https://doi.org/10.1140/epjp/s13360-025-07125-y
Regular Article
Fossa ship rescue optimization-enabled 3D convolutional neural network for severity-level classification of brain tumor using 3D-MRI images
1
Department of Electronics and Communication Engineering, BLDEA’s V. P. Dr. P. G. Halakatti College of Engineering and Technology, 586103, Vijayapura, Karnataka, India
2
Visvesvaraya Technological University, 590018, Belagavi, Karnataka, India
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
3
July
2025
Accepted:
27
November
2025
Published online:
18
December
2025
Abstract
Brain tumors arise from the abnormal growth of cells in the brain, often invading surrounding tissues and forming metastases, making them life-threatening. Effective treatment planning depends on an accurate diagnosis. Classical brain tumor recognition models often underperform due to inadequate feature selection and overfitting on limited datasets, which affects severity grading. To address this limitation, this paper presents a Fossa Ship Rescue Optimization_3D Convolutional Neural Network (FSRO_3D-CNN) for brain tumor severity classification. The tumor region is then segmented using SwinBTS trained with the proposed fossa ship rescue optimization (FSRO), which integrates the fossa optimization algorithm (FOA) and ship rescue optimization (SRO). After segmentation, feature extraction is performed to obtain effective discriminative representations. Finally, severity-level classification is carried out using a 3D-CNN, with hyperparameters optimized by FSRO. Experiments on the BRATS-2018 dataset show that the method achieves 92.82% accuracy, 91.87% true-positive rate (TPR), 93.86% true-negative rate (TNR), and a 92.861% F1-score, demonstrating strong capability for automated brain tumor severity assessment.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
