Motivation: Wuhan pneumonia is an acute infectious disease caused by the 2019 novel coronavirus (COVID-19). It is being treated as a Class A infectious disease though it was classified as Class B according to the Infectious Disease Prevention Act of China. Accurate estimation of the incubation period of the coronavirus is essential to the prevention and control. However, it remains unclear about its exact incubation period though it is believed that symptoms of COVID-19 can appear in as few as 2 days or as long as 14 or even more after exposure. The accurate incubation period calculation requires original chain-of-infection data that may not be fully available in the Wuhan regions. In this study, we aim to accurately calculate the incubation period of COVID-19 by taking advantage of the chain-of-infection data, which is well-documented and epidemiologically informative, outside the Wuhan regions. Methods: We acquired and collected officially reported COVID-19 data from 10 regions in China except for Hubei province. To achieve the accurate calculation of the incubation period, we only involved the officially confirmed cases with a clear history of exposure and time of onset. We excluded those without relevant epidemiological descriptions, working or living in Wuhan for a long time, or hard to determine the possible exposure time. We proposed a Monte Caro simulation approach to estimate the incubation of COVID-19 as well as employed nonparametric ways. We also employed manifold learning and related statistical analysis to decipher the incubation relationships between different age/gender groups. Result: The incubation period of COVID-19 did not follow general incubation distributions such as lognormal, Weibull, and Gamma distributions. We estimated that the mean and median of its incubation were 5.84 and 5.0 days via bootstrap and proposed Monte Carlo simulations. We found that the incubation periods of the groups with age>=40 years and age

/pdf/estimate-the-incubation-period-of-coronavirus-2019-covid-19-36lgzuun89.pdf

Estimate the incubation period of coronavirus 2019 (COVID-19)

Class-aware sample reweighting optimal transport for multi-source domain adaptation

A mixed reality-based navigation method for dental implant navigation method: A pilot study

LDANet: Automatic lung parenchyma segmentation from CT images

Dynamic Individual Selection and Crossover Boosted Forensic-based Investigation Algorithm for Global Optimization and Feature Selection

Accurate iris segmentation and recognition using an end-to-end unified framework based on MADNet and DSANet

Study on a novel method of speaker recognition: Study on a novel method of speaker recognition

In the field of Human-Computer Interaction (HCI), speech emotion recognition technology plays an important role. Facing a small number of speech emotion data, a novel speech emotion recognition method based on feature construction and ensemble learning is proposed in this paper. Firstly, the acoustic features are extracted from the speech signal and combined to form different original feature sets. Secondly, based on Light Gradient Boosting Machine (LightGBM) and Sequential Forward Selection (SFS) method, a novel feature selection method named L-SFS is proposed. And then, the softmax regression model is used to learn automatically the weights of the four single weak learners including Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Extreme Gradient Boosting (XGBoost) and LightGBM. Lastly, based on the learned automatically weights and the weighted average probability voting strategy, an ensemble classification model named Sklex is constructed, which integrates the above four single weak learners. In conclusion, the method reflects the effectiveness of feature construction and the superiority and stability of ensemble learning, and gets good speech emotion recognition accuracy.

Liang-jun Xu

Papers

Accurate iris segmentation and recognition using an end-to-end unified framework based on MADNet and DSANet

Study on a novel method of speaker recognition: Study on a novel method of speaker recognition

A novel speech emotion recognition method based on feature construction and ensemble learning