article · BMC Medical Imaging
Conventional diagnosis and classification of brain tumours rely on histological evaluation of biopsy samples, a method that is invasive, slow, and prone to manual error. To address the need for automated solutions, three distinct deep convolutional neural network models were developed for specific diagnostic stages. The first model detects the presence of brain tumours with 99.53 per cent accuracy. The second model categorises brain scans into five groups, comprising normal tissue, glioma, meningioma, pituitary, and metastatic tumours, achieving 93.81 per cent accuracy. The third model identifies tumour grades with 98.56 per cent accuracy. Hyperparameters across all models were automatically tuned using grid search optimisation. When evaluated on large, publicly available clinical datasets, these models outperformed several established architectures, including AlexNet, DenseNet121, ResNet-101, VGG-19, and GoogleNet.
Standard brain tumour diagnosis depends heavily on invasive biopsies that take time and are vulnerable to human error. Reliable automated image classification offers a non-invasive alternative that can accelerate early detection and grading. By accurately distinguishing between several tumour types and stages, deep learning tools can assist healthcare providers in establishing faster, more consistent diagnostic workflows.
This work could enable automated diagnostic decision-support software for radiologists and clinical oncologists analysing brain scans. Based on tests using publicly available clinical datasets, the technology sits at an applied and tested research stage. Moving towards real-world hospital deployment would require further clinical integration and regulatory clearance.
AI-generated from the published abstract. Always read the original work before citing.
The current approach to diagnosing and classifying brain tumors relies on the histological evaluation of biopsy samples, which is invasive, time-consuming, and susceptible to manual errors. These limitations underscore the pressing need for a fully automated, deep-learning-based multi-classification system for brain malignancies. This article aims to leverage a deep convolutional neural network (CNN) to enhance early detection and presents three distinct CNN models designed for different types of classification tasks. The first CNN model achieves an impressive detection accuracy of 99.53% for brain tumors. The second CNN model, with an accuracy of 93.81%, proficiently categorizes brain tumors into five distinct types: normal, glioma, meningioma, pituitary, and metastatic. Furthermore, the third CNN model demonstrates an accuracy of 98.56% in accurately classifying brain tumors into their different grades. To ensure optimal performance, a grid search optimization approach is employed to automatically fine-tune all the relevant hyperparameters of the CNN models. The utilization of large, publicly accessible clinical datasets results in robust and reliable classification outcomes. This article conducts a comprehensive comparison of the proposed models against classical models, such as AlexNet, DenseNet121, ResNet-101, VGG-19, and GoogleNet, reaffirming the superiority of the deep CNN-based approach in advancing the field of brain tumor classification and early detection.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1186/s12880-024-01195-7
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.
