Hybrid Deep Feature Extraction and mRMR-Based Optimization-Assisted Machine Learning Approaches in MRI-Based Brain Tumor Classification

Abstract

Brain tumors are among the important neurological diseases that require early and accurate diagnosis due to their high mortality rates and serious clinical outcomes. Although Magnetic Resonance Imaging (MRI) is widely used in the evaluation of brain tumors, structural and tissue similarities between different tumor types can complicate the diagnostic process. This increases the need for computer-aided decision systems. This study conducted a comparative analysis of machine learning approaches supported by mRMR-based feature selection and hyperparameter optimization for determining the optimal number of features in MRI-based brain tumor classification. Features extracted from MRI images belonging to glioma, meningioma, no tumor, and pituitary tumor classes were used in the study. The baseline and optimized performances of KNN, SVM, RF, XGBoost, and LightGBM algorithms were evaluated based on accuracy, precision, recall, F1-Score, and runtime criteria. After optimization, the accuracy of the KNN algorithm increased from 81.10% to 88.13%, while the accuracy of the SVM algorithm increased from 88.57% to 92.75%. The SVM classifier exhibited greater performance relative to other approaches, attaining the best accuracy of 92.75% and F1-Score of 92.73% on the test set.

Keywords

• Brain tumor
• deep feature extraction
• feature selection
• optimization
• classification.

MRI Tabanlı Beyin Tümörü Sınıflandırmasında Hibrit Derin Öznitelik Çıkarımı ve mRMR Tabanlı Optimizasyon Destekli Makine Öğrenmesi Yaklaşımları

Abstract

Beyin tümörleri, yüksek ölüm oranları ve ciddi klinik sonuçları nedeniyle erken ve doğru tanı gerektiren önemli nörolojik hastalıklar arasındadır. Manyetik Rezonans Görüntüleme (MRG), beyin tümörlerinin değerlendirilmesinde yaygın olarak kullanılsa da farklı tümör tipleri arasındaki yapısal ve doku benzerlikleri tanı sürecini karmaşıklaştırabilir. Bu durum, bilgisayar destekli karar sistemlerine olan ihtiyacı artırmaktadır. Bu çalışma, MRG tabanlı beyin tümörü sınıflandırmasında optimal özellik sayısını belirlemek için mRMR tabanlı özellik seçimi ve hiperparametre optimizasyonu ile desteklenen makine öğrenme yaklaşımlarının karşılaştırmalı bir analizini gerçekleştirmiştir. Çalışmada, gliom, menenjiom, normal ve hipofiz tümörü sınıflarına ait MRG görüntülerinden çıkarılan özellikler kullanılmıştır. KNN, SVM, RF, XGBoost ve LightGBM algoritmalarının temel ve optimize edilmiş performansları doğruluk, hassasiyet, geri çağırma, F1-Skoru ve çalışma süresi kriterlerine göre değerlendirilmiştir. Optimizasyon sonrasında, KNN algoritmasının doğruluğu %81.10'dan %88.13'e, SVM algoritmasının doğruluğu ise %88.57'den %92.75'e yükselmiştir. SVM sınıflandırıcı, diğer yaklaşımlara kıyasla daha yüksek performans sergileyerek test setinde %92.75 doğruluk ve %92.73 F1-Skorunu elde etmiştir.

Keywords

• beyin tümörü
• derin öznitelik çıkarımı
• öznitelik seçimi
• optimizasyon
• sınıflandırma

