Finally, five features were extracted each from T2w, T2 FLAIR, and T1c images, and 12 features were extracted for the combined model. Four models were established using the optimal features. In the test cohort, the combined model performed the best out of all models. The AUCs of the T2w, T2 FLAIR, T1c, and combined models were 0.73, 0.78, 0.74, and 0.87, respectively, and accuracies were 0.72, 0.76, 0.72, and 0.84, respectively. The ROC curves and DCA showed that the combined model had the highest efficiency and most favorable clinical benefits.

CONCLUSION

The combined radiomics model based on multi-parameter MRI features provided a reliable non-invasive method for the prognostic prediction of midline gliomas.

Imaging diagnosis and treatment selection for brain tumors in the era of molecular therapeutics

Currently, most CNS tumors require tissue sampling to discern their molecular/genomic landscape. However, growing research has shown the powerful role imaging can play in non-invasively and accurately detecting the molecular signature of these tumors. The overarching theme of this review article is to provide neuroradiologists and neurooncologists with a framework of several important molecular markers, their associated imaging features and the accuracy of those features. A particular emphasis is placed on those tumors and mutations that have specific or promising imaging correlates as well as their respective therapeutic potentials.

Prediction of H3 K27M-mutant in midline gliomas by magnetic resonance imaging: a systematic review and meta-analysis

PURPOSE

To summarize the predictive value of MRI for H3 K27M-mutant in midline gliomas using meta-analysis.

METHODS

Systematic electronic searches of the PubMed, Embase, ISI Web of Science, and Cochrane Library up to Jun 31, 2021, were conducted by two experienced neuroradiologists with the keywords of "MRI," "Glioma," and "H3 K27M." The hierarchical summary receiver-operating characteristic (HSROC) model was used to calculate the pooled sensitivity, specificity, positive likelihood ratio (LR +), negative likelihood ratio (LR -), and diagnostic odds ratio (DOR). Coupled forest plots were used to evaluate the heterogeneity of the included studies.

RESULTS

Of seven original studies with a total of 593 patients, 240 glioma patients were included, with 45.5-70.6% H3 K27M-mutant gliomas. Using MRI, a pooled sensitivity of 0.78 (95% CI, 0.66-0.87), specificity of 0.85 (95% CI, 0.76-0.91), LR + of 5.07 (95% CI, 3.19-8.08), LR - of 0.26 (95% CI, 0.16-0.42), and DOR of 19.80 (95% CI, 9.28-42.28) were achieved for H3 K27M-mutant prediction. Significant heterogeneity was observed among the studies in terms of sensitivity (Q = 16.83, df = 7, p = 0.02; I² = 58.40 [95% CI, 25.83-90.97]), LR - (Q = 16.61, df = 7, p = 0.02; I² = 57.87 [95% CI, 24.81-90.93]), and DOR (Q = 14.05, df = 7, p = 0.05; I² = 50.18 [95% CI, 10.06-90.31]).

CONCLUSIONS

This meta-analysis demonstrated a clinical value of MRI to predict H3 K27M-mutant in midline gliomas with a pooled sensitivity of 0.78 and specificity of 0.85.

Multiparametric MRI-Based Radiomics Model for Predicting H3 K27M Mutant Status in Diffuse Midline Glioma: A Comparative Study Across Different Sequences and Machine Learning Techniques

Objectives

The performance of multiparametric MRI-based radiomics models for predicting H3 K27M mutant status in diffuse midline glioma (DMG) has not been thoroughly evaluated. The optimal combination of multiparametric MRI and machine learning techniques remains undetermined. We compared the performance of various radiomics models across different MRI sequences and different machine learning techniques.

Methods

A total of 102 patients with pathologically confirmed DMG were retrospectively enrolled (27 with H3 K27M-mutant and 75 with H3 K27M wild-type). Radiomics features were extracted from eight sequences, and 18 feature sets were conducted by independent combination. There were three feature matrix normalization algorithms, two dimensionality-reduction methods, four feature selectors, and seven classifiers, consisting of 168 machine learning pipelines. Radiomics models were established across different feature sets and machine learning pipelines. The performance of models was evaluated using receiver operating characteristic curves with area under the curve (AUC) and compared with DeLong’s test.

