AUTHOR=Montaha Sidratul , Azam Sami , Rafid A. K. M. Rakibul Haque , Hasan Md. Zahid , Karim Asif , Hasib Khan Md. , Patel Shobhit K. , Jonkman Mirjam , Mannan Zubaer Ibna 

TITLE=MNet-10: A robust shallow convolutional neural network model performing ablation study on medical images assessing the effectiveness of applying optimal data augmentation technique

JOURNAL=Frontiers in Medicine

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2022.924979

DOI=10.3389/fmed.2022.924979

ISSN=2296-858X

ABSTRACT=The interpretation of medical images with computer-aided diagnosis (CAD) system is arduous due to the complex structure of cancerous lesions within different imaging modalities, high degree of resemblance between the inter-classes, presence of dissimilar characteristics in intra-classes, the scarcity of medical data and the presence of artifacts and noises. In this study, these challenges are addressed by developing a shallow Convolutional Neural Network (CNN) model with optimal configuration performing ablation study by altering layer structure and hyper-parameters and utilizing suitable augmentation technique. Eight medical datasets with different modalities are investigated where the proposed model, named MNet-10, with low computational complexity is able to yield optimal performance across all the datasets. The impact of photometric and geometric augmentation techniques on different datasets is also evaluated. We selected the mammogram dataset to proceed with the ablation study for being one of the most challenging imaging modalities. Before generating the model, the dataset is augmented using the two approaches. A base CNN model is constructed first and applied to both the augmented and non-augmented mammogram datasets where the highest accuracy is obtained with photometric dataset. Therefore, the architecture and hyper-parameters of the model are determined by performing ablation study on the base model using the mammogram photometric dataset. Afterwards, the robustness of the network and impact of different augmentation techniques are assessed by training the model with the rest of the seven datasets. We obtain test accuracy of 97.34% on the mammogram, 98.43% on the skin cancer, 99.54% on the brain tumor magnetic resonance imaging (MRI), 97.29% on the COVID chest X-ray, 96.31% on the tympanic membrane, 99.82% on the chest computed tomography (CT) scan, 98.75% on the breast cancer ultrasound datasets with photometric augmentation and 96.76% on the breast cancer microscopic biopsy dataset with geometric augmentation. Moreover, some elastic deformation augmentation methods are explored with the proposed model using all the datasets to evaluate its effectiveness. Finally, VGG16, InceptionV3 and ResNet50 were trained on the best-performing augmented datasets and compared the performance consistency with MNet-10 model. The findings may aid future researchers in medical data analysis involving ablation studies and augmentation techniques.