Radiomics and Qualitative Features From Multiparametric MRI Predict Molecular Subtypes in Patients With Lower-Grade Glioma

MRI-based machine learning reveals proteasome subunit PSMB8-mediated malignant glioma phenotypes through activating TGFBR1/2-SMAD2/3 axis

Gliomas are the most prevalent and aggressive neoplasms of the central nervous system, representing a major challenge for effective treatment and patient prognosis. This study identifies the proteasome subunit beta type-8 (PSMB8/LMP7) as a promising prognostic biomarker for glioma. Using a multiparametric radiomic model derived from preoperative magnetic resonance imaging (MRI), we accurately predicted PSMB8 expression levels. Notably, radiomic prediction of poor prognosis was highly consistent with elevated PSMB8 expression. Our findings demonstrate that PSMB8 depletion not only suppressed glioma cell proliferation and migration but also induced apoptosis via activation of the transforming growth factor beta (TGF-β) signaling pathway. This was supported by downregulation of key receptors (TGFBR1 and TGFBR2). Furthermore, interference with PSMB8 expression impaired phosphorylation and nuclear translocation of SMAD2/3, critical mediators of TGF-β signaling. Consequently, these molecular alterations resulted in reduced tumor progression and enhanced sensitivity to temozolomide (TMZ), a standard chemotherapeutic agent. Overall, our findings highlight PSMB8's pivotal role in glioma pathophysiology and its potential as a prognostic marker. This study also demonstrates the clinical utility of MRI radiomics for preoperative risk stratification and pre-diagnosis. Targeted inhibition of PSMB8 may represent a therapeutic strategy to overcome TMZ resistance and improve glioma patient outcomes.

MRI-based habitat imaging predicts high-risk molecular subtypes and early risk assessment of lower-grade gliomas

BACKGROUND

In lower-grade gliomas (LrGGs, histological grades 2-3), there exist a minority of high-risk molecular subtypes with malignant transformation potential, associated with unfavorable clinical outcomes and shorter survival prognosis. Identifying high-risk molecular subtypes early in LrGGs and conducting preoperative prognostic evaluations are crucial for precise clinical diagnosis and treatment.

MATERIALS AND METHODS

We retrospectively collected data from 345 patients with LrGGs and comprehensively screened key high-risk molecular markers. Based on preoperative MRI sequences (CE-T1WI/T2-FLAIR), we employed seven classifiers to construct models based on habitat, radiomics, and combined. Eventually, we identified Extra Trees based on habitat features as the optimal predictive model for identifying high-risk molecular subtypes of LrGGs. Moreover, we developed a prognostic prediction model based on radiomics score (Radscore) to assess the survival outlook of patients with LrGGs. We utilized Kaplan-Meier (KM) survival analysis alongside the log-rank test to discern variations in survival probabilities among high-risk and low-risk cohorts. The concordance index was employed to gauge the efficacy of habitat, clinical, and amalgamated prognosis models. Calibration curves were utilized to appraise the congruence between the anticipated survival probability and the actual survival probability projected by the models.

RESULTS

The habitat model for predicting high-risk molecular subtypes of LrGGs, achieved AUCs of 0.802, 0.771, and 0.768 in the training set, internal test set, and external test set, respectively. Comparison among habitat, clinical, combined prognostic models revealed that the combined prognostic model exhibited the highest performance (C-index = 0.781 in the training set, C-index = 0.778 in the internal test set, C-index = 0.743 in the external test set), followed by the habitat prognostic model (C-index = 0.749 in the training set, C-index = 0.716 in the internal test set, C-index = 0.707 in the external test set), while the clinical prognostic model performed the worst (C-index = 0.717 in the training set, C-index = 0.687 in the internal test set, C-index = 0.649 in the external test set). Furthermore, the calibration curves of the combined model exhibited satisfactory alignment when forecasting the 1-year, 2-year, and 3-year survival probabilities of patients with LrGGs.

CONCLUSION

The MRI-based habitat model simultaneously achieves the objectives of non-invasive prediction of high-risk molecular subtypes of LrGGs and assessment of survival prognosis. This has incremental value for early non-invasive warning of malignant transformation in LrGGs and risk-stratified management.