Results

The multiparametric MRI-based radiomics models could accurately predict the H3 K27M mutant status in DMG (highest AUC: 0.807–0.969, for different sequences or sequence combinations). However, the results varied significantly between different machine learning techniques. When suitable machine learning techniques were used, the conventional MRI-based radiomics models shared similar performance to advanced MRI-based models (highest AUC: 0.875–0.915 vs. 0.807–0.926; DeLong’s test, p > 0.05). Most models had a better performance when generated with a combination of MRI sequences. The optimal model in the present study used a combination of all sequences (AUC = 0.969).

Conclusions

The multiparametric MRI-based radiomics models could be useful for predicting H3 K27M mutant status in DMG, but the performance varied across different sequences and machine learning techniques.

Application of Artificial Intelligence in Diagnosis of Craniopharyngioma

Craniopharyngioma is a congenital brain tumor with clinical characteristics of hypothalamic-pituitary dysfunction, increased intracranial pressure, and visual field disorder, among other injuries. Its clinical diagnosis mainly depends on radiological examinations (such as Computed Tomography, Magnetic Resonance Imaging). However, assessing numerous radiological images manually is a challenging task, and the experience of doctors has a great influence on the diagnosis result. The development of artificial intelligence has brought about a great transformation in the clinical diagnosis of craniopharyngioma. This study reviewed the application of artificial intelligence technology in the clinical diagnosis of craniopharyngioma from the aspects of differential classification, prediction of tissue invasion and gene mutation, prognosis prediction, and so on. Based on the reviews, the technical route of intelligent diagnosis based on the traditional machine learning model and deep learning model were further proposed. Additionally, in terms of the limitations and possibilities of the development of artificial intelligence in craniopharyngioma diagnosis, this study discussed the attentions required in future research, including few-shot learning, imbalanced data set, semi-supervised models, and multi-omics fusion.

Radiomics and radiogenomics in pediatric neuro-oncology: A review

The current era of advanced computing has allowed for the development and implementation of the field of radiomics. In pediatric neuro-oncology, radiomics has been applied in determination of tumor histology, identification of disseminated disease, prognostication, and molecular classification of tumors (ie, radiogenomics). The field also comes with many challenges, such as limitations in study sample sizes, class imbalance, generalizability of the methods, and data harmonization across imaging centers. The aim of this review paper is twofold: first, to summarize existing literature in radiomics of pediatric neuro-oncology; second, to distill the themes and challenges of the field and discuss future directions in both a clinical and technical context.

Magnetic Resonance Imaging Characteristics of Molecular Subgroups in Pediatric H3 K27M Mutant Diffuse Midline Glioma

Purpose

Recent research identified histone H3 K27M mutations to be associated with a dismal prognosis in pediatric diffuse midline glioma (pDMG); however, data on detailed MRI characteristics with respect to H3 K27 mutation status and molecular subgroups (H3.1 and H3.3 K27M mutations) are limited.

Methods

Standardized magnetic resonance imaging (MRI) parameters and epidemiologic data of 68 pDMG patients (age <18 years) were retrospectively reviewed and compared in a) H3 K27M mutant versus H3 K27 wildtype (WT) tumors and b) H3.1 versus H3.3 K27M mutant tumors.

Results

Intracranial gliomas (n = 58) showed heterogeneous phenotypes with isointense to hyperintense signal in T2-weighted images and frequent contrast enhancement. Hemorrhage and necrosis may be present. Comparing H3 K27M mutant to WT tumors, there were significant differences in the following parameters: i) tumor localization (p = 0.001), ii) T2 signal intensity (p = 0.021), and iii) T1 signal homogeneity (p = 0.02). No significant imaging differences were found in any parameter between H3.1 and H3.3 K27M mutant tumors; however, H3.1 mutant tumors occurred at a younger age (p = 0.004). Considering spinal gliomas (n = 10) there were no significant imaging differences between the analyzed molecular groups.

Conclusion

With this study, we are the first to provide detailed MR imaging data on H3 K27M mutant pDMG with respect to molecular subgroup status in a large patient cohort. Our findings may support diagnosis and future targeted therapeutic trials of pDMG within the framework of the radiogenomics concept.

Supplementary Information

The online version of this article (10.1007/s00062-021-01120-3) contains supplementary material, which is available to authorized users.