Molecular Mutation Classification of Glioblastoma Multiforme Using Metric-Meta-Learning Approach Based on Siamese Neural Networks: A Retrospective Study
International Clinical Neuroscience Journal,
Vol. 12 No. Continuous (2025),
25 January 2026
,
Page e1
https://doi.org/10.22037/icnj.v12i1.50472
Abstract
Background: Glioblastoma is the most common malignant primary brain tumor in adults. It has been classified into mutant and wild-type subtypes of isocitrate dehydrogenase (IDH). Patient therapeutic options in Glioblastoma vary across subtypes. Therefore, preoperative classification of IDH mutation status is important for therapeutic decision-making. This study aimed to develop and evaluate a metric-meta learning approach based on a Siamese Neural Network model for noninvasive classification of IDH mutation status in patients from magnetic resonance imaging (MRI) scans, based on their similarity.
Methods: MRI scans from 93 patients (2018-2024) with glioblastoma tumors were collected and subdivided into IDH-mutant and IDH-wildtype scans. After preprocessing MRI scans, a Siamese Convolutional Neural Network was combined with two pre-trained models, InceptionResNetV2 and ResNet152V2, as an embedding function to differentiate between classes using a small dataset. The network extracts relevant features to classify patients' IDH-mutation status. We propose a self-attention mechanism combined with a Siamese neural network to improve model accuracy.
Results: We evaluate a Siamese Neural Network on MRI scans of brain tumors. After adding an attention mechanism to the Siamese Neural network, we obtain classification accuracy (73.33% - 70%), precision (68.42% - 68%), recall (86%- 73.33%), and F1-score (76.20%- 70.96%) on our dataset.
Conclusion: Our proposed model demonstrated promising results after combined with an attention mechanism to perform classification of IDH-mutation status using a limited number of brain tumor images and accuracy of this model has improved.
- Glioblastoma, IDH-mutation, magnetic resonance imaging, Image classification, Siamese Neural Network.
How to Cite
References
1. Zhnyu T, Yuyun X, Lei J, Abudumijiti A, Junfeng L, Zhicheng J, et al. Deep Learning of Imaging Phenotype and Genotype for Predicting Overall Survival Time of Glioblastoma Patients. IEEE Trans Med Imaging. 2020;39(6):2100–9. doi: 10.1109/TMI.2020.2964310
2. Yang L, Fei Z, Heba A, Justin L, Peiwen C. Immunotherapy for glioblastoma: current state, challenges, and future perspectives. Cellular & Molecular Immunology. 2024;1354–75. doi: 10.1038/s41423-024-01226-x
3. Joanna W, Chetan B. Genomic discoveries in adult astrocytoma. Current Opinion in Genetics & Development. 2015;17-24. doi: 10.1016/j.gde.2014.12.002
4. Sue H, Yang L, Sabrina JC, Mingyu Q. IDH mutation in glioma: molecular mechanisms and potential therapeutic targets. Br J Cancer 122. 2020;1580–9. doi: 10.1038/s41416-020-0814-x
5. Mustafa Emre S, Zeki B, Ümit K, Tolga A, Ahmet Hamit C, er al. Potential Biomarkers for IDH-Mutant and ID H-Wild-Type Glioblastomas: A Single-Center Retrospective Study. J Clin Med. 2025;14(7):2518. doi: 10.3390/jcm14072518
6. Chandan Y, Benjamin W, Nghi T, James MH, Divya R, Niloufar S, et al. MRI-Based Deep Learning Method for Classification of IDH Mutation Status. Bioengineering (Basel) 2023;10(9):1045. doi: 10.3390/bioengineering10091045
7. Sara L, Sara L, Angela D, Alessandro C, Andrea V. The Many Facets of Tumor Heterogeneity: Is Metabolism Lagging Behind? Cancers (Basel). 2019;11(10):1574. doi: 10.3390/cancers11101574
8. Xiaoli C, Junqiang L, Shuaiwen W, Jing Z, Lubin G. Diagnostic accuracy of a machine learning-based radiomics approach of MR in predicting IDH mutations in glioma patients: a systematic review and meta-analysis. Front. Oncol. 2024. doi: 10.3389/fonc.2024.1409760
9. Sebastian V, Fatih I, Maarten W, Georgios K, Renske G, Joost WS, et al. Combined molecular subtyping, grading, and segmentation of glioma using multi-task deep learning. Neuro Oncol. 2023;25(2):279-89. doi: 10.1093/neuonc/noac166
10. Kester P, David K, David S. Disease Assessments in Patients with Glioblastoma. Curr Oncol Rep. 2023;25(9):1057-69. doi: 10.1007/s11912-023-01440-2
11. Mengfang L, Yuanyuan J,Yanzhou Z, Haisheng Z. Medical image analysis using deep learning algorithms. Front. Public Health. 2023. doi: 10.3389/fpubh.2023.1273253
12. Hasan I, Tareque Abu A, Suriya C, Muhammad A. Few Shot Learning for Medical Imaging: A Review of Categorized Images. 2023 IEEE 7th Conference on Information and Communication Technology.