Utilizing machine-learning techniques on MRI radiomics to identify primary tumors in brain metastases

Objective

To develop a machine learning-based clinical and/or radiomics model for predicting the primary site of brain metastases using multiparametric magnetic resonance imaging (MRI).

Materials and methods

A total of 202 patients (87 males, 115 females) with 439 brain metastases were retrospectively included, divided into training sets (brain metastases of lung cancer [BMLC] n = 194, brain metastases of breast cancer [BMBC] n = 108, brain metastases of gastrointestinal tumor [BMGiT] n = 48) and test sets (BMLC n = 50, BMBC n = 27, BMGiT n = 12). A total of 3,404 quantitative image features were obtained through semi-automatic segmentation from MRI images (T1WI, T2WI, FLAIR, and T1-CE). Intra-class correlation coefficient (ICC) was used to examine segmentation stability between two radiologists. Radiomics features were selected using analysis of variance (ANOVA), recursive feature elimination (RFE), and Kruskal-Wallis test. Three machine learning classifiers were used to build the radiomics model, which was validated using five-fold cross-validation on the training set. A comprehensive model combining radiomics and clinical features was established, and the diagnostic performance was compared by area under the curve (AUC) and evaluated in an independent test set.

Results

The radiomics model differentiated BMGiT from BMLC (13 features, AUC = 0.915 ± 0.071) or BMBC (20 features, AUC = 0.954 ± 0.064) with high accuracy, while the classification between BMLC and BMBC was unsatisfactory (11 features, AUC = 0.729 ± 0.114). However, the combined model incorporating radiomics and clinical features improved the predictive performance, with AUC values of 0.965 for BMLC vs. BMBC, 0.991 for BMLC vs. BMGiT, and 0.935 for BMBC vs. BMGiT.

Conclusion

The machine learning-based radiomics model demonstrates significant potential in distinguishing the primary sites of brain metastases, and may assist screening of primary tumor when brain metastasis is suspected whereas history of primary tumor is absent.

Predictive machine learning models based on VASARI features for WHO grading, isocitrate dehydrogenase mutation, and 1p19q co-deletion status: a multicenter study

The objective of our study was to develop predictive models using Visually Accessible Rembrandt Images (VASARI) magnetic resonance imaging (MRI) features combined with machine learning techniques to predict the World Health Organization (WHO) grade, isocitrate dehydrogenase (IDH) mutation status, and 1p19q co-deletion status of high-grade gliomas. To achieve this, we retrospectively included 485 patients with high-grade glioma from the First Affiliated Hospital of Xinjiang Medical University, of which 312 patients were randomly divided into a training set (n=218) and a test set (n=94) in a 7:3 ratio. Twenty-five VASARI MRI features were selected from an initial set of 30, and three machine learning models - Multilayer Perceptron (MP), Bernoulli Naive Bayes (BNB), and Logistic Regression (LR) - were trained using the training set. The most informative features were identified using recursive feature elimination. Model performance was assessed using the test set and an independent validation set of 173 patients from Beijing Tiantan Hospital. The results indicated that the MP model exhibited the highest predictive accuracy on the training set, achieving an area under the curve (AUC) close to 1, indicating perfect discrimination. However, its performance decreased in the test and validation sets; particularly for predicting the 1p19q co-deletion status, the AUC was only 0.703, suggesting potential overfitting. On the other hand, the BNB model demonstrated robust generalization on the test and validation sets, with AUC values of 0.8292 and 0.8106, respectively, for predicting IDH mutation status and 1p19q co-deletion status, indicating high accuracy, sensitivity, and specificity. The LR model also showed good performance with AUCs of 0.7845 and 0.8674 on the test and validation sets, respectively, for predicting IDH mutation status, although it was slightly inferior to the BNB model for the 1p19q co-deletion status. In conclusion, integrating VASARI MRI features with machine learning techniques shows promise for the non-invasive prediction of glioma molecular markers, which could guide treatment strategies and improve prognosis in glioma patients. Nonetheless, further model optimization and validation are necessary to enhance its clinical utility.

Ten Years of VASARI Glioma Features: Systematic Review and Meta-Analysis of Their Impact and Performance

BACKGROUND

Visually Accessible Rembrandt (Repository for Molecular Brain Neoplasia Data) Images (VASARI) features, a vocabulary to establish reproducible terminology for glioma reporting, have been applied for a decade, but a systematic performance evaluation is lacking.

