EUVIMEDEUVIMED entdecken
Crossref

Advanced diffusion models in real-world glioma assessment: A comparative analysis of diagnostic performance, pathological correlations, and prognostic value

Nengjin Zhu, Liwei Mazu, Zuqi Xia, Siyuan He, Ling Wang, Zhenhua Yu, Rulin Xu iD, Longfei Zhao iD, Huawei Jin, Yingqian Huang, Jianping Chu

Neuroradiology · 11.11.2026

DOI: 10.1007/s00234-026-04229-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Neuroradiology, online erschienen am 11.11.2026. Verfasst von Nengjin Zhu und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. Die Arbeit nennt 42 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Various advanced diffusion models have shown robust diagnostic performance in glioma assessment, yet their relative efficacy has not been comprehensively evaluated. This study aimed to compare continuous-time random walk (CTRW), free water corrected-diffusion tensor imaging (FW-DTI), and mean apparent propagator (MAP) in predicting isocitrate dehydrogenase (IDH) status and Ki-67 levels of gliomas, investigate their associations with tumor microenvironment (TME) markers, and evaluate their prognostic value in IDH-wildtype patients. METHODS: This retrospective study included 151 patients (83 men; mean age, 47.45 years) with gliomas. CTRW, FW-DTI, MAP, and apparent diffusion coefficient (ADC) were processed from diffusion spectrum imaging (DSI), and mean parameter values were extracted. Logistic regression assessed model performance for predicting IDH status and Ki-67 levels. Correlation and Cox regression analyses evaluated their associations with TME markers and prognosis, respectively. RESULTS: CTRW, FW-DTI, MAP, and the combined models showed higher diagnostic performance than ADC, with the combined models achieved the highest numerical AUCs (AUCs 0.930 and 0.859) in identifying IDH status and Ki-67 levels. Except for ADC, no statistically significant differences were observed in diagnostic performance among these models. Additionally, multiple advanced diffusion parameters showed correlations with TME markers, and MAP showed associations with overall survival for IDH-wildtype patients (HR = 4.070, P = 0.020). CONCLUSION: CTRW, FW-DTI, and MAP showed similar predictive performance and outperformed ADC for IDH status and tumor proliferation in gliomas. Their associations with TME markers and the prognostic relevance of MAP support the potential of advanced diffusion imaging for noninvasive pathological and prognostic assessment of gliomas.

Schlagwörter

Stichwörter der Autor:innen

Diffusion magnetic resonance imagingGenotypeGliomaKi-67Prognosis

Autor:innen und Institutionen

  • Department of Radiology, First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.
  • Chengdu Zhongying Medical Technology, Chengdu, China.
  • Department of Neurosurgery, First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.
  • Clinical Science Research Department, CT & MR Business Division, Canon Medical Systems, Guangzhou, China.
  • Research and Development Center, Canon Medical Systems, Beijing, China.
  • Department of Neurosurgery, First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. jinhw@mail.sysu.edu.cn.

ORCID-Profile: Rulin Xu, Longfei Zhao

Zitieren

APA

Zhu, N., Mazu, L., Xia, Z., He, S., Wang, L., Yu, Z., Xu, R., Zhao, L., Jin, H., Huang, Y., & Chu, J. (2026). Advanced diffusion models in real-world glioma assessment: A comparative analysis of diagnostic performance, pathological correlations, and prognostic value. Neuroradiology. https://doi.org/10.1007/s00234-026-04229-9

Vancouver

Zhu N, Mazu L, Xia Z, He S, Wang L, Yu Z, et al. Advanced diffusion models in real-world glioma assessment: A comparative analysis of diagnostic performance, pathological correlations, and prognostic value. Neuroradiology. 2026. doi:10.1007/s00234-026-04229-9

BibTeX
@article{Zhu2026,
  author = {Zhu, Nengjin and Mazu, Liwei and Xia, Zuqi and He, Siyuan and Wang, Ling and Yu, Zhenhua and Xu, Rulin and Zhao, Longfei and Jin, Huawei and Huang, Yingqian and Chu, Jianping},
  title = {Advanced diffusion models in real-world glioma assessment: A comparative analysis of diagnostic performance, pathological correlations, and prognostic value},
  journal = {Neuroradiology},
  year = {2026},
  doi = {10.1007/s00234-026-04229-9}
}
Kostenlos starten →