Background: Lung ultrasound (LUS) can be an important imaging tool for the diagnosis and assessment of lung involvement. In this study, we determined the ultrasound manifestations of the lung associated with COVID-19 pneumonia, and obtained the ultrasound image changes of the patients from the initial diagnosis to rehabilitation. Methods: The purpose of this study is to establish a lung involvement assessment model based on deep learning. A channel attention classification method based on squeeze-and-excitation network combining with ResNeXt (SE_ResNeXt) is proposed, which can automatically learn the importance of different channel features, so as to achieve selective learning of channels and further achieve more accurate classification results. Results and conclusion: Among 104 patients' data from multicenter and multi-mode ultrasound, the diagnostic model can achieve 97.11% accuracy. The lung involvement severity of COVID-19 pneumonia and the trend of lesion were evaluated quantitatively.

Elastography ultrasound (EUS) imaging is a vital ultrasound imaging modality. The current use of EUS faces many challenges, such as vulnerability to subjective manipulation, echo signal attenuation, and unknown risks of elastic pressure in certain delicate tissues. The hardware requirement of EUS also hinders the trend of miniaturization of ultrasound equipment. We therefore present a cost-efficient solution by designing a deep neural network to synthesize augmented reality EUS (AR-EUS) from conventional B-mode images. By using 4580 cases from 15 medical centers, we evaluate the performance of AR-EUS on breast cancer diagnosis. The quantitative metric and blind evaluation results show no significant difference between AR-EUS and real EUS in image authenticity and in clinical diagnosis. The performance of pocket-sized ultrasound in breast tumor diagnosis is also significantly improved after AR-EUS is equipped. These results highlight the potential of AR-EUS in clinical application.

Background: Lung ultrasound (LUS) can be an important imaging tool for the diagnosis and assessment of lung involvement. Ultrasound sonograms have been confirmed to illustrate damage to a person’s lungs, which means that the correct classification and scoring of a patient’s sonogram can be used to assess lung involvement.Methods: The purpose of this study was to establish a lung involvement assessment model based on deep learning. A novel multimodal channel and receptive field attention network combined with ResNeXt (MCRFNet) was proposed to classify sonograms, and the network can automatically fuse shallow features and determine the importance of different channels and respective fields. Finally, sonogram classes were transformed into scores to evaluate lung involvement from the initial diagnosis to rehabilitation.Results and conclusion: Using multicenter and multimodal ultrasound data from 104 patients, the diagnostic model achieved 94.39% accuracy, 82.28% precision, 76.27% sensitivity, and 96.44% specificity. The lung involvement severity and the trend of COVID-19 pneumonia were evaluated quantitatively.