PURPOSE

Our aim was to conduct a systematic review and meta-analysis of the performance of the VASARI features set for glioma assessment.

DATA SOURCES

MEDLINE, Web of Science, EMBASE, and the Cochrane Library were systematically searched until September 26, 2023.

STUDY SELECTION

Original articles predicting diagnosis, progression, and survival in patients with glioma were included.

DATA ANALYSIS

The modified Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was applied to evaluate the risk-of-bias. The meta-analysis used a random effects model and forest plot visualizations, if ≥5 comparable studies with a low or medium risk of bias were provided.

DATA SYNTHESIS

Thirty-five studies (3304 patients) were included. Risk-of-bias scores were medium (n = 33) and low (n = 2). Recurring objectives were overall survival (n = 18) and isocitrate dehydrogenase mutation (IDH; n = 12) prediction. Progression-free survival was examined in 7 studies. In 4 studies (glioblastoma n = 2, grade 2/3 glioma n = 1, grade 3 glioma n = 1), a significant association was found between progression-free survival and single VASARI features. The single features predicting overall survival with the highest pooled hazard ratios were multifocality (hazard ratio = 1.80; 95%-CI, 1.21-2.67; I2 = 53%), ependymal invasion (hazard ratio = 1.73; 95% CI, 1.45-2.05; I2 = 0%), and enhancing tumor crossing the midline (hazard ratio = 2.08; 95% CI, 1.35-3.18; I2 = 52%). IDH mutation-predicting models combining VASARI features rendered a pooled area under the receiver operating characteristic curve of 0.82 (95% CI, 0.76-0.88) at considerable heterogeneity (I2 = 100%). Combined input models using VASARI plus clinical and/or radiomics features outperformed single data-type models in all relevant studies (n = 17).

LIMITATIONS

Studies were heterogeneously designed and often with a small sample size. Several studies used The Cancer Imaging Archive database, with likely overlapping cohorts. The meta-analysis for IDH was limited due to a high study heterogeneity.

CONCLUSIONS

Some VASARI features perform well in predicting overall survival and IDH mutation status, but combined models outperform single features. More studies with less heterogeneity are needed to increase the evidence level.

Diffusion kurtosis imaging-based habitat analysis identifies high-risk molecular subtypes and heterogeneity matching in diffuse gliomas

OBJECTIVE

High-risk types of diffuse gliomas in adults include isocitrate dehydrogenase (IDH) wild-type glioblastomas and grade 4 astrocytomas. Achieving noninvasive prediction of high-risk molecular subtypes of gliomas is important for personalized and precise diagnosis and treatment.

METHODS

We retrospectively collected data from 116 patients diagnosed with adult diffuse gliomas. Multiple high-risk molecular markers were tested, and various habitat models and whole-tumor models were constructed based on preoperative routine and diffusion kurtosis imaging (DKI) sequences to predict high-risk molecular subtypes of gliomas. Feature selection and model construction utilized Least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM). Finally, the Wilcoxon rank-sum test was employed to explore the correlation between habitat quantitative features (intra-tumor heterogeneity score，ITH score) and heterogeneity, as well as high-risk molecular subtypes.

RESULTS

The results showed that the habitat analysis model based on DKI performed remarkably well (with AUC values reaching 0.977 and 0.902 in the training and test sets, respectively). The model's performance was further enhanced when combined with clinical variables. (The AUC values were 0.994 and 0.920, respectively.) Additionally, we found a close correlation between ITH score and heterogeneity, with statistically significant differences observed between high-risk and non-high-risk molecular subtypes.

INTERPRETATION

The habitat model based on DKI is an ideal means for preoperatively predicting high-risk molecular subtypes of gliomas, holding significant value for noninvasively alerting malignant gliomas and those with malignant transformation potential.

