Share：
Share this content in WeChat
X
[Chinese] [PDF] 8 1
• Review •
Research progress of convolutional neural network and vision transformer in gliomas
YANG Haohui  XU Tao  WANG Wei  AN Liangliang  AO Yongfang  ZHU Jiabao 

DOI:10.12015/issn.1674-8034.2026.01.026.


[Abstract] Gliomas pose significant challenges to traditional diagnosis and treatment due to their high heterogeneity, strong invasiveness, and poor prognosis. The introduction of deep learning (DL) technology has opened up a new avenue for their precise diagnosis and treatment, among which convolutional neural network (CNN) and Vision Transformer (ViT) are core tools. CNN inherently excels in local feature extraction (e.g., tumor edges, texture details) through hierarchical convolution operations, while ViT stands out in global context modeling (e.g., cross-regional heterogeneity of tumors, multimodal correlations) based on the self-attention mechanism. The fusion strategy of CNN and ViT integrates local fine-grained features with global associated information, demonstrating remarkable advantages in addressing clinical dilemmas such as blurred glioma boundaries and cross-modal data heterogeneity. This article reviews the research progress of CNN and ViT in key clinical tasks of gliomas, including detection and segmentation, pathological grading, molecular subtyping, and prognosis assessment. It elaborates on their principles, individual applications, and fusion strategies. Furthermore, it discusses the prevailing challenges in the field, such as the heavy reliance on annotated data and insufficient model interpretability, and outlines promising future research directions, including the development of lightweight architectures, the advancement of self-supervised learning paradigms, and the promotion of multi-omics integration. This review thereby provides a systematic reference for the intelligent diagnosis of gliomas.
[Keywords] glioma；deep learning；convolutional neural network；vision transformer；magnetic resonance imaging

YANG Haohui1, 2   XU Tao3   WANG Wei4   AN Liangliang5   AO Yongfang6   ZHU Jiabao2*  

1 Changzhi Medical College, Changzhi 046000, China

2 Department of Neurosurgery, Yuncheng Central Hospital Affiliated to Shanxi Medical University, Yuncheng 044000, China

3 Department of Pathology, Yuncheng Central Hospital Affiliated to Shanxi Medical University, Yuncheng 044000, China

4 Department of Radiation Oncology, Yuncheng Central Hospital Affiliated to Shanxi Medical University, Yuncheng 044000, China

5 Department of Ultrasound, Yuncheng Central Hospital Affiliated to Shanxi Medical University, Yuncheng 044000, China

6 Department of Emergency Medicine, the Third People's Hospital of Guizhou Province, Guiyang 550001, China

Corresponding author: ZHU J B, E-mail: zhujiabao1982@163.com

Conflicts of interest   None.

Received  2025-08-26
Accepted  2025-12-09
DOI: 10.12015/issn.1674-8034.2026.01.026
DOI:10.12015/issn.1674-8034.2026.01.026.

