Prediction of IDH1 Mutation Status in Glioblastoma Using Machine Learning Technique Based on Quantitative Radiomic Data

Lee, Min Ho; Kim, Junhyung; Kim, Sung-Tae; Shin, Hong-Jae; You, Hyejin; Choi, Jung Won; Seol, Ho Jun; Nam, Do‐Hyun; Lee, Jung-Il; Kong, Doo‐Sik

doi:10.1016/j.wneu.2019.01.157

Cited by 34 publications

(28 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Also emerging AI algorithms can provide the voxel-based assessment of data obtained by histogram analysis or other methods. The voxel-based assessment of imaging data provides new quantitative information, which is invisible to human assessment and can more precisely extract and use thousands of different and new radiomic features, which are validated as the quantitative imaging biomarkers to characterize intratumoral dynamics throughout diagnosis and treatment [60–63]. The second problem is the analysis of data.…”

Section: Possible Pitfalls and Solutionsmentioning

confidence: 99%

Advanced imaging in adult diffusely infiltrating low-grade gliomas

Bulakbaşı¹,

Paksoy

2019

Insights Imaging

View full text Add to dashboard Cite

The adult diffusely infiltrating low-grade gliomas (LGGs) are typically IDH mutant and slow-growing gliomas having moderately increased cellularity generally without mitosis, necrosis, and microvascular proliferation. Supra-total resection of LGG significantly increases the overall survival by delaying malignant transformation compared with a simple debulking so accurate MR diagnosis is crucial for treatment planning. Data from meta-analysis support the addition of diffusion and perfusion-weighted MR imaging and MR spectroscopy in the diagnosis of suspected LGG. Typically, LGG has lower cellularity (ADCmin), angiogenesis (rCBVmax), capillary permeability (Ktrans), and mitotic activity (Cho/Cr ratio) compared to high-grade glioma. The identification of 2-hydroxyglutarate by MR spectroscopy can reflect the IDH status of the tumor. The initial low ADCmin, high rCBVmax, and Ktrans values are consistent with the poor prognosis. The gradual increase in intratumoral Cho/Cr ratio and rCBVmax values are well correlated with tumor progression. Besides MR-based technical artifacts, which are minimized by the voxel-based assessment of data obtained by histogram analysis, the problems derived from the diversity and the analysis of imaging data should be solved by using artificial intelligence techniques. The quantitative multiparametric MR imaging of LGG can either improve the diagnostic accuracy of their differential diagnosis or assess their prognosis.

show abstract

Section: Possible Pitfalls and Solutionsmentioning

confidence: 99%

Advanced imaging in adult diffusely infiltrating low-grade gliomas

Bulakbaşı¹,

Paksoy

2019

Insights Imaging

View full text Add to dashboard Cite

show abstract

“…IDH is a very important prognostic, diagnostic and therapeutic biomarker for glioma, and triggered the integrated genomic-histological characterization of brain tumours proposed in the 2016 World Health Organization (WHO) classification system 1 . Recently, some studies have shown IDH mutational status may be predicted using neuroimaging with good accuracy (between 78.2% and 92.8%) [11][12][13][14][15][16][17][18][19][20] , and also with very good diagnostic performance when using 2-hydroxyglutarate MR spectroscopy (2HG-MRS, with a pooled 91% sensitivity and 95% specificity) 21,22 . However, neuroimaging is not yet state-of-the-art in detecting IDH mutations in glioma, which is one of the reasons tumour sampling is often still necessary, also because surgical resection/debulking is part of the current mainstay of treatment 23 .…”

mentioning

confidence: 99%

“…However, it is not clear how the patients were selected in that study. Furthermore, the performance of previous deep learning methods on either MRI or H&E slides remains unclear because of the small sample sizes and unbalanced sample distributions in past studies [11][12][13][14][15][16][17][18][19][20] .…”

mentioning

confidence: 99%

Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning

Liu

Shah

Sav

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Recent improvements in ML algorithms and computational power provide an attractive venue for exploring MR radiomic features, an excellent fit for ML-approach analysis that considers the large data size and multimodal nature. Therefore, ML methods have been recently explored to predict glioma genetic biomarkers from MRI radiomic features [10][11][12][21][22][23][24][25][26][27][28][29][30][31][32][33].…”

Section: Discussionmentioning

confidence: 99%

“…Some studies only used conventional MRI sequences and achieved AUC ranging 0.84-0.96 [10,[22][23][24][25]30]. Others explored the added value of advanced MRI imaging, such as MR diffusion or perfusion, with mixed results [21,[26][27][28][29]31]. The highest performance of predicting IDH1 genotype with AUC of 0.96 was observed with a RF model that was trained with conventional MRI, but this study focused only on patients with glioblastomas [10].…”

Section: Discussionmentioning

confidence: 99%

MRI Radiomic Features to Predict IDH1 Mutation Status in Gliomas: A Machine Learning Approach using Gradient Tree Boosting

Sakai

Yang

Kihira

et al. 2020

IJMS

View full text Add to dashboard Cite

Patients with gliomas, isocitrate dehydrogenase 1 (IDH1) mutation status have been studied as a prognostic indicator. Recent advances in machine learning (ML) have demonstrated promise in utilizing radiomic features to study disease processes in the brain. We investigate whether ML analysis of multiparametric radiomic features from preoperative Magnetic Resonance Imaging (MRI) can predict IDH1 mutation status in patients with glioma. This retrospective study included patients with glioma with known IDH1 status and preoperative MRI. Radiomic features were extracted from Fluid-Attenuated Inversion Recovery (FLAIR) and Diffused Weighted Imaging (DWI). The dataset was split into training, validation, and testing sets by stratified sampling. Synthetic Minority Oversampling Technique (SMOTE) was applied to the training sets. eXtreme Gradient Boosting (XGBoost) classifiers were trained, and the hyperparameters were tuned. Receiver operating characteristic curve (ROC), accuracy, and f1-scores were collected. A total of 100 patients (age: 55 ± 15, M/F 60/40); with IDH1 mutant (n = 22) and IDH1 wildtype (n = 78) were included. The best performance was seen with a DWI-trained XGBoost model, which achieved ROC with Area Under the Curve (AUC) of 0.97, accuracy of 0.90, and f1-score of 0.75 on the test set. The FLAIR-trained XGBoost model achieved ROC with AUC of 0.95, accuracy of 0.90, f1-score of 0.75 on the test set. A model that was trained on combined FLAIR-DWI radiomic features did not provide incremental accuracy. The results show that a XGBoost classifier using multiparametric radiomic features derived from preoperative MRI can predict IDH1 mutation status with > 90% accuracy.

show abstract

Prediction of IDH1 Mutation Status in Glioblastoma Using Machine Learning Technique Based on Quantitative Radiomic Data

Cited by 34 publications

References 45 publications

Advanced imaging in adult diffusely infiltrating low-grade gliomas

Advanced imaging in adult diffusely infiltrating low-grade gliomas

Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning

MRI Radiomic Features to Predict IDH1 Mutation Status in Gliomas: A Machine Learning Approach using Gradient Tree Boosting

Contact Info

Product

Resources

About