The T2-FLAIR mismatch sign in oncologic neuroradiology: History, current use, emerging data, and future directions

The T2-Fluid-Attenuated Inversion Recovery (T2-FLAIR) mismatch sign is a radiogenomic marker that is easily discernible on preoperative conventional MR imaging. Application of strict criteria (adult population, cerebral hemisphere location, and classic imaging morphology) permits the noninvasive preoperative diagnosis of isocitrate dehydrogenase (IDH)-mutant 1p/19q-non-codeleted diffuse astrocytoma with near-perfect specificity, albeit with variably low sensitivity. This leads to improved preoperative planning and patient counseling. More recent research has shown that the application of less strict criteria compromises the near-perfect specificity of the sign but remains adequate for ruling out IDH-wildtype (glioblastoma) phenotype, which bears a far grimmer prognosis compared to IDH-mutant diffuse astrocytic disease. In this review, we elaborate on the various definitions of the T2-FLAIR mismatch sign present in the literature, illustrate these with images obtained at a comprehensive cancer center, discuss the potential of the mismatch sign for application to certain pediatric-type brain tumors, namely dysembryoplastic neuroepithelial tumor and diffuse midline glioma, and elaborate upon the clinical, histologic, and molecular associations of the T2-FLAIR mismatch sign as recognized to date. Finally, the sign's correlates in diffusion- and perfusion-weighted imaging are presented, and opportunities to further maximize the diagnostic and prognostic applications of the sign in the context of the 2021 revision of the WHO Classification of Central Nervous System Tumors are discussed.

Novel Imaging Approaches for Glioma Classification in the Era of the World Health Organization 2021 Update: A Scoping Review

The 2021 WHO classification of CNS tumors is a challenge for neuroradiologists due to the central role of the molecular profile of tumors. The potential of novel data analysis tools in neuroimaging must be harnessed to maintain its role in predicting tumor subgroups. We performed a scoping review to determine current evidence and research gaps. A comprehensive literature search was conducted regarding glioma subgroups according to the 2021 WHO classification and the use of MRI, radiomics, machine learning, and deep learning algorithms. Sixty-two original articles were included and analyzed by extracting data on the study design and results. Only 8% of the studies included pediatric patients. Low-grade gliomas and diffuse midline gliomas were represented in one-third of the research papers. Public datasets were utilized in 22% of the studies. Conventional imaging sequences prevailed; data on functional MRI (DWI, PWI, CEST, etc.) are underrepresented. Multiparametric MRI yielded the best prediction results. IDH mutation and 1p/19q codeletion status prediction remain in focus with limited data on other molecular subgroups. Reported AUC values range from 0.6 to 0.98. Studies designed to assess generalizability are scarce. Performance is worse for smaller subgroups (e.g., 1p/19q codeleted or IDH1/2 mutated gliomas). More high-quality study designs with diversity in the analyzed population and techniques are needed.

T2-FLAIR mismatch sign and machine learning-based multiparametric MRI radiomics in predicting IDH mutant 1p/19q non-co-deleted diffuse lower-grade gliomas

AIM

To investigate the application of the T2-weighted (T2)-fluid-attenuated inversion recovery (FLAIR) mismatch sign and machine learning-based multiparametric magnetic resonance imaging (MRI) radiomics in predicting 1p/19q non-co-deletion of lower-grade gliomas (LGGs).

MATERIALS AND METHODS

One hundred and forty-six patients, who had pathologically confirmed isocitrate dehydrogenase (IDH) mutant LGGs were assigned randomly to the training cohort (n=102) and the testing cohort (n=44) at a ratio of 7:3. The T2-FLAIR mismatch sign and conventional MRI features were evaluated. Radiomics features extracted from T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), FLAIR, apparent diffusion coefficient (ADC), and contrast-enhanced T1WI images (CE-T1WI). The models that displayed the best performance of each sequence were selected, and their predicted values as well as the T2-FLAIR mismatch sign data were collected to establish a final stacking model. Receiver operating characteristic curve (ROC) analyses and area under the curve (AUC) values were applied to evaluate and compare the performance of the models.

RESULTS

The T2-FLAIR mismatch sign was more common in the IDH mutant 1p/19q non-co-deleted group (p<0.05) and the area under the curve (AUC) value was 0.692 with sensitivity 0.397, specificity 0.987, and accuracy 0.712, respectively. The stacking model showed a favourable performance with an AUC of 0.925 and accuracy of 0.882 in the training cohort and an AUC of 0.886 and accuracy of 0.864 in the testing cohort.

