Performance optimisation of deep learning models using majority voting algorithm for brain tumour classification

• Published in: Computers in biology and medicine Vol. 135; p. 104564
• Main Authors: Tandel, Gopal S., Tiwari, Ashish, Kakde, O.G.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.08.2021; Elsevier Limited
• Subjects: Accuracy; Algorithms; Artificial intelligence; Artificial neural networks; Bayesian analysis; Biopsy; Brain; Brain cancer; Brain tumors; Classification; Computer-aided diagnosis; Convolutional neural network; Datasets; Decision trees; Deep learning; Ensemble; Feature selection; Image classification; Internal Medicine; Learning algorithms; Literature reviews; Machine learning; Magnetic resonance imaging; Majority voting; Medical diagnosis; Metastasis; Methods; Neural networks; Neuroimaging; Optimization; Other; Principal components analysis; Spinal cord; Support vector machines; Transfer learning; Tumors; Women; Deep learning; Magnetic resonance imaging; Transfer learning; Convolutional neural network; Ensemble; Machine learning; Computer-aided diagnosis; Majority voting
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104564

Although biopsy is the gold standard for tumour grading, being invasive, this procedure also proves fatal to the brain. Thus, non-invasive methods for brain tumour grading are urgently needed. Here, a magnetic resonance imaging (MRI)-based non-invasive brain tumour grading method has been proposed using deep learning (DL) and machine learning (ML) techniques. Four clinically applicable datasets were designed. The four datasets were trained and tested on five DL-based models (convolutional neural networks), AlexNet, VGG16, ResNet18, GoogleNet, and ResNet50, and five ML-based models, Support Vector Machine, K-Nearest Neighbours, Naïve Bayes, Decision Tree, and Linear Discrimination using five-fold cross-validation. A majority voting (MajVot)-based ensemble algorithm has been proposed to optimise the overall classification performance of five DL and five ML-based models. The average accuracy improvement of four datasets using the DL-based MajVot algorithm against AlexNet, VGG16, ResNet18, GoogleNet, and ResNet50 models was 2.02%, 1.11%, 1.04%, 2.67%, and 1.65%, respectively. Further, a 10.12% improvement was seen in the average accuracy of four datasets using the DL method against ML. Furthermore, the proposed DL-based MajVot algorithm was validated on synthetic face data and improved the male versus female face image classification accuracy by 2.88%, 0.71%, 1.90%, 2.24%, and 0.35% against AlexNet, VGG16, ResNet18, GoogleNet, and ResNet50, respectively. The proposed MajVot algorithm achieved promising results for brain tumour classification and is able to utilise the combined potential of multiple models. •A computer-aided diagnostic tool is proposed for automated brain tumour classification using MRI.•Machine and deep learning-based ensemble algorithms are proposed to enhance the performance of many models.•The proposed method is validated on synthetic face data.•The effect of convolutional layers is analysed on accuracy and training time.