[1]
LAPOINTE S, PERRY A, BUTOWSKI N A. Primary brain tumours in adults[J]. Lancet, 2018, 392(10145): 432-446. DOI: 10.1016/s0140-6736(18)30990-5.
[2]
SHOUSHTARI F K, ELAHI R, VALIZADEH G, et al. Current trends in glioma tumor segmentation: A survey of deep learning modules[J/OL]. Phys Med, 2025, 135: 104988 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40460759/. DOI: 10.1016/j.ejmp.2025.104988.
[3]
GORE S, CHOUGULE T, JAGTAP J, et al. A review of radiomics and deep predictive modeling in glioma characterization[J]. Acad Radiol, 2021, 28(11): 1599-1621. DOI: 10.1016/j.acra.2020.06.016.
[4]
ALLEMAN K, KNECHT E, HUANG J, et al. Multimodal deep learning-based prognostication in glioma patients: A systematic review[J/OL]. Cancers (Basel), 2023, 15(2): 545 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36672494/. DOI: 10.3390/cancers15020545.
[5]
CHEN C, MAT ISA N A, LIU X. A review of convolutional neural network based methods for medical image classification[J/OL]. Comput Biol Med, 2025, 185: 109507 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39631108/. DOI: 10.1016/j.compbiomed.2024.109507.
[6]
ABURASS S, DORGHAM O, SHAQSI J AL, et al. Vision transformers in medical imaging: A comprehensive review of advancements and applications across multiple diseases[J/OL]. J Imaging Inform Med, 2025 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40164818/. DOI: 10.1007/s10278-025-01481-y.
[7]
TAKAHASHI S, SAKAGUCHI Y, KOUNO N, et al. Comparison of vision transformers and convolutional neural networks in medical image analysis: A systematic review[J/OL]. J Med Syst, 2024, 48(1): 84 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39264388/. DOI: 10.1007/s10916-024-02105-8.
[8]
XU H, XU Q, CONG F, et al. Vision transformers for computational histopathology[J]. IEEE Rev Biomed Eng, 2024, 17: 63-79. DOI: 10.1109/rbme.2023.3297604.
[9]
PAPANASTASIOU G, DIKAIOS N, HUANG J, et al. Is attention all you need in medical image analysis? A review[J]. IEEE J Biomed Health Inform, 2024, 28(3): 1398-1411. DOI: 10.1109/jbhi.2023.3348436.
[10]
FUKUSHIMA K. Neocognitron: A self organizing neural network model for a mechanism of pattern recognition unaffected by shift in position[J]. Biol Cybern, 1980, 36(4): 193-202. DOI: 10.1007/bf00344251.
[11]
YANG R, YU Y. Artificial convolutional neural network in object detection and semantic segmentation for medical imaging analysis[J/OL]. Front Oncol, 2021, 11: 638182 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/33768000/. DOI: 10.3389/fonc.2021.638182.
[12]
KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90. DOI: 10.1145/3065386.
[13]
ZHANG H, OGASAWARA K. Grad-cam-based explainable artificial intelligence related to medical text processing[J/OL]. Bioengineering (Basel), 2023, 10(9): 1070 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/37760173/. DOI: 10.3390/bioengineering10091070.
[14]
AL-HAMMURI K, GEBALI F, KANAN A, et al. Vision transformer architecture and applications in digital health: A tutorial and survey[J/OL]. Vis Comput Ind Biomed Art, 2023, 6(1): 14 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/37428360/. DOI: 10.1186/s42492-023-00140-9.
[15]
LI Z, CONG Y, CHEN X, et al. Vision transformer-based weakly supervised histopathological image analysis of primary brain tumors[J/OL]. iScience, 2023, 26(1): 105872 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36647383/. DOI: 10.1016/j.isci.2022.105872.
[16]
LIU Z, LV Q, YANG Z, et al. Recent progress in transformer-based medical image analysis[J/OL]. Comput Biol Med, 2023, 164: 107268 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/37494821/. DOI: 10.1016/j.compbiomed.2023.107268.
[17]