CONCLUSION

The stacking model based on multiparametric MRI can serve as a supplementary tool for pathological diagnosis, offering valuable guidance for clinical practice.

Amide proton transfer weighted and diffusion weighted imaging based radiomics classification algorithm for predicting 1p/19q co-deletion status in low grade gliomas

BACKGROUND

1p/19q co-deletion in low-grade gliomas (LGG, World Health Organization grade II and III) is of great significance in clinical decision making. We aim to use radiomics analysis to predict 1p/19q co-deletion in LGG based on amide proton transfer weighted (APTw), diffusion weighted imaging (DWI), and conventional MRI.

METHODS

This retrospective study included 90 patients histopathologically diagnosed with LGG. We performed a radiomics analysis by extracting 8454 MRI-based features form APTw, DWI and conventional MR images and applied a least absolute shrinkage and selection operator (LASSO) algorithm to select radiomics signature. A radiomics score (Rad-score) was generated using a linear combination of the values of the selected features weighted for each of the patients. Three neuroradiologists, including one experienced neuroradiologist and two resident physicians, independently evaluated the MR features of LGG and provided predictions on whether the tumor had 1p/19q co-deletion or 1p/19q intact status. A clinical model was then constructed based on the significant variables identified in this analysis. A combined model incorporating both the Rad-score and clinical factors was also constructed. The predictive performance was validated by receiver operating characteristic curve analysis, DeLong analysis and decision curve analysis. P < 0.05 was statistically significant.

RESULTS

The radiomics model and the combined model both exhibited excellent performance on both the training and test sets, achieving areas under the curve (AUCs) of 0.948 and 0.966, as well as 0.909 and 0.896, respectively. These results surpassed the performance of the clinical model, which achieved AUCs of 0.760 and 0.766 on the training and test sets, respectively. After performing Delong analysis, the clinical model did not significantly differ in predictive performance from three neuroradiologists. In the training set, both the radiomic and combined models performed better than all neuroradiologists. In the test set, the models exhibited higher AUCs than the neuroradiologists, with the radiomics model significantly outperforming resident physicians B and C, but not differing significantly from experienced neuroradiologist.

CONCLUSIONS

Our results suggest that our algorithm can noninvasively predict the 1p/19q co-deletion status of LGG. The predictive performance of radiomics model was comparable to that of experienced neuroradiologist, significantly outperforming the diagnostic accuracy of resident physicians, thereby offering the potential to facilitate non-invasive 1p/19q co-deletion prediction of LGG.

Advances in Molecular Pathology, Diagnosis and Treatment of Spinal Cord Astrocytomas

Spinal cord astrocytoma (SCA) is a rare subtype of astrocytoma, posing challenges in diagnosis and treatment. Low-grade SCA can achieve long-term survival solely through surgery, while high-grade has a disappointing prognosis even with comprehensive treatment. Diagnostic criteria and standard treatment of intracranial astrocytoma have shown obvious limitations in SCA. Research on the molecular mechanism in SCA is lagging far behind that on intracranial astrocytoma. In recent years, huge breakthroughs have been made in molecular pathology of astrocytoma, and novel techniques have emerged, including DNA methylation analysis and radiomics. These advances are now making it possible to provide a precise diagnosis and develop corresponding treatment strategies in SCA. Our aim is to review the current status of diagnosis and treatment of SCA, and summarize the latest research advancement, including tumor subtype, molecular characteristics, diagnostic technology, and potential therapy strategies, thus deepening our understanding of this uncommon tumor type and providing guidance for accurate diagnosis and treatment.

Evaluation of glial tumors: correlation between magnetic resonance imaging and histopathological analysis

Objective

To determine the correlation of conventional and diffusion-weighted imaging findings on magnetic resonance imaging (MRI) of the brain, based on Visually AcceSAble Rembrandt Images (VASARI) criteria, with the histopathological grading of gliomas: low-grade or high-grade.

Materials and Methods

Preoperative MRI scans of 178 patients with brain gliomas and pathological confirmation were rated by two neuroradiologists for tumor size, location, and tumor morphology, using a standardized imaging feature set based on the VASARI criteria.

