article · Neural Computing and Applications
Brain tumours represent a severe health risk where early, accurate detection is vital for patient recovery. Training deep learning networks from scratch often requires weeks of computation. To address this, an ensemble transfer learning approach combines modified versions of the pre-trained DenseNet121 and InceptionV3 architectures. The original classifiers within these networks were replaced to tailor them to tumour classification. The system detects three specific tumour types: meningioma, glioma, and pituitary tumours. To provide transparency and trust in diagnostic outcomes, the approach integrates gradient-weighted class activation maps, known as Grad-CAM, which visually explain the model decisions. Tested against a publicly available magnetic resonance imaging dataset, the ensemble system attained 99.02% accuracy, 98.75% precision, 98.98% recall, and an F1 score of 98.86%, outperforming other methods while offering explainable results.
Accurate early identification of brain tumours significantly improves patient survival rates. Because deep learning systems are often viewed as opaque black boxes, clinicians can be reluctant to rely on them. Integrating explainable visual maps directly into highly accurate diagnostic tools helps medical practitioners verify artificial intelligence assessments, providing a dependable second opinion during clinical decision-making.
This research could support software tools for radiologists and clinical oncologists analysing brain MRI scans. The integration of explainability features provides a clear pathway towards clinical decision-support systems. Given that the tool has been validated solely on a publicly available benchmark dataset, the technology sits at an applied research stage and requires clinical validation and regulatory clearance before deployment in hospitals.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Brain tumors are very dangerous as they cause death. A lot of people die every year because of brain tumors. Therefore, accurate classification and detection in the early stages can help in recovery. Various deep learning techniques have achieved good results in brain tumor classification. The traditional deep learning methods and training the neural network from scratch are time-consuming and can last for weeks of training. Therefore, in this work, we proposed an ensemble approach depending on transfer learning that utilizes pre-trained models of DenseNet121 and InceptionV3 to detect three forms of brain tumors: meningioma, glioma, and pituitary. While developing the ensemble model, some changes were made to the architecture of pre-trained models by replacing their classifiers (fully connected and SoftMax layers) with a new classifier to adopt the recent task. In addition, gradient-weighted class activation maps (Grad-CAM) are an explainable model to verify results and achieve high confidence. The suggested model was validated using a publicly available dataset and achieved 99.02% accuracy, 98.75% precision, 98.98% recall, and a 98.86% F1 score. The suggested approach outperformed others in detecting and classifying brain tumor MRI data, and verifying results using the explainable model achieved a high degree of trust.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1007/s00521-024-10401-0
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