CHENG Z, WU Y, LI Y, et al. A comprehensive review of explainable artificial intelligence (xai) in computer vision[J/OL]. Sensors (Basel), 2025, 25(13): 4166 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40648421/. DOI: 10.3390/s25134166.
[18]
SINGH Y, HATHAWAY Q A, KEISHING V, et al. Beyond post hoc explanations: A comprehensive framework for accountable ai in medical imaging through transparency, interpretability, and explainability[J/OL]. Bioengineering (Basel), 2025, 12(8): 879 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40868392/. DOI: 10.3390/bioengineering12080879.
[19]
GUPTA R K, BHARTI S, KUNHARE N, et al. Brain tumor detection and classification using cycle generative adversarial networks[J]. Interdiscip Sci, 2022, 14(2): 485-502. DOI: 10.1007/s12539-022-00502-6.
[20]
DISCI R, GURCAN F, SOYLU A. Advanced brain tumor classification in mr images using transfer learning and pre-trained deep cnn models[J/OL]. Cancers (Basel), 2025, 17(1): 121 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39796749/. DOI: 10.3390/cancers17010121.
[21]
RAHIMPOUR M, BOELLAARD R, JENTJENS S, et al. A multi-label cnn model for the automatic detection and segmentation of gliomas using [(18)f]fet pet imaging[J]. Eur J Nucl Med Mol Imaging, 2023, 50(8): 2441-2452. DOI: 10.1007/s00259-023-06193-5.
[22]
TRIPATHI P C, BAG S. An attention-guided cnn framework for segmentation and grading of glioma using 3d mri scans[J]. IEEE/ACM Trans Comput Biol Bioinform, 2023, 20(3): 1890-1904. DOI: 10.1109/tcbb.2022.3220902.
[23]
RUDIE J D, CALABRESE E, SALUJA R, et al. Longitudinal assessment of posttreatment diffuse glioma tissue volumes with three-dimensional convolutional neural networks[J/OL]. Radiol Artif Intell, 2022, 4(5): e210243 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36204543/. DOI: 10.1148/ryai.210243.
[24]
TUMMALA S, KADRY S, BUKHARI S A C, et al. Classification of brain tumor from magnetic resonance imaging using vision transformers ensembling[J]. Curr Oncol, 2022, 29(10): 7498-7511. DOI: 10.3390/curroncol29100590.
[25]
REDDY C K K, REDDY P A, JANAPATI H, et al. A fine-tuned vision transformer based enhanced multi-class brain tumor classification using mri scan imagery[J/OL]. Front Oncol, 2024, 14: 1400341 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39091923/. DOI: 10.3389/fonc.2024.1400341.
[26]
WANG S, WU J, CHEN M, et al. Balanced transformer: Efficient classification of glioblastoma and primary central nervous system lymphoma[J/OL]. Phys Med Biol, 2024, 69(4) [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/38232389/. DOI: 10.1088/1361-6560/ad1f88.
[27]
ASIRI A A, SHAF A, ALI T, et al. Advancing brain tumor detection: Harnessing the swin transformer's power for accurate classification and performance analysis[J/OL]. PeerJ Comput Sci, 2024, 10: e1867 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/38435590/. DOI: 10.7717/peerj-cs.1867.
[28]
XIONG D, REN X, HUANG W, et al. Noninvasive classification of glioma subtypes using multiparametric mri to improve deep learning[J/OL]. Diagnostics (Basel), 2022, 12(12): 3063 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36553070/. DOI: 10.3390/diagnostics12123063.
[29]
LI R, LIAO Y, HUANG Y, et al. Deepglioseg: Advanced glioma mri data segmentation with integrated local-global representation architecture[J/OL]. Front Oncol, 2025, 15: 1449911 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39968077/. DOI: 10.3389/fonc.2025.1449911.
[30]
ZEINELDIN R A, KARAR M E, ELSHAER Z, et al. Explainable hybrid vision transformers and convolutional network for multimodal glioma segmentation in brain mri[J/OL]. Sci Rep, 2024, 14(1): 3713 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/38355678/. DOI: 10.1038/s41598-024-54186-7.
[31]
ZHUANG Y, LIU H, FANG W, et al. A 3d hierarchical cross-modality interaction network using transformers and convolutions for brain glioma segmentation in mr images[J]. Med Phys, 2024, 51(11): 8371-8389. DOI: 10.1002/mp.17354.