Results

In the univariate analysis, more than half of the MRI characteristics evaluated showed a significant association with the tumor grade. The characteristics most significantly associated with the tumor grade were hemorrhage; restricted diffusion; pial invasion; enhancement; and a non-contrast-enhancing tumor crossing the midline. In a multivariable regression model, the presence of enhancement and hemorrhage maintained a significant association with high tumor grade. The absence of contrast enhancement and restricted diffusion were associated with the presence of an isocitrate dehydrogenase gene mutation.

Conclusion

Our data illustrate that VASARI MRI features, especially intratumoral hemorrhage, contrast enhancement, and multicentricity, correlate strongly with glial tumor grade.

Artificial Intelligence Applications in Glioma With 1p/19q Co-Deletion: A Systematic Review

As an important genomic marker for oligodendrogliomas, early determination of 1p/19q co-deletion status is critical for guiding therapy and predicting prognosis in patients with glioma. The purpose of this study is to systematically review the literature concerning the magnetic resonance imaging (MRI) with artificial intelligence (AI) methods for predicting 1p/19q co-deletion status in glioma. PubMed, Scopus, Embase, and IEEE Xplore were searched in accordance with the Preferred Reporting Items for systematic reviews and meta-analyses guidelines. Methodological quality of studies was assessed according to the Quality Assessment of Diagnostic Accuracy Studies-2. Finally, 28 studies were included in the quantitative analysis. Diagnostic test accuracy reached an area under the ROC curve of 0.71-0.98 were reported in 24 studies. The remaining four studies with no available AUC provided an accuracy of 0.75-0. 89. The included studies varied widely in terms of imaging sequences, input features, and modeling methods. The current review highlighted that integrating MRI with AI technology is a potential tool for determination 1p/19q status pre-operatively and noninvasively, which can possibly help clinical decision-making. However, the reliability and feasibility of this approach still need to be further validated and improved in a real clinical setting. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: 2.

Machine learning models based on multi-parameter MRI radiomics for prediction of molecular glioblastoma: a new study based on the 2021 World Health Organization classification

BACKGROUND

The 2021 World Health Organization (WHO) classification considers a histological low grade glioma with specific molecular characteristics as molecular glioblastoma (mGBM). Accurate identification of mGBM will aid in risk stratification of glioma patients.

PURPOSE

To explore the value of machine learning models based on magnetic resonance imaging (MRI) radiomics features in predicting mGBM.

MATERIAL AND METHODS

In total, 166 patients histologically diagnosed as low-grade diffuse glioma (WHO II and III) were included in the study. Fifty-three cases were reclassified as mGBM based on molecular status. Four dimensionality reduction methods including distance correlation (DC), gradient boosted decision tree (GBDT), least absolute shrinkage and selection operator (LASSO) and minimal redundancy maximal relevance (MRMR) were used to select the optimal signatures. Six machine learning algorithms including support vector machine (SVM), linear discriminant analysis (LDA), neural network (NN), logistic regression (LR), K-nearest neighbour (KNN) and decision tree (DT) were used to develop the classifiers. The relative SD was used to evaluate the stability of the models, and the area under the curve values in the independent test group were used to evaluate their performances.

RESULTS

NN_DC was determined as the optimal classifier due to the highest area under the curve of 0.891 in the test group. The classification accuracy, sensitivity, specificity, positive predictive value and negative predictive value of NN_DC were 0.915, 0.842, 0.950, 0.889 and 0.927, respectively.

CONCLUSION

Machine learning models can predict mGBM non-invasively, which may help to develop personalized treatment strategies for neurosurgeons and provide an effective tool for accurate stratification in clinical trials.

Improving Noninvasive Classification of Molecular Subtypes of Adult Gliomas With Diffusion-Weighted MR Imaging: An Externally Validated Machine Learning Algorithm

BACKGROUND

Genetic testing for molecular markers of gliomas sometimes is unavailable because of time-consuming and expensive, even limited tumor specimens or nonsurgery cases.

PURPOSE

To train a three-class radiomic model classifying three molecular subtypes including isocitrate dehydrogenase (IDH) mutations and 1p/19q-noncodeleted (IDHmut-noncodel), IDH wild-type (IDHwt), IDH-mutant and 1p/19q-codeleted (IDHmut-codel) of adult gliomas and investigate whether radiomic features from diffusion-weighted imaging (DWI) could bring additive value.