References

1. Şimşek AB, Dicle A. Primer beyin tümörü: Klinik özellikler, tedavi ve bakım. Dokuz Eylül Üniversitesi Hemşirelik Fakültesi Elektronik Dergisi 2013; 6(2): 102-113.
2. Özcan İ, Öztürk S. Beyin MR görüntülerinden tümör tespiti için derin öğrenmeye dayalı hibrit modeller. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 2023; 26(3): 718-733.
3. National Brain Tumor Society. Brain tumor facts. Available at: https://braintumor.org/brain-tumors/about-brain-tumors/brain-tumor-facts/ (accessed 25 November 2025).
4. Szturm T, Beheshti I, Mahana B, Hobson DE, Goertzen A, Ko JH. Imaging cerebral glucose metabolism during dual-task walking in patients with Parkinson's disease. J Neuroimaging 2021; 31(2): 356-362.
5. de Verdier MC, et al. The 2024 Brain Tumor Segmentation (BraTS) Challenge: Glioma segmentation on post-treatment MRI. arXiv 2024; arXiv:2405.18368.
6. Prakash BV, et al. Meningioma brain tumor detection and classification using hybrid CNN method and RIDGELET transform. Sci Rep 2023; 13(1): 14522.
7. Melmed S, et al. Clinical biology of the pituitary adenoma. Endocr Rev 2022; 43(6): 1003-1037.
8. Alcantara Llaguno SR, Parada LF. Cell of origin of glioma: Biological and clinical implications. Br J Cancer 2016; 115(12): 1445-1450.