[32]
ZHUGE Y, NING H, MATHEN P, et al. Automated glioma grading on conventional mri images using deep convolutional neural networks[J]. Med Phys, 2020, 47(7): 3044-3053. DOI: 10.1002/mp.14168.
[33]
XU C, PENG Y, ZHU W, et al. An automated approach for predicting glioma grade and survival of lgg patients using cnn and radiomics[J/OL]. Front Oncol, 2022, 12: 969907 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36033433/. DOI: 10.3389/fonc.2022.969907.
[34]
FASIHI SHIREHJINI O, BABAPOUR MOFRAD F, SHAHMOHAMMADI M, et al. Grading of gliomas using transfer learning on mri images[J]. Magma, 2023, 36(1): 43-53. DOI: 10.1007/s10334-022-01046-y.
[35]
MUNQUAD S, DAS A B. Deepautoglioma: A deep learning autoencoder-based multi-omics data integration and classification tools for glioma subtyping[J/OL]. BioData Min, 2023, 16(1): 32 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/37968655/. DOI: 10.1186/s13040-023-00349-7.
[36]
SHIRAE S, DEBSARKAR S S, KAWANAKA H, et al. Multimodal ensemble fusion deep learning using histopathological images and clinical data for glioma subtype classification[J]. IEEE Access, 2025, 13: 57780-57797. DOI: 10.1109/access.2025.3556713.
[37]
ELAZAB N, KHALIFA F, ALLAH W GAB, et al. Histopathological-based brain tumor grading using 2d-3d multi-modal cnn-transformer combined with stacking classifiers[J/OL]. Sci Rep, 2025, 15(1): 27764 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40739310/. DOI: 10.1038/s41598-025-11754-9.
[38]
LOUIS D N, PERRY A, WESSELING P, et al. The 2021 who classification of tumors of the central nervous system: A summary[J]. Neuro Oncol, 2021, 23(8): 1231-1251. DOI: 10.1093/neuonc/noab106.
[39]
CHOI Y S, BAE S, CHANG J H, et al. Fully automated hybrid approach to predict the idh mutation status of gliomas via deep learning and radiomics[J]. Neuro Oncol, 2021, 23(2): 304-313. DOI: 10.1093/neuonc/noaa177.
[40]
CLUCERU J, INTERIAN Y, PHILLIPS J J, et al. Improving the noninvasive classification of glioma genetic subtype with deep learning and diffusion-weighted imaging[J]. Neuro Oncol, 2022, 24(4): 639-652. DOI: 10.1093/neuonc/noab238.
[41]
SACLI-BILMEZ B, BAS A, ERŞEN DANYELI A, et al. Detecting idh and tertp mutations in diffuse gliomas using (1)H-MRS with attention deep-shallow networks[J/OL]. Comput Biol Med, 2025, 186: 109736 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39874812/. DOI: 10.1016/j.compbiomed.2025.109736.
[42]
YUAN Y, LI G, MEI S, et al. Deep mutual learning on hybrid amino acid pet predicts h3k27m mutations in midline gliomas[J/OL]. NPJ Precis Oncol, 2024, 8(1): 274 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39587279/. DOI: 10.1038/s41698-024-00760-1.
[43]
CHEN H, LIN F, ZHANG J, et al. Deep learning radiomics to predict pten mutation status from magnetic resonance imaging in patients with glioma[J/OL]. Front Oncol, 2021, 11: 734433 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/34671557/. DOI: 10.3389/fonc.2021.734433.
[44]
KOSKA İ, KOSKA Ç. Deep learning classification of mgmt status of glioblastomas using multiparametric mri with a novel domain knowledge augmented mask fusion approach[J/OL]. Sci Rep, 2025, 15(1): 3273 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39863759/. DOI: 10.1038/s41598-025-87803-0.
[45]
KUDUS K, WAGNER M W, NAMDAR K, et al. Beyond hand-crafted features for pretherapeutic molecular status identification of pediatric low-grade gliomas[J/OL]. Sci Rep, 2024, 14(1): 19102 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/39154039/. DOI: 10.1038/s41598-024-69870-x.
[46]
TAK D, YE Z, ZAPAISCHYKOVA A, et al. Noninvasive molecular subtyping of pediatric low-grade glioma with self-supervised transfer learning[J/OL]. Radiol Artif Intell, 2024, 6(3): e230333 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/38446044/. DOI: 10.1148/ryai.230333.