STUDY TYPE

Retrospective.

POPULATION

A total of 755 patients including 111 IDHmut-noncodel, 571 IDHwt, and 73 IDHmut-codel cases were divided into training (n = 480) and internal validation set (n = 275); 139 patients including 21 IDHmut-noncodel, 104 IDHwt, and 14 IDHmut-codel cases were utilized as external validation set.

FIELD STRENGTH/SEQUENCE

A 1.5 T or 3.0 T/multiparametric MRI, including T1-weighted (T1), T1-weighted gadolinium contrast-enhanced (T1c), T2-weighted (T2), fluid attenuated inversion recovery (FLAIR), and DWI.

ASSESSMENT

The performance of multiparametric radiomic model (random-forest model) using 22 selected features from T1, T2, FLAIR, T1c images and apparent diffusion coefficient (ADC) maps, and conventional radiomic model using 20 selected features from T1, T2, FLAIR, and T1c images was assessed in internal and external validation sets by comparing probability values and actual incidence.

STATISTICAL TESTS

Mann-Whitney U test, Chi-Squared test, Wilcoxon test, receiver operating curve (ROC), and area under the curve (AUC); DeLong analysis. P < 0.05 was statistically significant.

RESULTS

The multiparametric radiomic model achieved AUC values for IDHmut-noncodel, IDHwt, and IDHmut-codel of 0.8181, 0.8524, and 0.8502 in internal validation set and 0.7571, 0.7779, and 0.7491 in external validation set, respectively. Multiparametric radiomic model showed significantly better diagnostic performance after DeLong analysis, especially in classifying IDHwt and IDHmut-noncodel subtypes.

DATA CONCLUSION

Radiomic features from DWI could bring additive value and improve the performance of conventional MRI-based radiomic model for classifying the molecular subtypes especially IDHmut-noncodel and IDHwt of adult gliomas.

LEVEL OF EVIDENCE: 3

TECHNICAL EFFICACY

Stage 2.

Research on application of radiomics in glioma: a bibliometric and visual analysis

Background

With the continuous development of medical imaging informatics technology, radiomics has become a new and evolving field in medical applications. Radiomics aims to be an aid to support clinical decision making by extracting quantitative features from medical images and has a very wide range of applications. The purpose of this study was to perform a bibliometric and visual analysis of scientific results and research trends in the research application of radiomics in glioma.

Methods

We searched the Web of Science Core Collection (WOScc) for publications related to glioma radiomics. A bibliometric and visual analysis of online publications in this field related to countries/regions, authors, journals, references and keywords was performed using CiteSpace and R software.

Results

A total of 587 relevant literature published from 2012 to September 2022 were retrieved in WOScc, and finally a total of 484 publications were obtained according to the filtering criteria, including 393 (81.20%) articles and 91 (18.80%) reviews. The number of relevant publications increases year by year. The highest number of publications was from the USA (171 articles, 35.33%) and China (170 articles, 35.12%). The research institution with the highest number of publications was Chinese Acad Sci (24), followed by Univ Penn (22) and Fudan Univ (21). WANG Y (27) had the most publications, followed by LI Y (22), and WANG J (20). Among the 555 co-cited authors, LOUIS DN (207) and KICKINGEREDER P (207) were the most cited authors. FRONTIERS IN ONCOLOGY (42) was the most published journal and NEURO-ONCOLOGY (412) was the most co-cited journal. The most frequent keywords in all publications included glioblastoma (187), survival (136), classification (131), magnetic resonance imaging (113), machine learning (100), tumor (82), and feature (79), central nervous system (66), IDH (57), and radiomics (55). Cluster analysis was performed on the basis of keyword co-occurrence, and a total of 16 clusters were formed, indicating that these directions are the current hotspots of radiomics research applications in glioma and may be the future directions of continuous development.

Conclusion

In the past decade, radiomics has received much attention in the medical field and has been widely used in clinical research applications. Cooperation and communication between countries/regions need to be enhanced in future research to promote the development of radiomics in the field of medicine. In addition, the application of radiomics has improved the accuracy of pre-treatment diagnosis, efficacy prediction and prognosis assessment of glioma and helped to promote the development into precision medicine, the future still faces many challenges.

