Recognition of honeycomb lung in CT images based on improved MobileNet model

Abstract

Purpose The research is to improve the efficiency and accuracy of recognition of honeycomb lung in CT images. Methods Deep learning methods are used to achieve automatic recognition of honeycomb lung in CT images, however, are time consuming and less accurate due to the large amount of structural parameters. In this paper, a novel recognition method based on MobileNetV1 network, multiscale feature fusion method (MSFF), and dilated convolution is explored to deal with honeycomb lung in CT image classification. Firstly, the dilated convolution with different dilated rate is used to extract features to obtain receptive fields of different sizes, and then fuse the features of different scales at multiscale feature fusion block is used to solve the problem of feature loss and incomplete feature extraction. After that, by using linear activation functions (Sigmoid) instead of nonlinear activation functions (ReLu) in the improved deep separable convolution blocks to retain the feature information of each channel. Finally, by reducing the number of improved deep separable blocks to reduce the computation and resource consumption of the model. Results The experimental results show that improved MobileNet model has the best performance and the potential for recognition of honeycomb lung image datasets, which includes 6318 images. By comparing with 4 traditional models (SVM, RF, decision tree, and KNN) and 11 deep learning models (LeNet‐5, AlexNet, VGG‐16, GoogleNet, ResNet18, DenseNet121, SENet18, InceptionV3, InceptionV4, Xception, and MobileNetV1), our model achieved the performance with an accuracy of 99.52%, a sensitivity of 99.35%, and a specificity of 99.89%. Conclusion Improved MobileNet model is designed for the automatic recognition and classification of honeycomb lung in CT images. Through experiments comparative analysis of other models of machine learning and deep learning, it is proved that the proposed improved MobileNet method has the best recognition accuracy with fewer the model parameters and less the calculation time.

Diagnostic performance of corona virus disease 2019 chest computer tomography image recognition based on deep learning: Systematic review and meta-analysis

“…During the outbreak of COVID-19, some models can achieve an accuracy nearly 100% in classifying COVID-19 positive cases from combined Pneumonia and healthy cases. [45][46][47] According to the review, Resnet has the best performance in detecting the nidus, however, this finding is also confirmed in previous studies. [48] Some studies indicated the combination of different models can improve the speed and efficiency.…”

Semi-Supervised Deep Learning Semantic Segmentation for 3D Volumetric Computed Tomographic Scoring of Chronic Rhinosinusitis: Clinical Correlations and Comparison with Lund-Mackay Scoring

“…Compared to the above approaches, our architecture was augmented to expand the effective receptive field and calculate context information both in high-and low-resolution feature maps. Similar to DeepLab-V3+, MobileNet, which was used in this study, employs depthwise separable and pointwise convolutions for concatenation in up-sampled operators, which results in a faster and stronger network [ 15 ]. The dilated convolutions and atrous SPP can expand the receptive field that helps to further integrate information around the sinuses.…”

Random Graph-Based Neuromorphic Learning with a Layer-Weaken Structure

“…Accordingly, their originative internal connections have made them outperform many traditional NNs. Advancing this trend, as mentioned above, ResNet [13,14] and MobileNet [15,16] show more powerful abilities in classifying images. Moreover, neural architecture search (NAS) [17] in deep learning has emerged as a promising direction for optimizing the wiring patterns.…”