EUVIMEDEUVIMED entdecken
Crossref

A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models

Feng-Ying Zhu, Wen-Jing Chen, Hao-Yan Chen, Si-Yu Ren, Li-Yong Zhuo, Tian-Da Wang, Cong-Cong Ren, Xiao-Ping Yin, Jia-Ning Wang

European Journal of Radiology · 2025-11 · 192 · 112406

DOI: 10.1016/j.ejrad.2025.112406 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Journal of Radiology, 2025-11. Verfasst von Feng-Ying Zhu und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Affiliated Hospital of Hebei University, Baoding 071000, China. Hauptthemen laut MEDLINE: Glioma, Brain Neoplasms, Deep Learning, DNA Modification Methylases. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 29 Literaturangaben. Gefördert von: Hebei University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: The present study aimed to develop a noninvasive predictive framework that integrates clinical data, conventional radiomics, habitat imaging, and deep learning for the preoperative stratification of MGMT gene promoter methylation in glioma. MATERIALS AND METHODS: This retrospective study included 410 patients from the University of California, San Francisco, USA, and 102 patients from our hospital. Seven models were constructed using preoperative contrast-enhanced T1-weighted MRI with gadobenate dimeglumine as the contrast agent. Habitat radiomics features were extracted from tumor subregions by k-means clustering, while deep learning features were acquired using a 3D convolutional neural network. Model performance was evaluated based on area under the curve (AUC) value, F1-score, and decision curve analysis. RESULTS: The combined model integrating clinical data, conventional radiomics, habitat imaging features, and deep learning achieved the highest performance (training AUC = 0.979 [95 % CI: 0.969-0.990], F1-score = 0.944; testing AUC = 0.777 [0.651-0.904], F1-score = 0.711). Among the single-modality models, habitat radiomics outperformed the other models (training AUC = 0.960 [0.954-0.983]; testing AUC = 0.724 [0.573-0.875]). CONCLUSION: The proposed multimodal framework considerably enhances preoperative prediction of MGMT gene promoter methylation, with habitat radiomics highlighting the critical role of tumor heterogeneity. This approach provides a scalable tool for personalized management of glioma.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GliomaBrain NeoplasmsDeep LearningDNA Modification MethylasesTumor Suppressor ProteinsDNA Repair EnzymesDNA MethylationMagnetic Resonance ImagingHumansFemaleMaleMiddle AgedPromoter Regions, GeneticRetrospective StudiesAdultAgedContrast MediaMultimodal ImagingRadiomics

Stichwörter der Autor:innen

GliomaHabitat radiomicsMGMT gene promoter methylationMultimodal deep learningRadiogenomics

Substanzen

MGMT protein, humanDNA Modification MethylasesTumor Suppressor ProteinsDNA Repair EnzymesContrast Media

Autor:innen und Institutionen

  • Department of Radiology, Affiliated Hospital of Hebei University, Baoding 071000, China.
  • Department of Research and Development, United Imaging Intelligence (Beijing) Co., Ltd., Beijing 100089, China.
  • Department of Radiology, Affiliated Hospital of Hebei University, Baoding 071000, China. Electronic address: jianing0218@163.com.

Förderung

  • Hebei University

Zitieren

APA

Zhu, F. Y., Chen, W. J., Chen, H. Y., Ren, S. Y., Zhuo, L. Y., Wang, T. D., Ren, C. C., Yin, X. P., & Wang, J. N. (2025). A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models. European Journal of Radiology. https://doi.org/10.1016/j.ejrad.2025.112406

Vancouver

Zhu FY, Chen WJ, Chen HY, Ren SY, Zhuo LY, Wang TD, et al. A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models. European Journal of Radiology. 2025. doi:10.1016/j.ejrad.2025.112406

BibTeX
@article{Zhu2025,
  author = {Zhu, Feng-Ying and Chen, Wen-Jing and Chen, Hao-Yan and Ren, Si-Yu and Zhuo, Li-Yong and Wang, Tian-Da and Ren, Cong-Cong and Yin, Xiao-Ping and Wang, Jia-Ning},
  title = {A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models},
  journal = {European Journal of Radiology},
  year = {2025},
  doi = {10.1016/j.ejrad.2025.112406}
}
Kostenlos starten →