BiT-MAC: Mortality prediction by bidirectional time and multi-feature attention coupled network on multivariate irregular time series.

Mortality prediction is crucial to evaluate the severity of illness and assist in improving the prognosis of patients. In clinical settings, one way is to analyze the multivariate time series (MTSs) of patients based on their medical data, such as heart rates and invasive mean arterial blood pressure. However, this suffers from sparse, irregularly sampled, and incomplete data issues. These issues can compromise the performance of follow-up MTS-based analytic applications. Plenty of existing methods try to deal with such irregular MTSs with missing values by capturing the temporal dependencies within a time series, yet in-depth research on modeling inter-MTS couplings remains rare and lacks model interpretability. To this end, we propose a bidirectional time and multi-feature attention coupled network (BiT-MAC) to capture the temporal dependencies (i.e., intra-time series coupling) and the hidden relationships among variables (i.e., inter-time series coupling) with a bidirectional recurrent neural network and multi-head attention, respectively. The resulting intra- and inter-time series coupling representations are then fused to estimate the missing values for a more robust MTS-based prediction. We evaluate BiT-MAC by applying it to the missing-data corrupted mortality prediction on two real-world clinical datasets, i.e., PhysioNet'2012 and COVID-19. Extensive experiments demonstrate the superiority of BiT-MAC over cutting-edge models, verifying the great value of the deep and hidden relations captured by MTSs. The interpretability of features is further demonstrated through a case study.

Global burden of upper respiratory infections in 204 countries and territories, from 1990 to 2019.

BACKGROUND: Upper respiratory infections (URIs) are among the most common diseases. However, the related burden has not been comprehensively evaluated. Thus, we designed the present study to describe the global and regional burden of URIs from 1990 to 2019. METHODS: A secondary analysis was performed on the incidence, mortality, and disability-adjusted life years (DALYs) of URIs in different sex and age groups, from 21 geographic regions, 204 countries and territories, between 1990 and 2019, using the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019. Countries and territories were categorized according to Socio-demographic Index (SDI) quintiles. FINDINGS: Globally, the incident cases of URIs reached 17·2 (95% uncertainty interval: 15·4 to 19·3) billion in 2019, which accounted for 42·83% (40·01% to 45·77%) cases from all causes in the GBD 2019 study. The age-standardized incidence rate remained stable from 1990 to 2019, while significant decreases were found in the mortality and DALY rate. The highest age-standardized incidence rates from 1990 to 2019 and the highest age-standardized DALY rates after 2011 were observed in high SDI regions. Among all the age groups, children under five years old suffered from the highest incidence and DALY rates, both of which were decreased with increasing age. Fatal consequences of URIs occurred mostly in the elderly and children under five years old. INTERPRETATION: The present study provided comprehensive estimates of URIs burden for the first time. Our findings, highlighting the substantial incidence and considerable DALYs due to URIs, are expected to attract more attention to URIs and provide future explorations in the prevention and treatment with epidemiological evidence. FUNDING: The study was funded by the National Natural Science Foundation of China (81770057).