The burning of fossil fuels in industry results in significant carbon emissions,and the heat generated is often not fully utilized.For high-temperature industries,thermophotovoltaics(TPVs)is an effective method for wa...The burning of fossil fuels in industry results in significant carbon emissions,and the heat generated is often not fully utilized.For high-temperature industries,thermophotovoltaics(TPVs)is an effective method for waste heat recovery.This review covers two aspects of high-efficiency TPV systems and industrial waste heat applications.At the system level,representative results of TPV complete the systems,while selective emitters and photovoltaic cells in the last decade are compiled.The key points of components to improve the energy conversion efficiency are further analyzed,and the related micro/nano-fabrication methods are introduced.At the application level,the feasibility of TPV applications in high-temperature industries is shown from the world waste heat utilization situation.The potential of TPV in waste heat recovery and carbon neutrality is illustrated with the steel industry as an example.展开更多
This paper aims to develop and validate a deep learning-based short-term mortality risk prediction model for critically ill patients by using routinely collected data in a large Chinese cohort and explore the explaina...This paper aims to develop and validate a deep learning-based short-term mortality risk prediction model for critically ill patients by using routinely collected data in a large Chinese cohort and explore the explainability of the model decision.A total of 10925 critically ill patients between January 2014 and June 2020 are included in this study.Data routinely collected in the electronic health records(EHRs)system are extracted and used to develop a short-term mortality risk prediction model based on a deep artificial neural network(ANN).The features include demographic characteristics,vital signs,laboratory tests,and the daily dose of intravenous medications.The developed deep learning model(AUROC:0.88,AUPRC:0.63,Brier score:0.108)is superior to the model based on APACHEⅡscores(AUROC:0.78,AURPC:0.52,Brier score:0.124)in the prediction of hospital mortality for critically ill patients.Further attribution analysis based on the integrated gradients method shows that measurements observed at a later time seem to have a more significant influence on mortality,while earlier usage of amiodarone or dexmedetomidine contributed to lower mortality.This well-performing and interpretable model may have practical implications for improving the quality of care for critically ill patients.展开更多
基金supported by the National Natural Science Foundation of China(No.52227813)China Postdoctoral Science Foundation(Nos.2023M740905,2023T160164)+3 种基金National Key ResearchDevelopment Program of China(No.2022YFE0210200)Natural Science Foundation of Heilongjiang Province(No.LH2023E043)the Fundamental Research Funds for the Central Universities(Nos.2022ZFJH04,HIT.OCEF.2021023)。
文摘The burning of fossil fuels in industry results in significant carbon emissions,and the heat generated is often not fully utilized.For high-temperature industries,thermophotovoltaics(TPVs)is an effective method for waste heat recovery.This review covers two aspects of high-efficiency TPV systems and industrial waste heat applications.At the system level,representative results of TPV complete the systems,while selective emitters and photovoltaic cells in the last decade are compiled.The key points of components to improve the energy conversion efficiency are further analyzed,and the related micro/nano-fabrication methods are introduced.At the application level,the feasibility of TPV applications in high-temperature industries is shown from the world waste heat utilization situation.The potential of TPV in waste heat recovery and carbon neutrality is illustrated with the steel industry as an example.
基金Supported by Xiangya Clinical Big Data Construction Project。
文摘This paper aims to develop and validate a deep learning-based short-term mortality risk prediction model for critically ill patients by using routinely collected data in a large Chinese cohort and explore the explainability of the model decision.A total of 10925 critically ill patients between January 2014 and June 2020 are included in this study.Data routinely collected in the electronic health records(EHRs)system are extracted and used to develop a short-term mortality risk prediction model based on a deep artificial neural network(ANN).The features include demographic characteristics,vital signs,laboratory tests,and the daily dose of intravenous medications.The developed deep learning model(AUROC:0.88,AUPRC:0.63,Brier score:0.108)is superior to the model based on APACHEⅡscores(AUROC:0.78,AURPC:0.52,Brier score:0.124)in the prediction of hospital mortality for critically ill patients.Further attribution analysis based on the integrated gradients method shows that measurements observed at a later time seem to have a more significant influence on mortality,while earlier usage of amiodarone or dexmedetomidine contributed to lower mortality.This well-performing and interpretable model may have practical implications for improving the quality of care for critically ill patients.