Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets
Abstract Accurate detection of brain tumors remains a significant challenge due to the diversity of tumor types along with human interventions during diagnostic process. This study proposes a novel ensemble deep learning system for accurate brain tumor classification using MRI data. The proposed sys...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-14917-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849332874441916416 |
|---|---|
| author | Retinderdeep Singh Sheifali Gupta Ashraf Osman Ibrahim Lubna A. Gabralla Salil Bharany Ateeq Ur Rehman Seada Hussen |
| author_facet | Retinderdeep Singh Sheifali Gupta Ashraf Osman Ibrahim Lubna A. Gabralla Salil Bharany Ateeq Ur Rehman Seada Hussen |
| author_sort | Retinderdeep Singh |
| collection | DOAJ |
| description | Abstract Accurate detection of brain tumors remains a significant challenge due to the diversity of tumor types along with human interventions during diagnostic process. This study proposes a novel ensemble deep learning system for accurate brain tumor classification using MRI data. The proposed system integrates fine-tuned Convolutional Neural Network (CNN), ResNet-50 and EfficientNet-B5 to create a dynamic ensemble framework that addresses existing challenges. An adaptive dynamic weight distribution strategy is employed during training to optimize the contribution of each networks in the framework. To address class imbalance and improve model generalization, a customized weighted cross-entropy loss function is incorporated. The model obtains improved interpretability through explainabile artificial intelligence (XAI) techniques, including Grad-CAM, SHAP, SmoothGrad, and LIME, providing deeper insights into prediction rationale. The proposed system achieves a classification accuracy of 99.4% on the test set, 99.48% on the validation set, and 99.31% in cross-dataset validation. Furthermore, entropy-based uncertainty analysis quantifies prediction confidence, yielding an average entropy of 0.3093 and effectively identifying uncertain predictions to mitigate diagnostic errors. Overall, the proposed framework demonstrates high accuracy, robustness, and interpretability, highlighting its potential for integration into automated brain tumor diagnosis systems. |
| format | Article |
| id | doaj-art-267cca1811914bb9bbfca537dea9ef18 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-267cca1811914bb9bbfca537dea9ef182025-08-20T03:46:04ZengNature PortfolioScientific Reports2045-23222025-08-0115113510.1038/s41598-025-14917-wAdvanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasetsRetinderdeep Singh0Sheifali Gupta1Ashraf Osman Ibrahim2Lubna A. Gabralla3Salil Bharany4Ateeq Ur Rehman5Seada Hussen6Chitkara University Institute of Engineering and Technology, Chitkara UniversityChitkara University Institute of Engineering and Technology, Chitkara UniversityDepartment of Computing, Universiti Teknologi PETRONASDepartment of Computer Science, Applied College, Princess Nourah bint Abdulrahman UniversityChitkara University Institute of Engineering and Technology, Chitkara UniversitySchool of Computing, Gachon UniversityDepartment of Electrical Power, Adama Science and Technology UniversityAbstract Accurate detection of brain tumors remains a significant challenge due to the diversity of tumor types along with human interventions during diagnostic process. This study proposes a novel ensemble deep learning system for accurate brain tumor classification using MRI data. The proposed system integrates fine-tuned Convolutional Neural Network (CNN), ResNet-50 and EfficientNet-B5 to create a dynamic ensemble framework that addresses existing challenges. An adaptive dynamic weight distribution strategy is employed during training to optimize the contribution of each networks in the framework. To address class imbalance and improve model generalization, a customized weighted cross-entropy loss function is incorporated. The model obtains improved interpretability through explainabile artificial intelligence (XAI) techniques, including Grad-CAM, SHAP, SmoothGrad, and LIME, providing deeper insights into prediction rationale. The proposed system achieves a classification accuracy of 99.4% on the test set, 99.48% on the validation set, and 99.31% in cross-dataset validation. Furthermore, entropy-based uncertainty analysis quantifies prediction confidence, yielding an average entropy of 0.3093 and effectively identifying uncertain predictions to mitigate diagnostic errors. Overall, the proposed framework demonstrates high accuracy, robustness, and interpretability, highlighting its potential for integration into automated brain tumor diagnosis systems.https://doi.org/10.1038/s41598-025-14917-wBrain tumorEfficientNetResNetEnsemble modelDynamic weightsExplainable AI |
| spellingShingle | Retinderdeep Singh Sheifali Gupta Ashraf Osman Ibrahim Lubna A. Gabralla Salil Bharany Ateeq Ur Rehman Seada Hussen Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets Scientific Reports Brain tumor EfficientNet ResNet Ensemble model Dynamic weights Explainable AI |
| title | Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| title_full | Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| title_fullStr | Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| title_full_unstemmed | Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| title_short | Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| title_sort | advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets |
| topic | Brain tumor EfficientNet ResNet Ensemble model Dynamic weights Explainable AI |
| url | https://doi.org/10.1038/s41598-025-14917-w |
| work_keys_str_mv | AT retinderdeepsingh advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT sheifaligupta advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT ashrafosmanibrahim advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT lubnaagabralla advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT salilbharany advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT ateequrrehman advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets AT seadahussen advanceddynamicensembleframeworkwithexplainabilitydriveninsightsforprecisionbraintumorclassificationacrossdatasets |