ABSTRACT
The Traditional Chinese Medicine (TCM) has demonstrated its significant medical value over the decades, particularly during the COVID-19 pandemic. TCM-AI interdisciplinary models have been proposed to model TCM knowledge, diagnosis, and treatment experiments in clinical practice. Among them, numerous models have been developed to simulate the syndrome differentiation process of human TCM doctors for automatic syndrome diagnosis. However, these models are designed for normal scenarios and trained using a supervised learning paradigm which needs tens of thousands of training samples. They fail to effectively differentiate syndromes in rare disease scenarios where the available TCM electronic medical records (EMRs) are very limited for each unique syndrome. To address the challenge of rare diseases, this study proposes a simple yet effective method called Transfer Learning based Dual-Augmentation (TLDA). TLDA aims to augment the limited EMRs at both the sample-level and feature-level, enriching the pathological and medical information during training. Extended experiments involving 11 comparison models, including the state-of-the-art model, demonstrate the effectiveness of TLDA. TLDA outperforms all comparison models by a significant margin. Furthermore, TLDA can also be extended to other medical tasks when the EMRs for diagnosis are limited in samples.
Subject(s)
COVID-19 , Rare DiseasesABSTRACT
Based on monitoring data of atmospheric pollutants in Nanjing from Jan. 1, 2015 to Feb. 10, 2021, the spatial-temporal distribution characteristics of Nanjing's ambient air quality and the contribution of potential source areas were analyzed. The average concentrations of six air pollutants (CO, NO2 , SO2 , O3, PM10, and PM2.5) were 800 µg. m-3, 43. 1 µg . m-3, 13. 0 µg . m-3, 106. 0 µg . m-3, 77. 1 µg . m-3, and 43. 0µg . m-3, respectively. The average concentration of ozone in Nanjing was higher than that in China's other typical cities (Beijing, Shanghai, Guangzho, Chengd, Lanzhou, and Wuhan). The number of pollution days for NO2,PMl0,and PM2.5 were reduced by 29. 1% ,38. 1% ,and 28. 1% during 2015 to 2020. However, the frequency of ozone pollution days was increasing (the highest value in summer and the lowest value in winter). The potential source analysis of fine particulate matter in January of 2015-2020 was carried out. It was found that the potential source for Nanjing's PM2.5 was surrounding industrial areas (Anhui province,north of Jiangsu province, and Shandong province). The concentration of air pollutants in Nanjing in 2020 was lower than that in 2019 and 2021. It indicated that the reduction of human activity caused by COVID-19 pandemic has resulted in less air pollutant emissions and improved air quality in Nanjing.
ABSTRACT
Background The study aim was to investigate one-year all-cause mortality and health consequences of cancer COVID-19 patients in China, stratified by primary tumor subtype.Methods In this multicenter cohort study, 166 cancer COVID-19 patients were studied along with 498 gender- and age-matched non-cancer COVID-19 patients in four hospitals in Wuhan, China, admitted 2020/01/01-2020/03/18, as well as with 498 parallel gender-, age-, and cancer subtype- matched non-COVID cancer patients hospitalized between 2019/01/01-2020/03/17. All patients were followed-up with a telephone survey to assess health consequences. Cox proportional hazards regression were used for risk analysis.Results In the three cohorts of median age of 65 ± 1 year and 49% male, the one-year all-cause mortality and hospital mortality rates of Cancer COVID-19 Cohort, 30% and 20% respectively, were significantly higher than COVID-19 Cohort (9% and 8%), and Cancer Cohort (16% and 2%). The 12-month all-cause post-discharge mortality rate of Cancer COVID-19 Cohort (11%) was higher than COVID-19 Cohort (0.4%), but similar to Cancer Cohort (15%). The high 1-year all-cause mortality was among hematologic malignancies (65%) and then nasopharyngeal, brain and skin tumors (45%), digestive system (43%), and lung (32%). the rate was low among genitourinary (14%), female genital (13%), breast (11%), and thyroid (0). As for patients having at least one symptom at the 1-year follow-up, Cancer COVID-19 Cohort (23%, 26/114) is similar to COVID-19 Cohort (30%, 130/432).Discussion Cancer COVID-19 patients showed a high rate of hospitalization mortality, but not after discharged, signifying the strong acute adverse effect of COVID-19 on cancer patients while little was in long-term effect. Risk stratification showed that hematologic malignancies, nasopharyngeal, brain, digestive system and lung tumors were high risk, while genitourinary, female genital, breast and thyroid were low risk which was similar to non-cancer COVID-19.Conclusions COVID-19 had little effect of 1-year mortality and sequelae for cancer survivors discharged from SARS-CoV-2 virus infection. Different tumor subtypes had different effect of COVID-19. COVID-19 patients with thyroid, breast, female genital, genitourinary tumor had low risk mortality which was similar to non-cancer patients.