PIXEL-LEVEL AND AFFINITY-LEVEL KNOWLEDGE DISTILLATION FOR UNSUPERVISED SEGMENTATION OF COVID-19 LESIONS

Automatic segmentation of COVID-19 lesions is essential for computer-aided diagnosis. However, this task remains challenging because widely-used supervised based methods require large-scale annotated data that is difficult to obtain. Although an unsupervised method based on anomaly detection has shown promising results in [1], its performance is relatively poor. We address this problem by proposing a pixel-level and affinity-level knowledge distillation method. It obtains a pre-trained teacher network with rich semantic knowledge of CT images by constructing and training an auto-encoder at first, and then trains a student network with the same architecture as the teacher by distilling the teacher's knowledge only from normal CT images, and finally localizes COVID-19 lesions using the feature discrepancy between the teacher and the student networks. Besides, except for the traditional pixel-level distillation, we design the affinity-level distillation that takes into account the pairwise relationship of features to fully distill effective knowledge. We evaluate this method by using three different COVID-19 datasets and the experimental results show that the segmentation performance is largely improved when it is compared with the other existing unsupervised anomaly detection methods. © 2022 IEEE

Association between Decreased Social Participation and Depressive Symptom Onset among Community-Dwelling Older Adults: A Longitudinal Study during the COVID-19 Pandemic.

OBJECTIVES: The coronavirus disease (COVID-19) pandemic has imposed restrictions on people's social behavior. However, there is limited evidence regarding the relationship between changes in social participation and depressive symptom onset among older adults during the pandemic. We examined the association between changes in social participation and the onset of depressive symptoms among community-dwelling older adults during the COVID-19 pandemic. DESIGN: This was a longitudinal study. SETTING: Communities in Minokamo City, a semi-urban area in Japan. PARTICIPANTS: We recruited community-dwelling older adults aged ≥ 65 years using random sampling. Participants completed a questionnaire survey at baseline (March 2020) and follow-up (October 2020). MEASUREMENTS: Depressive symptoms were assessed using the Two-Question Screen. Based on their social participation status in March and October 2020, participants were classified into four groups: "continued participation," "decreased participation," "increased participation," and "consistent non-participation." RESULTS: A total of 597 older adults without depressive symptoms at baseline were analyzed (mean age = 79.8 years; 50.4% females). Depressive symptoms occurred in 20.1% of the participants during the observation period. Multivariable Poisson regression analysis showed that decreased social participation was significantly associated with the onset of the depressive symptoms, compared to continued participation, after adjusting for all covariates (incidence rate ratio = 1.59, 95% confidence interval = 1.01-2.50, p = 0.045). CONCLUSION: Older adults with decreased social participation during the COVID-19 pandemic demonstrated a high risk of developing depressive symptoms. We recommend that resuming community activities and promoting the participation of older adults, with sufficient consideration for infection prevention, are needed to maintain mental health among older adults.

BG-Net: Boundary-Guided Network for Lung Segmentation on Clinical CT Images

Lung segmentation on CT images is a crucial step for a computer-aided diagnosis system of lung diseases. The existing deep learning based lung segmentation methods are less efficient to segment lungs on clinical CT images, especially that the segmentation on lung boundaries is not accurate enough due to complex pulmonary opacities in practical clinics. In this paper, we propose a boundary-guided network (BG-Net) to address this problem. It contains two auxiliary branches that seperately segment lungs and extract the lung boundaries, and an aggregation branch that efficiently exploits lung boundary cues to guide the network for more accurate lung segmentation on clinical CT images. We evaluate the proposed method on a private dataset collected from the Osaka university hospital and four public datasets including StructSeg [1], HUG [2], VESSEL12 [3], and a Novel Coronavirus 2019 (COVID-19) dataset [4]. Experimental results show that the proposed method can segment lungs more accurately and outperform several other deep learning based methods.

Unsupervised Detection of Pulmonary Opacities for Computer-Aided Diagnosis of COVID-19 on CT Images

COVID-19 emerged towards the end of 2019 which was identified as a global pandemic by the world heath organization (WHO). With the rapid spread of COVID-19, the number of infected and suspected patients has increased dramatically. Chest computed tomography (CT) has been recognized as an efficient tool for the diagnosis of COVID-19. However, the huge CT data make it difficult for radiologist to fully exploit them on the diagnosis. In this paper, we propose a computer-aided diagnosis system that can automatically analyze CT images to distinguish the COVID-19 against to community-acquired pneumonia (CAP). The proposed system is based on an unsupervised pulmonary opacity detection method that locates opacity regions by a detector unsupervisedly trained from CT images with normal lung tissues. Radiomics based features are extracted insides the opacity regions, and fed into classifiers for classification. We evaluate the proposed CAD system by using 200 CT images collected from different patients in several hospitals. The accuracy, precision, recall, fl-score and AUC achieved are 95.5%, 100%, 91%, 95.1% and 95.9% respectively, exhibiting the promising capacity on the differential diagnosis of COVID-19 from CT images.