EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable Deep Ensemble Meta-Learning Framework for Brain Tumor Classification Using MRI Images

Shawon Chakrabarty Kakon iD, Zawad Al Sazid, Ismat Ara Begum, Md Abdus Samad iD, A S M Sanwar Hosen iD

Cancers · 30.08.2025 · 17 (17) · 2853

DOI: 10.3390/cancers17172853 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cancers, online erschienen am 30.08.2025. Verfasst von Shawon Chakrabarty Kakon und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Artificial Intelligence and Big Data, Woosong University, Daejeon 34606, Republic of Korea. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Woosong University Academic Research Fund 2025, South Korea, Woosong University Academic Research Fund 2025, Republic of Korea.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Brain tumors can severely impair neurological function, leading to symptoms such as headaches, memory loss, motor coordination deficits, and visual disturbances. In severe cases, they may cause permanent cognitive damage or become life-threatening without early detection.

Methods

To address this, we propose an interpretable deep ensemble model for tumor detection in Magnetic Resonance Imaging (MRI) by integrating pre-trained Convolutional Neural Networks—EfficientNetB7, InceptionV3, and Xception—using a soft voting ensemble to improve classification accuracy. The framework is further enhanced with a Light Gradient Boosting Machine as a meta-learner to increase prediction accuracy and robustness within a stacking architecture. Hyperparameter tuning is conducted using Optuna, and overfitting is mitigated through batch normalization, L2 weight decay, dropout, early stopping, and extensive data augmentation.

Results

These regularization strategies significantly enhance the model’s generalization ability within the BR35H dataset. The framework achieves a classification accuracy of 99.83 on the MRI dataset of 3060 images.

Conclusions

To improve interpretability and build clinical trust, Explainable Artificial Intelligence methods Grad-CAM++, LIME, and SHAP are employed to visualize the factors influencing model predictions, effectively highlighting tumor regions within MRI scans. This establishes a strong foundation for further advancements in radiology decision support systems.

Schlagwörter

Stichwörter der Autor:innen

MRI imagesbrain tumor detectiondeep learningexplainable artificial intelligence

Autor:innen und Institutionen

  • Department of Artificial Intelligence and Big Data, Woosong University, Daejeon 34606, Republic of Korea.
  • Department of Biomedical Sciences and Institute for Medical Science, Jeonbuk National University Medical School, Jeonju 54907, Republic of Korea.
  • Department of Information and Communication Engineering, Yeungnam University, Gyeongsan 38541, Republic of Korea.

ORCID-Profile: Shawon Chakrabarty Kakon, Md Abdus Samad, A S M Sanwar Hosen

Förderung

  • Woosong University Academic Research Fund 2025, South Korea
  • Woosong University Academic Research Fund 2025, Republic of Korea

Zitieren

APA

Kakon, S. C., Sazid, Z. A., Begum, I. A., Samad, M. A., & Hosen, A. S. M. S. (2025). Explainable Deep Ensemble Meta-Learning Framework for Brain Tumor Classification Using MRI Images. Cancers, 17(17), 2853. https://doi.org/10.3390/cancers17172853

Vancouver

Kakon SC, Sazid ZA, Begum IA, Samad MA, Hosen ASMS. Explainable Deep Ensemble Meta-Learning Framework for Brain Tumor Classification Using MRI Images. Cancers. 2025;17(17):2853. doi:10.3390/cancers17172853

BibTeX
@article{Kakon2025,
  author = {Kakon, Shawon Chakrabarty and Sazid, Zawad Al and Begum, Ismat Ara and Samad, Md Abdus and Hosen, A S M Sanwar},
  title = {Explainable Deep Ensemble Meta-Learning Framework for Brain Tumor Classification Using MRI Images},
  journal = {Cancers},
  year = {2025},
  volume = {17},
  number = {17},
  pages = {2853},
  doi = {10.3390/cancers17172853}
}
Kostenlos starten →