[47]
XU Q, XU Q Q, SHI N, et al. A multitask classification framework based on vision transformer for predicting molecular expressions of glioma[J/OL]. Eur J Radiol, 2022, 157: 110560 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/36327857/. DOI: 10.1016/j.ejrad.2022.110560.
[48]
NIU W, YAN J, HAO M, et al. Mri transformer deep learning and radiomics for predicting idh wild type tert promoter mutant gliomas[J/OL]. NPJ Precis Oncol, 2025, 9(1): 89 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40148588/. DOI: 10.1038/s41698-025-00884-y.
[49]
CHENG J, LIU J, KUANG H, et al. A fully automated multimodal mri-based multi-task learning for glioma segmentation and idh genotyping[J]. IEEE Trans Med Imaging, 2022, 41(6): 1520-1532. DOI: 10.1109/tmi.2022.3142321.
[50]
ZHAO Y, WANG W, JI Y, et al. Computational pathology for prediction of isocitrate dehydrogenase gene mutation from whole slide images in adult patients with diffuse glioma[J]. Am J Pathol, 2024, 194(5): 747-758. DOI: 10.1016/j.ajpath.2024.01.009.
[51]
LYU Q, PARRENO-CENTENO M, PAPA J P, et al. Survnet: A low-complexity convolutional neural network for survival time classification of patients with glioblastoma[J/OL]. Heliyon, 2024, 10(12): e32870 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/38988550/. DOI: 10.1016/j.heliyon.2024.e32870.
[52]
LEE J O, AHN S S, CHOI K S, et al. Added prognostic value of 3d deep learning-derived features from preoperative mri for adult-type diffuse gliomas[J]. Neuro Oncol, 2024, 26(3): 571-580. DOI: 10.1093/neuonc/noad202.
[53]
YIN Q, CHEN W, ZHANG C, et al. A convolutional neural network model for survival prediction based on prognosis-related cascaded wx feature selection[J]. Lab Invest, 2022, 102(10): 1064-1074. DOI: 10.1038/s41374-022-00801-y.
[54]
HOLLON T C, PANDIAN B, URIAS E, et al. Rapid, label-free detection of diffuse glioma recurrence using intraoperative stimulated raman histology and deep neural networks[J/OL]. Neuro Oncol, 2021, 23(1): 144-155. DOI: 10.1093/neuonc/noaa162.
[55]
GUO P, UNBERATH M, HEO H Y, et al. Learning-based analysis of amide proton transfer-weighted mri to identify true progression in glioma patients[J/OL]. Neuroimage Clin, 2022, 35: 103121 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/35905666/. DOI: 10.1016/j.nicl.2022.103121.
[56]
ZHAO C, CHEN M, WEN X, et al. Three-step-guided visual prediction of glioblastoma recurrence from multimodality images[J/OL]. Comput Med Imaging Graph, 2025, 124: 102585 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40532633/. DOI: 10.1016/j.compmedimag.2025.102585.
[57]
OH J H, CHOI W, KO E, et al. Pathcnn: Interpretable convolutional neural networks for survival prediction and pathway analysis applied to glioblastoma[J]. Bioinformatics, 2021, 37(Suppl_1): i443-i450. DOI: 10.1093/bioinformatics/btab285.
[58]
GOMAA A, HUANG Y, STEPHAN P, et al. A self-supervised multimodal deep learning approach to differentiate post-radiotherapy progression from pseudoprogression in glioblastoma[J/OL]. Sci Rep, 2025, 15(1): 17133 [2025-08-26]. https://pubmed.ncbi.nlm.nih.gov/40382400/. DOI: 10.1038/s41598-025-02026-7.
[59]
GUO J, XU P, WU Y, et al. Cromam: A cross-magnification attention feature fusion model for predicting genetic status and survival of gliomas using histological images[J]. IEEE J Biomed Health Inform, 2024, 28(12): 7345-7356. DOI: 10.1109/jbhi.2024.3431471.

PREV Advances in the application of diffusion tensor imaging combined with artificial intelligence in cerebral small vessel disease
NEXT Research progress in multimodal MRI radiomics for predicting molecular typing of gliomas
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn