AUTHOR=Wang Shui-Hua , Zhu Ziquan , Zhang Yu-Dong 

TITLE=RETRACTED: PSCNN: PatchShuffle Convolutional Neural Network for COVID-19 Explainable Diagnosis

JOURNAL=Frontiers in Public Health

VOLUME=Volume 9 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2021.768278

DOI=10.3389/fpubh.2021.768278

ISSN=2296-2565

ABSTRACT=(Aim) COVID-19 is a sort of infectious disease caused by a new strain of coronavirus. This study aims to develop a more accurate COVID-19 diagnosis system. (Methods) First, the n-conv module (nCM) is introduced. Then we build a 12-layer convolutional neural network (12l-CNN) as the backbone network. Afterwards, PatchShuffle is introduced to integrate with 12l-CNN as a regularization term of the loss function. Our model is named PSCNN. Moreover, multiple-way data augmentation and Grad-CAM are employed to avoid overfitting and locating lung lesions. (Results) The mean and standard variation values of the seven measures of our The MSD values of the seven measures of our model are 95.28±1.03 (sensitivity), 95.78±0.87 (specificity), 95.76±0.86 (precision), 95.53±0.83 (accuracy), 95.52±0.83 (F1 score), 91.07±1.65 (MCC), and 95.52±0.83 (FMI). (Conclusion) Our PSCNN is better than ten state-of-the-art models. Further, we validate the optimal hyperparameters in our model and demonstrate the effectiveness of PatchShuffle.