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.

Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma

Determining the optimal course of treatment for low grade glioma (LGG) patients is challenging and frequently reliant on subjective judgment and limited scientific evidence. Our objective was to develop a comprehensive deep learning assisted radiomics model for assessing not only overall survival in LGG, but also the likelihood of future malignancy and glioma growth velocity. Thus, we retrospectively included 349 LGG patients to develop a prediction model using clinical, anatomical, and preoperative MRI data. Before performing radiomics analysis, a U2-model for glioma segmentation was utilized to prevent bias, yielding a mean whole tumor Dice score of 0.837. Overall survival and time to malignancy were estimated using Cox proportional hazard models. In a postoperative model, we derived a C-index of 0.82 (CI 0.79-0.86) for the training cohort over 10 years and 0.74 (Cl 0.64-0.84) for the test cohort. Preoperative models showed a C-index of 0.77 (Cl 0.73-0.82) for training and 0.67 (Cl 0.57-0.80) test sets. Our findings suggest that we can reliably predict the survival of a heterogeneous population of glioma patients in both preoperative and postoperative scenarios. Further, we demonstrate the utility of radiomics in predicting biological tumor activity, such as the time to malignancy and the LGG growth rate.

Vasari Scoring System in Discerning between Different Degrees of Glioma and IDH Status Prediction: A Possible Machine Learning Application?

(1) The aim of our study is to evaluate the capacity of the Visually AcceSAble Rembrandt Images (VASARI) scoring system in discerning between the different degrees of glioma and Isocitrate Dehydrogenase (IDH) status predictions, with a possible application in machine learning. (2) A retrospective study was conducted on 126 patients with gliomas (M/F = 75/51; mean age: 55.30), from which we obtained their histological grade and molecular status. Each patient was analyzed with all 25 features of VASARI, blinded by two residents and three neuroradiologists. The interobserver agreement was assessed. A statistical analysis was conducted to evaluate the distribution of the observations using a box plot and a bar plot. We then performed univariate and multivariate logistic regressions and a Wald test. We also calculated the odds ratios and confidence intervals for each variable and the evaluation matrices with receiver operating characteristic (ROC) curves in order to identify cut-off values that are predictive of a diagnosis. Finally, we did the Pearson correlation test to see if the variables grade and IDH were correlated. (3) An excellent ICC estimate was obtained. For the grade and IDH status prediction, there were statistically significant results by evaluation of the degree of post-contrast impregnation (F4) and the percentage of impregnated area (F5), not impregnated area (F6), and necrotic (F7) tissue. These models showed good performances according to the area under the curve (AUC) values (>70%). (4) Specific MRI features can be used to predict the grade and IDH status of gliomas, with important prognostic implications. The standardization and improvement of these data (aim: AUC > 80%) can be used for programming machine learning software.

The combination of radiomics features and VASARI standard to predict glioma grade

Background and Purpose

Radiomics features and The Visually AcceSAble Rembrandt Images (VASARI) standard appear to be quantitative and qualitative evaluations utilized to determine glioma grade. This study developed a preoperative model to predict glioma grade and improve the efficacy of clinical strategies by combining these two assessment methods.

Materials and Methods

Patients diagnosed with glioma between March 2017 and September 2018 who underwent surgery and histopathology were enrolled in this study. A total of 3840 radiomic features were calculated; however, using the least absolute shrinkage and selection operator (LASSO) method, only 16 features were chosen to generate a radiomic signature. Three predictive models were developed using radiomic features and VASARI standard. The performance and validity of models were evaluated using decision curve analysis and 10-fold nested cross-validation.

Results

Our study included 102 patients: 35 with low-grade glioma (LGG) and 67 with high-grade glioma (HGG). Model 1 utilized both radiomics and the VASARI standard, which included radiomic signatures, proportion of edema, and deep white matter invasion. Models 2 and 3 were constructed with radiomics or VASARI, respectively, with an area under the receiver operating characteristic curve (AUC) of 0.937 and 0.831, respectively, which was less than that of Model 1, with an AUC of 0.966.

Conclusion

The combination of radiomics features and the VASARI standard is a robust model for predicting glioma grades.