13. Archit P, Minwoo L. Learning from Few Examples: A Summary of Approaches to Few-Shot Learning. Computer Vision and Pattern Recognition. 2022. doi: 10.48550/arXiv.2203.04291
14. Zhaorui F, Xin C,Yaling X. Few-shot rumor detection based on prompt learning and meta learning. Proc. SPIE 13228, Fifth International Conference on Computer Communication and Network Security (CCNS 2024), 1322811 (22 August 2024). doi: 10.1117/12.3038277
15. Hassan G, Fereshteh M, Fang C, Amir G. Meta-learning approaches for few-shot learning: A survey of recent advances. Machine Learning. 2023. doi: 10.48550/arXiv.2303.07502
16. Siva J, Brintha T. Sharpening enhancement technique for MR images to enhance the segmentation. Biomedical Signal Processing and Control. 2018;21-30. doi: 10.1016/j.bspc.2017.11.007
17. Kyle M, Nirmalie W, Sadiq S, Stewart M, Jeremie C. A Convolutional Siamese Network for Developing Similarity Knowledge in the SelfBACK Dataset. 25th International conference on case-based reasoning (ICCBR). Trondheim, Norway. 2017.
18. Nicolás S, Alejandro B. Siamese neural networks in recommendation. Neural Computing and Applications. 2023. doi:10.1007/s00521-023-08610-0
19. Sounak D, Anjan D, J. Ignacio T, Suman G, Josep L. SigNet: Convolutional Siamese Network for Writer Independent Offline Signature Verification. Computer Vision and Pattern Recognition. 2017. doi: 10.48550/arXiv.1707.02131
20. Christian S, Sergey L, Vincent V, Alex A. Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning. Computer Vision and Pattern Recognition. 2016. doi: 10.48550/arXiv.1602.07261
21. Long N, Zhiping L, Dongyun L, Jiuwen C. Deep CNNs for microscopic image classification by exploiting transfer learning and feature concatenation. 2018 IEEE International Symposium on Circuits and Systems (ISCAS). doi: 10.1109/ISCAS.2018.8351550
22. Kousalya K. Terrain identification and land price estimation using deep learning. Proceedings Of The 4th National Conference on Current And Emerging Process Technologies E-Concept-2021. doi: 10.1063/5.0068625
23. Amarnath A, Ali B, Jeremy H. Transfer-Learning Approach for Enhanced Brain Tumor Classification in MRI Imaging. BioMed Informatics. 2024;4(3):1745-56. doi: 10.3390/biomedinformatics4030095
24. Olga R, Jia D, Hao S, Jonathan K, Sanjeev S, et al . ImageNet Large Scale Visual Recognition Challenge. Computer Vision and Pattern Recognition. 2015. doi: 10.48550/arXiv.1409.0575
25. Zijun C, Aijun Y. A case weighted similarity deep measurement method based on a self-attention Siamese neural network. Industrial artificial intelligence. 2023. doi: 10.1007/s44244-022-00002-y
26. Jinfan Z, Xiaolong Z, Xiaoli L. A Diabetic Retinopathy Classification Method Based on Novel Attention Mechanism. Lecture Notes in Computer Science, Intelligent Computing Theories and Application. 2021. doi: 10.1007/978-3-030-84522-3_10
27. Zaenab A, Laith A, Jinglan Z, Yuefeng L, Yuantong G. Attention-Enhanced Deep Learning for Accurate Musculoskeletal X-Ray Classification. doi: 10.2139/ssrn.5143442
28. Vishal A, Mamta T, Amit Y, Gesu T, Mamata P, et al. Optimizing brain tumor detection in MRI scans through InceptionResNetV2 and deep stacked Autoencoders with SwiGLU activation and sparsity regularization. MethodsX. 2025;14. doi: 10.1016/j.mex.2025.103255
29. Mehdi N, Muktar A, Hossein A, Menasha T. Hybrid Inception Architecture with Residual Connection: Fine-tuned Inception-ResNet Deep Learning Model for Lung Inflammation Diagnosis from Chest Radiographs. Procedia Computer Science. 2024;235(1):1841-50. doi: 10.1016/j.procs.2024.04.175
30. Nallamotu P, MuneeswariG. Modified ResNet152v2: Binary Classification and Hybrid Segmentation of Brain Stroke Using Transfer Learning-Based Approach. Polish Journal of Medical Physics and Engineering. 2024;30:1. doi: 10.2478/pjmpe-2024-0004
31. Hosseini A, Hosseini E, Hajianfar G, Shiri I, Servaes S, et al. MRI-Based Radiomics Combined with Deep Learning for Distinguishing IDH-Mutant WHO Grade 4 Astrocytomas from IDH-Wild-Type Glioblastomas. Cancers. 2023;15(3):951. doi: 10.3390/cancers15030951
- Abstract Viewed: 126 times
- PDF Downloaded: 65 times