ORIGINAL RESEARCH article
Front. Oncol.
Sec. Cancer Imaging and Image-directed Interventions
Volume 15 - 2025 | doi: 10.3389/fonc.2025.1647254
This article is part of the Research TopicRadiomics and AI-Driven Deep Learning for Cancer Diagnosis and TreatmentView all 10 articles
Better Performance of Cerebral Blood Volume Images Synthesized from Arterial Spin Labeling and Standard MRI in Separating Glioblastoma Recurrence from Treatment Response than Arterial Spin Labeling
Provisionally accepted
- 1Northwestern Polytechnical University, Xi'an, China
- 2Qilu Hospital, Shandong University, Jinan, China
- 3Shandong Provincial Hospital, Jinan, China
- 4Siemens Healthineers China, Shanghai, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Cerebral blood volume (CBV) maps play an important role in differential diagnosis between glioblastoma recurrence and treatment response but it needs high injective velocity. This study aimed to synthesize CBV maps from arterial spin labeling (ASL) and standard MRI sequences by deep learning method and validate its separating value.A total of 744 MRI scans from 364 patients were included in this retrospective, singleinstitution study. A 3D incrementable encoder-decoder network (IEDN) designed for asymmetrical sample size was trained to synthesize CBV maps from ASL and standard MRIs. The synthetic performance was evaluated quantitatively using the structural similarity index (SSIM), peak signal-to-noise ratio (PSNR) and qualitatively using a 4point Likert scale (from 0 to 3). In 96 patients suspected of glioblastoma recurrence vs treatment response as external testing from a hospital-based cohort, difference between synthetic CBV maps and ASL in adding values to standard MRIs were tested by using the Z test. The best algorithm seemed to be achieved by 3D IEDN with ASL+T1WI+T2WI+ADC maps+postcontrast T1WI {SSIM, 88.69±3.97 (%); PSNR, 32.76±3.39 (dB)}. For image quality scores, the mean image quality score for all the synthetic CBV maps was 2.90. Standard MRI plus synthetic CBV maps had better performance than standard MRI and ASL scans in differential diagnosis between tumor recurrence and treatment response (p=0.019). Therefore, 3D IEDN produced qualified synthetic CBV maps without need of high injective velocity from ASL and standard MRIs. The synthetic CBV maps achieve better performance in the differential diagnosis between glioblastoma recurrence and treatment response.
Keywords: Arterial Spin Labeling, cerebral blood volume, tumor recurrence, Glioblastoma, Magnetic Resonance Imaging
Received: 15 Jun 2025; Accepted: 11 Aug 2025.
Copyright: © 2025 Pan, Zhou, Wu, Wang, He, Xiang and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Bao Wang, Qilu Hospital, Shandong University, Jinan, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.