9. Zong H, Verhaak RG, Canoll P. The cellular origin for malignant glioma and prospects for clinical advancements. Expert Rev Mol Diagn 2012; 12(4): 383-394.
10. Louis DN, et al. The 2016 World Health Organization classification of tumors of the central nervous system: A summary. Acta Neuropathol 2016; 131(6): 803-820.
11. Ogasawara C, Philbrick BD, Adamson DC. Meningioma: A review of epidemiology, pathology, diagnosis, treatment, and future directions. Biomedicines 2021; 9(3): 319.
12. Markou M, Lavrentaki A, Ntali G. Recent advances in understanding and managing pituitary adenomas. Fac Rev 2023; 12: 6.
13. Omuro AM, Leite CC, Mokhtari K, Delattre JY. Pitfalls in the diagnosis of brain tumours. Lancet Neurol 2006; 5(11): 937-948.
14. Villanueva-Meyer JE, Mabray MC, Cha S. Current clinical brain tumor imaging. Neurosurgery 2017; 81(3): 397-415.
15. Abdelaziz Ismael SA, Mohammed A, Hefny H. An enhanced deep learning approach for brain cancer MRI images classification using residual networks. Artif Intell Med 2020; 102: 101779.
16. Amin J, Sharif M, Haldorai A, Yasmin M, Nayak RS. Brain tumor detection and classification using machine learning: A comprehensive survey. Complex Intell Syst 2022; 8(4): 3161-3183.
17. Anantharajan S, Gunasekaran S, Subramanian T, Venkatesh R. MRI brain tumor detection using deep learning and machine learning approaches. Meas Sens 2024; 31: 101026.
18. Alam MA, Sohel A, Hasan KM, Ahmad I. Advancing brain tumor detection using machine learning and artificial intelligence: A systematic literature review of predictive models and diagnostic accuracy. Strategic Data Management and Innovation 2024; 1(1): 37-55.
19. Dorfner FJ, Patel JB, Kalpathy-Cramer J, Gerstner ER, Bridge CP. A review of deep learning for brain tumor analysis in MRI. npj Precis Oncol 2025; 9(1): 2.
20. Ali A, Li X, Mashwani WK, Abiad M, Karim FK, Mostafa SM. Multi-class brain tumor MRI segmentation and classification using deep learning and machine learning approaches. Cancer Imaging 2025; 25: 131.
21. Ibrahim AT, et al. Hybrid attention-enhanced Xception and dynamic chaotic whale optimization for brain tumor diagnosis. Bioengineering (Basel) 2025; 12(7): 747.
22. Babar NA, et al. Brain tumor classification in MRI scans using edge computing and a shallow attention-guided CNN. Biomedicines 2025; 13(10): 2571.
23. Das T, Vishwakarma DK, Ranga V. A comparative study of deep learning models for brain tumor classification. In: 2024 4th International Conference on Intelligent Technologies (CONIT); 21-23 June 2024; Bangalore, India. IEEE; 2024. pp. 1-5.
24. Ghufran F, Kumar A, Kumar A, Kaushik A, Saxena S. Detection of brain tumors in MRI images through machine learning, deep learning, Faster R-CNN and YOLOv5. In: 2024 1st International Conference on Advances in Computing, Communication and Networking (ICAC2N); 16-17 December 2024; Greater Noida, India. IEEE; 2024. pp. 754-759.
25. Akter S, Talukder MSH, Mondal SK, Aljaidi M, Sulaiman RB, Alshammari AA. Brain tumor classification utilizing pixel distribution and spatial dependencies higher-order statistical measurements through explainable ML models. Sci Rep 2024; 14(1): 25800.
26. Oh A, et al. Machine learning approach to brain tumor detection and classification. arXiv 2024; arXiv:2410.12692.
27. Samee NA, Ahmad T, Mahmoud NF, Atteia G, Abdallah HA, Rizwan A. Clinical decision support framework for segmentation and classification of brain tumor MRIs using a U-Net and DCNN cascaded learning algorithm. Healthcare (Basel) 2022; 10(12): 2340.
28. Cristin DR, Kumar DKS, Anbhazhagan DP. Severity level classification of brain tumor based on MRI images using fractional-chicken swarm optimization algorithm. Comput J 2021; 64(10): 1514-1530.
29. Yu Z, He Q, Yang J, Luo M. A supervised ML applied classification model for brain tumors MRI. Front Pharmacol 2022; 13: 884495.
30. Hao R, Namdar K, Liu L, Khalvati F. A transfer learning-based active learning framework for brain tumor classification. Front Artif Intell 2021; 4: 635766.
31. Yu Y, et al. Characterizing brain tumor regions using texture analysis in magnetic resonance imaging. Front Neurosci 2021; 15: 634926.
32. Tariciotti L, et al. A deep learning model for preoperative differentiation of glioblastoma, brain metastasis and primary central nervous system lymphoma: A pilot study. Front Oncol 2022; 12: 816638.
33. Kadam A, Bhuvaji S, Deshpande S. Brain tumor classification using deep learning algorithms. Int J Res Appl Sci Eng Technol 2021; 9(12): 417-426.
34. Kessler G. The impact of MD5 file hash collisions on digital forensic imaging. J Digit Forensics Secur Law 2016; 11(4): 129-138.
