@article{Wang_Lima_2021, title={Multiple sclerosis detection via 6-layer stochastic pooling convolutional neural network and multiple-way data augmentation}, volume={2}, url={https://stemedicine.org/index.php/stem/article/view/101}, DOI={10.37175/stemedicine.v2i8.101}, abstractNote={<p class="Indent" style="text-align: justify; text-justify: inter-ideograph; text-indent: 20.0pt;">Background: Multiple sclerosis is one of most widespread autoimmune neuroinflammatory diseases<br>which mainly damages body function such as movement, sensation, and vision. Despite of conventional<br>clinical presentation, brain magnetic resonance imaging of white matter lesions is often applied to<br>diagnose multiple sclerosis at the early stage.<br>Methods: In this article, we proposed a 6-layer stochastic pooling convolutional neural network (CNN)<br>with multiple-way data augmentation for multiple sclerosis detection in brain magnetic resonance imaging.<br>Our approach does not demand hand-crafted features unlike those traditional machine learning methods.<br>Via application of stochastic pooling and multiple-way data augmentation, our 6-layer CNN achieved<br>equivalent performance against those deep learning methods which consist of so many layers and<br>parameters that ordinarily bring difficulty to training. Further, we also conducted ablation experiments to<br>examine the contribution of stochastic pooling and multiple-way data augmentation to the original CNN<br>model.<br>Results: The results showed that this 6-layer CNN obtained a sensitivity of 95.98 ± 0.46%, a specificity of<br>95.67 ± 0.92%, and an accuracy of 95.82 ± 0.58%. According to comparison experiments, our results are<br>better than state-of-the-art approaches.<br>Conclusion: Our scheme of stochastic pooling and multiple-way data augmentation enhanced the original<br>6-layer CNN model compared to those using maximum pooling or average pooling and inadequate data<br>augmentation</p&gt;}, number={8}, journal={STEMedicine}, author={Wang, Jian and Lima, Dimas}, year={2021}, month={Sep.}, pages={e101} }