Epidemiologic Characteristics of COVID-19 in Guizhou Province, China.

INTRODUCTION: At the end of 2019, the COVID-19 broke out, and spread to Guizhou province in January of 2020. METHODOLOGY: To acquire the epidemiologic characteristics of COVID-19 in Guizhou province, we collected data from 169 laboratory-confirmed COVID-19 related cases. We described the demographic characteristics of the cases and estimated the incubation period, serial interval and the effective reproduction number. We also presented two representative case studies in Guizhou province: Case Study 1 was an example of the asymptomatic carrier; while Case Study 2 was an example of a large and complex infection chain that involved four different regions, spanning three provinces and eight families. RESULTS: Two peaks in the incidence distribution associated with COVID-19 in Guizhou province were related to the 6.04 days (95% CI: 5.00 - 7.10) of incubation period and 6.14±2.21 days of serial interval. We also discussed the effectiveness of the control measures based on the instantaneous effective reproduction number that was a constantly declining curve. CONCLUSIONS: As of February 2, 2020, the estimated effective reproduction number was below 1, and no new cases were reported since February 26. These showed that Guizhou Province had achieved significant progress in preventing the spread of the epidemic. The medical isolation of close contacts was consequential. Meanwhile, the asymptomatic carriers and the super-spreaders must be isolated in time, who would cause a widespread infection.

Learning Invariant Representations across Domains and Tasks (preprint)

Being expensive and time-consuming to collect massive COVID-19 image samples to train deep classification models, transfer learning is a promising approach by transferring knowledge from the abundant typical pneumonia datasets for COVID-19 image classification. However, negative transfer may deteriorate the performance due to the feature distribution divergence between two datasets and task semantic difference in diagnosing pneumonia and COVID-19 that rely on different characteristics. It is even more challenging when the target dataset has no labels available, i.e., unsupervised task transfer learning. In this paper, we propose a novel Task Adaptation Network (TAN) to solve this unsupervised task transfer problem. In addition to learning transferable features via domain-adversarial training, we propose a novel task semantic adaptor that uses the learning-to-learn strategy to adapt the task semantics. Experiments on three public COVID-19 datasets demonstrate that our proposed method achieves superior performance. Especially on COVID-DA dataset, TAN significantly increases the recall and F1 score by 5.0% and 7.8% compared to recently strong baselines. Moreover, we show that TAN also achieves superior performance on several public domain adaptation benchmarks.

Learning to Match Distributions for Domain Adaptation (preprint)

When the training and test data are from different distributions, domain adaptation is needed to reduce dataset bias to improve the model's generalization ability. Since it is difficult to directly match the cross-domain joint distributions, existing methods tend to reduce the marginal or conditional distribution divergence using predefined distances such as MMD and adversarial-based discrepancies. However, it remains challenging to determine which method is suitable for a given application since they are built with certain priors or bias. Thus they may fail to uncover the underlying relationship between transferable features and joint distributions. This paper proposes Learning to Match (L2M) to automatically learn the cross-domain distribution matching without relying on hand-crafted priors on the matching loss. Instead, L2M reduces the inductive bias by using a meta-network to learn the distribution matching loss in a data-driven way. L2M is a general framework that unifies task-independent and human-designed matching features. We design a novel optimization algorithm for this challenging objective with self-supervised label propagation. Experiments on public datasets substantiate the superiority of L2M over SOTA methods. Moreover, we apply L2M to transfer from pneumonia to COVID-19 chest X-ray images with remarkable performance. L2M can also be extended in other distribution matching applications where we show in a trial experiment that L2M generates more realistic and sharper MNIST samples.

COVID-19 patients with hypertension under potential risk of worsened organ injuries (preprint)

COVID -19 has rapidly spread from Wuhan to worldwide, and now has become a global health concern. Hypertension is the most common chronic illness in COVID-19, while the influence on those patients have not been well described. In this retrospective study, 82 confirmed patients with COVID-19 were enrolled, with epidemiological, demographic, clinical, laboratory, radiological, and therapies data analyzed and compared between COVID-19 patients with (29 cases) or without (53 cases) hypertension. Of all 82 patients with COVID-19, the median age of all patients was 60.5 years, including 49 females (59.8%) and 33 (40.2%) males. Hypertension (31[28.2%]) was the most chronic illness, followed by diabetes (16 [19.5%]) and cardiovascular disease (15 [18.3%]). Common symptoms included fatigue (55[67.1%]), dry cough (46 [56.1%]) and fever (≥37.3℃ (46 [56.1%]). The median time from illness onset to positive outcomes of RT-PCR analysis were 13.0 days, ranging from 3-25 days. In hypertension group, 6 (20.7%) patients died compared to 5 (9.4%) died in non-hypertension group. More hypertension patients with COVID-19 (8 [27.6%]) had at least one coexisting disease than those of non-hypertension patients (2 [3.8%]) (P=0.002). Compared with non-hypertension patients, higher levels of neutrophil counts, serum amyloid A, C-reactive protein, and NT-proBNP were observed in hypertension group, whereas levels of lymphocyte count and eGFR were decreased. Dynamic observations displayed more significant and worsened outcomes in hypertension group after hospital admission. COVID-19 patients with hypertension take more risks of severe inflammatory reactions, worsened internal organ injuries, and deteriorated progress.