35. Kumar MR, Jeya IJS, Athira S. Analysis of encryption techniques for secure data in cloud environment. In: 2025 International Conference on Intelligent Systems and Computational Networks (ICISCN); 24-25 January 2025; Bidar, India. IEEE; 2025. pp. 1-4.
36. Alhussen A, Arslan E. Avoiding data loss and corruption for file transfers with fast integrity verification. J Parallel Distrib Comput 2021; 152: 33-44.
37. Shaikh MS, Biswal A, Pandwal A, Khodifad N, Vaghela B. Image forgery detection using MD5 and OpenCV. In: 2024 International Conference on Emerging Research in Computational Science (ICERCS); 12-14 December 2024; Coimbatore, India. IEEE; 2024. pp. 1-8.
38. Farias F, Ludermir T, Bastos-Filho C. Similarity based stratified splitting: An approach to train better classifiers. arXiv 2020; arXiv:2010.06099.
39. Sadaiyandi J, Arumugam P, Sangaiah AK, Zhang C. Stratified sampling-based deep learning approach to increase prediction accuracy of unbalanced dataset. Electronics 2023; 12(21): 4423.
40. Ammari S, et al. Influence of magnetic field strength on magnetic resonance imaging radiomics features in brain imaging, an in vitro and in vivo study. Front Oncol 2021; 10: 541663.
41. Bagci U, Udupa JK, Bai L. The influence of intensity standardization on medical image registration. In: Medical Imaging 2010: Image-Guided Procedures, Robotic Interventions, and Modeling; 13-18 February 2010; San Diego, CA, USA. Proc SPIE 7625; 2010. p. 76251X.
42. Gonca M, Sazak Ç, Gündoğdu Ş. Effects of contrast limited adaptive histogram equalization (CLAHE) on manual and automated tracing of lateral cephalometric radiographs. Clin Exp Health Sci 2024; 14(3): 733-744.
43. Tan M, Le QV. EfficientNet: Rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th International Conference on Machine Learning; 9-15 June 2019; Long Beach, CA, USA. PMLR 97; 2019. pp. 6105-6114.
44. Li K, Wang F, Yang L, Liu R. Deep feature screening: Feature selection for ultra-high-dimensional data via deep neural networks. arXiv 2023; arXiv:2204.01682.
45. Liu R, Gillies DF. Overfitting in linear feature extraction for classification of high-dimensional image data. Pattern Recognit 2016; 53: 73-86.
46. Peng H, Long F, Ding C. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 2005; 27(8): 1226-1238.
47. Öztürk N. Meta-sezgisel algoritma destekli makine öğrenmesi yöntemiyle tiroid hastalığının tespitinde yeni bir yaklaşım. Afyon Kocatepe Univ J Sci Eng 2025; 25(6): 1336-1347.
48. Toker KG, Yüksel SE. Accelerating classification time in hyperspectral images. In: 2015 23rd Signal Processing and Communications Applications Conference (SIU); 16-19 May 2015; Malatya, Türkiye. IEEE; 2015. pp. 2126-2129.
49. Kutlu F, Göleli K, Demir H, Erdiz G. Bulanık kümeleme ve sınıflandırma teknikleri ile müşteri davranış segmentasyonu. YYU JINAS 2025; 30(3): 990-1008.
50. Chen S, Bowman FD. A novel support vector classifier for longitudinal high-dimensional data and its application to neuroimaging data. Stat Anal Data Min 2011; 4(6): 604-611.
51. Hsu D, Muthukumar V, Xu J. On the proliferation of support vectors in high dimensions. In: Proceedings of the 24th International Conference on Artificial Intelligence and Statistics; 13-15 April 2021; virtual conference. PMLR 130; 2021. pp. 91-99.
52. Halabaku E, Bytyçi E. Overfitting in machine learning: A comparative analysis of decision trees and random forests. Intell Autom Soft Comput 2024; 39(6): 987-1006.
53. Wiens M, Verone-Boyle A, Henscheid N, Podichetty JT, Burton J. A tutorial and use case example of the eXtreme Gradient Boosting (XGBoost) artificial intelligence algorithm for drug development applications. Clin Transl Sci 2025; 18(3): e70172.
54. Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu TY. LightGBM: A highly efficient gradient boosting decision tree. In: Advances in Neural Information Processing Systems 30; 4-9 December 2017; Long Beach, CA, USA. Curran Associates; 2017. pp. 3146-3154.
55. Akiba T, Sano S, Yanase T, Ohta T, Koyama M. Optuna: A next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 4-8 August 2019; Anchorage, AK, USA. ACM; 2019. pp. 2623-2631.
56. Snoek J, Larochelle H, Adams RP. Practical Bayesian optimization of machine learning algorithms. In: Advances in Neural Information Processing Systems 25; 3-6 December 2012; Lake Tahoe, NV, USA. Curran Associates; 2012. pp. 2960-2968.
57. Sinap V. Optuna tabanlı hiper parametre optimizasyonu ile konut fiyat tahminlemede makine öğrenmesi tekniklerinin karşılaştırmalı analizi. Gazi Univ J Sci Part C Des Technol 2025; 13(1): 10-28.
58. Canbay Y, Adsiz S, Canbay P. Privacy-preserving transfer learning framework for kidney disease detection. Appl Sci 2024; 14(19): 8629.