Prediction of H3K27M-mutant brainstem glioma by amide proton transfer–weighted imaging and its derived radiomics

Zhuo, Zhizheng; Qu, Liangsheng; Zhang, Peng; Duan, Yunyun; Cheng, Dan; Xu, Xiaolu; Sun, Ting; Ding, Jinli; Xie, Conghua; Liu, Xing; Haller, Sven; Barkhof, Frederik; Zhang, Liwei

doi:10.1007/s00259-021-05455-4

Cited by 29 publications

(16 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of note, the multiparametric MRI-based model with SVM, LDA, and LR demonstrated more favorable results. The prior study used various classifiers (e.g., SVM, RF, and XGBoost) and generated an AUC value of 0.549-0.953, which were lower than ours (AUC = 0.969) (11)(12)(13)(14)(15). Several reasons may account for this variety, including patient data, MRI data, and machine learning techniques.…”

Section: Discussionmentioning

confidence: 57%

“…The previous study focused on H3 K27M mutant status prediction, which rarely considered these elements. Two of them made an effort to compare the predictive power of different classifiers and another for different feature selectors ( 11 , 15 ). The limitation of these studies on model development warrants extra caution in terms of result explanation.…”

Section: Discussionmentioning

confidence: 99%

“…Several recent attempts have been made to use the multiparametric MRI-based radiomics model to predict H3 K27M mutant status, but the results varied greatly (11)(12)(13)(14)(15)(16). Most of them focused on different kinds of conventional MRI (cMRI), which could only reflect the tumor's morphologic information and benefit limitedly to reveal tumor heterogeneity.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

et al. 2022

View full text Add to dashboard Cite

ObjectivesThe 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.MethodsA 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.ResultsThe 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).ConclusionsThe 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.

show abstract

Section: Discussionmentioning

confidence: 57%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

et al. 2022

View full text Add to dashboard Cite

show abstract

“…First, the retrospective study design was subject to potential selection bias. Second, PCC threshold 0.9 has been used in previous studies to screen radiomics features ( 25 , 26 ). In this study, we used a threshold value lower than 0.9 to more rigorously filter out redundant features.…”

Section: Discussionmentioning

confidence: 99%

Diagnostic Performance of 2D and 3D T2WI-Based Radiomics Features With Machine Learning Algorithms to Distinguish Solid Solitary Pulmonary Lesion

et al. 2021

View full text Add to dashboard Cite

ObjectiveTo evaluate the performance of 2D and 3D radiomics features with different machine learning approaches to classify SPLs based on magnetic resonance(MR) T2 weighted imaging (T2WI).Material and MethodsA total of 132 patients with pathologically confirmed SPLs were examined and randomly divided into training (n = 92) and test datasets (n = 40). A total of 1692 3D and 1231 2D radiomics features per patient were extracted. Both radiomics features and clinical data were evaluated. A total of 1260 classification models, comprising 3 normalization methods, 2 dimension reduction algorithms, 3 feature selection methods, and 10 classifiers with 7 different feature numbers (confined to 3–9), were compared. The ten-fold cross-validation on the training dataset was applied to choose the candidate final model. The area under the receiver operating characteristic curve (AUC), precision-recall plot, and Matthews Correlation Coefficient were used to evaluate the performance of machine learning approaches.ResultsThe 3D features were significantly superior to 2D features, showing much more machine learning combinations with AUC greater than 0.7 in both validation and test groups (129 vs. 11). The feature selection method Analysis of Variance(ANOVA), Recursive Feature Elimination(RFE) and the classifier Logistic Regression(LR), Linear Discriminant Analysis(LDA), Support Vector Machine(SVM), Gaussian Process(GP) had relatively better performance. The best performance of 3D radiomics features in the test dataset (AUC = 0.824, AUC-PR = 0.927, MCC = 0.514) was higher than that of 2D features (AUC = 0.740, AUC-PR = 0.846, MCC = 0.404). The joint 3D and 2D features (AUC=0.813, AUC-PR = 0.926, MCC = 0.563) showed similar results as 3D features. Incorporating clinical features with 3D and 2D radiomics features slightly improved the AUC to 0.836 (AUC-PR = 0.918, MCC = 0.620) and 0.780 (AUC-PR = 0.900, MCC = 0.574), respectively.ConclusionsAfter algorithm optimization, 2D feature-based radiomics models yield favorable results in differentiating malignant and benign SPLs, but 3D features are still preferred because of the availability of more machine learning algorithmic combinations with better performance. Feature selection methods ANOVA and RFE, and classifier LR, LDA, SVM and GP are more likely to demonstrate better diagnostic performance for 3D features in the current study.

show abstract

“…It can be applied to study alterations in proteins, peptides, cellularity, proliferation, necrosis, IDH and MGMT, and metabolites. In addition to APT and NOE, other CEST contrasts have been explored to further study the underlying molecular alterations, including the regional alterations, in brain tumors, especially with the aid of radiomics [6,12].…”

Section: Multiple Cest Contrast In Brain Tumorsmentioning

confidence: 99%

Molecular Imaging of Brain Tumors and Drug Delivery Using CEST MRI: Promises and Challenges

et al. 2022

View full text Add to dashboard Cite

Chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI) detects molecules in their natural forms in a sensitive and non-invasive manner. This makes it a robust approach to assess brain tumors and related molecular alterations using endogenous molecules, such as proteins/peptides, and drugs approved for clinical use. In this review, we will discuss the promises of CEST MRI in the identification of tumors, tumor grading, detecting molecular alterations related to isocitrate dehydrogenase (IDH) and O-6-methylguanine-DNA methyltransferase (MGMT), assessment of treatment effects, and using multiple contrasts of CEST to develop theranostic approaches for cancer treatments. Promising applications include (i) using the CEST contrast of amide protons of proteins/peptides to detect brain tumors, such as glioblastoma multiforme (GBM) and low-grade gliomas; (ii) using multiple CEST contrasts for tumor stratification, and (iii) evaluation of the efficacy of drug delivery without the need of metallic or radioactive labels. These promising applications have raised enthusiasm, however, the use of CEST MRI is not trivial. CEST contrast depends on the pulse sequences, saturation parameters, methods used to analyze the CEST spectrum (i.e., Z-spectrum), and, importantly, how to interpret changes in CEST contrast and related molecular alterations in the brain. Emerging pulse sequence designs and data analysis approaches, including those assisted with deep learning, have enhanced the capability of CEST MRI in detecting molecules in brain tumors. CEST has become a specific marker for tumor grading and has the potential for prognosis and theranostics in brain tumors. With increasing understanding of the technical aspects and associated molecular alterations detected by CEST MRI, this young field is expected to have wide clinical applications in the near future.

show abstract

Prediction of H3K27M-mutant brainstem glioma by amide proton transfer–weighted imaging and its derived radiomics

Cited by 29 publications

References 33 publications

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

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

Diagnostic Performance of 2D and 3D T2WI-Based Radiomics Features With Machine Learning Algorithms to Distinguish Solid Solitary Pulmonary Lesion

Molecular Imaging of Brain Tumors and Drug Delivery Using CEST MRI: Promises and Challenges

Contact Info

Product

Resources

About