The sudden deterioration of patients with novel coronavirus disease 2019 (COVID-19) into critical illness is of major concern. It is imperative to identify these patients early. We show that a deep learning-based survival model can predict the risk of COVID-19 patients developing critical illness based on clinical characteristics at admission. We develop this model using a cohort of 1590 patients from 575 medical centers, with internal validation performance of concordance index 0.894 We further validate the model on three separate cohorts from Wuhan, Hubei and Guangdong provinces consisting of 1393 patients with concordance indexes of 0.890, 0.852 and 0.967 respectively. This model is used to create an online calculation tool designed for patient triage at admission to identify patients at risk of severe illness, ensuring that patients at greatest risk of severe illness receive appropriate care as early as possible and allow for effective allocation of health resources.
Recording runtime status via logs is common for almost every computer system, and detecting anomalies in logs is crucial for timely identifying malfunctions of systems. However, manually detecting anomalies for logs is time-consuming, error-prone, and infeasible. Existing automatic log anomaly detection approaches, using indexes rather than semantics of log templates, tend to cause false alarms. In this work, we propose LogAnomaly, a framework to model unstructured a log stream as a natural language sequence. Empowered by template2vec, a novel, simple yet effective method to extract the semantic information hidden in log templates, LogAnomaly can detect both sequential and quantitive log anomalies simultaneously, which were not done by any previous work. Moreover, LogAnomaly can avoid the false alarms caused by the newly appearing log templates between periodic model retrainings. Our evaluation on two public production log datasets show that LogAnomaly outperforms existing log-based anomaly detection methods.
BackgroundHuman parainfluenza viruses (HPIVs) are important causes of upper respiratory tract illness (URTI) and lower respiratory tract illness (LRTI). To analyse epidemiologic and clinical characteristics of the four types of human parainfluenza viruses (HPIVs), patients with acute respiratory tract illness (ARTI) were studied in Guangzhou, southern China.MethodsThroat swabs (n=4755) were collected and tested from children and adults with ARTI over a 26-month period, and 4447 of 4755 (93.5%) patients’ clinical presentations were recorded for further analysis.ResultsOf 4755 patients tested, 178 (3.7%) were positive for HPIV. Ninety-nine (2.1%) samples were positive for HPIV-3, 58 (1.2%) for HPIV-1, 19 (0.4%) for HPIV-2 and 8 (0.2%) for HPIV-4. 160/178 (88.9%) HPIV-positive samples were from paediatric patients younger than 5 years old, but no infant under one month of age was HPIV positive. Seasonal peaks of HPIV-3 and HPIV-1 occurred as autumn turned to winter and summer turned to autumn. HPIV-2 and HPIV-4 were detected less frequently, and their frequency of isolation increased when the frequency of HPIV-3 and HPIV-1 declined. HPIV infection led to a wide spectrum of symptoms, and more “hoarseness” (p=0.015), “abnormal pulmonary breathing sound” (p<0.001), “dyspnoea” (p<0.001), “pneumonia” (p=0.01), and “diarrhoea” (p<0.001) presented in HPIV-positive patients than HPIV-negative patients. 10/10 (100%) HPIV-positive adult patients (≥14 years old) presented with systemic influenza-like symptoms, while 90/164 (54.9%) HPIV-positive paediatric patients (<14 years old) presented with these symptoms (p=0.005). The only significant difference in clinical presentation between HPIV types was “Expectoration” (p<0.001). Co-infections were common, with 33.3%–63.2% of samples positive for the four HPIV types also testing positive for other respiratory pathogens. However, no significant differences were seen in clinical presentation between patients solely infected with HPIV and patients co-infected with HPIV and other respiratory pathogens.ConclusionsHPIV infection led to a wide spectrum of symptoms, and similar clinical manifestations were found in the patients with four different types of HPIVs. The study suggested pathogenic activity of HPIV in gastrointestinal illness. The clinical presentation of HPIV infection may differ by patient age.
We describe how to convert the heuristic search algorithm A* into an anytime algorithm that finds a sequence of improved solutions and eventually converges to an optimal solution. The approach we adopt uses weighted heuristic search to find an approximate solution quickly, and then continues the weighted search to find improved solutions as well as to improve a bound on the suboptimality of the current solution. When the time available to solve a search problem is limited or uncertain, this creates an anytime heuristic search algorithm that allows a flexible tradeoff between search time and solution quality. We analyze the properties of the resulting Anytime A* algorithm, and consider its performance in three domains; sliding-tile puzzles, STRIPS planning, and multiple sequence alignment. To illustrate the generality of this approach, we also describe how to transform the memoryefficient search algorithm Recursive Best-First Search (RBFS) into an anytime algorithm.
To harness modern multicore processors, it is imperative to develop parallel versions of fundamental algorithms. In this paper, we compare different approaches to parallel best-first search in a shared-memory setting. We present a new method, PBNF, that uses abstraction to partition the state space and to detect duplicate states without requiring frequent locking. PBNF allows speculative expansions when necessary to keep threads busy. We identify and fix potential livelock conditions in our approach, proving its correctness using temporal logic. Our approach is general, allowing it to extend easily to suboptimal and anytime heuristic search. In an empirical comparison on STRIPS planning, grid pathfinding, and sliding tile puzzle problems using 8-core machines, we show that A*, weighted A* and Anytime weighted A* implemented using PBNF yield faster search than improved versions of previous parallel search proposals.
Acute Respiratory Infections (ARI) are some of the most common human diseases worldwide. However, they have a complex and diverse etiology, and the characteristics of the pathogens involved in respiratory infections in developing countries are not well understood. In this work, we analyzed the characteristics of 17 common respiratory pathogens in children (≤14 years old) with ARI in Guangzhou, southern China over a 3-year period using real-time polymerase chain reaction. Pathogens were identified in 2361/4242 (55.7%) patients, and the positivity rate varied seasonally. Ten of the 17 pathogens investigated showed positivity rates of more than 5%. The most frequently detected pathogens were respiratory syncytial virus (768/2361, 32.5%), influenza A virus (428/2361, 18.1%), enterovirus (138/2361, 13.3%), Mycoplasma pneumoniae (267/2361, 11.3%) and adenovirus (213/2361, 9.0%). Co-pathogens were common and found in 503 of 2361 (21.3%) positive samples. When ranked according to frequency of occurrence, the pattern of co-pathogens was similar to that of the primary pathogens, with the exception of human bocavirus, human coronavirus and human metapneumovirus. Significant differences were found in age prevalence in 10 of the 17 pathogens (p≤0.009): four basic patterns were observed, A: detection rates increased with age, B: detection rates declined with age, C: the detection rate showed distinct peaks or D: numbers of patients were too low to detect a trend or showed no significant difference among age groups (p>0.05). These data will be useful for planning vaccine research and control strategies and for studies predicting pathogen prevalence.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.