BACKGROUND A well-recognized class effect of immune checkpoint inhibitors(ICI)is immune-related adverse events(IrAEs)ranging from low grade toxicities to life-threatening end organ damage requiring permanent discontin...BACKGROUND A well-recognized class effect of immune checkpoint inhibitors(ICI)is immune-related adverse events(IrAEs)ranging from low grade toxicities to life-threatening end organ damage requiring permanent discontinuation of ICI.Deaths are reported in<5%of patients treated with ICI.There are,however,no reliable markers to predict the onset and severity of IrAEs.We tested the association between neutrophil-lymphocyte ratio(NLR)and platelet-lymphocyte ratio(PLR)at baseline with development of clinically significant IrAEs(grade≥2)in hepatocellular carcinoma(HCC)patients treated with ICI.AIM To test the association between NLR and PLR at baseline with development of clinically significant IrAEs(grade≥2)in HCC patients treated with ICI.METHODS Data was extracted from an international database from a consortium of 11 tertiary-care referral centers.NLR=absolute neutrophil count/absolute lymphocyte count(ALC)and PLR=platelet count/ALC.Cutoff of 5 was used for NLR and 300 for PLR based on literature.We also tested the association between RESULTS Data was collected from 361 patients treated between 2016-2020 across the United States(67%),Asia(14%)and Europe(19%).Most patients received Nivolumab(n=255,71%).One hundred sixty-seven(46%)patients developed at least one IrAE,highest grade 1 in 80(48%),grade≥2 in 87(52%)patients.In a univariable regression model PLR>300 was significantly associated with a lower incidence of grade≥2 IrAEs(OR=0.40;P=0.044).Similarly,a trend was observed between NLR>5 and lower incidence of grade≥2 IrAEs(OR=0.58;P=0.097).Multivariate analyses confirmed PLR>300 as an independent predictive marker of grade≥2 IrAEs(OR=0.26;P=0.011),in addition to treatment with programmed cell death ligand 1(PD-1)/cytotoxic T lymphocyte-associated protein-4(OR=2.57;P=0.037)and PD-1/tyrosine kinase inhibitor(OR=3.39;P=0.01)combinations.Antibiotic use was not associated with IrAE incidence(OR=1.02;P=0.954).Patients treated with steroids had a>2-fold higher incidence of grade≥2 IrAEs(OR=2.74;P<0.001),although 74%were prescribed steroids for the treatment of IrAEs.CONCLUSION Given that high baseline NLR and PLR are associated with a decreased incidence of IrAEs,lower baseline NLR and PLR may be predictive biomarkers for the appearance of IrAEs in HCC treated with ICI.This finding is in keeping with several studies in solid tumors that have shown that baseline NLR and PLR appear predictive of IrAEs.展开更多
This article explains how two AI systems have been incorporated into the everyday operations of two Singapore public healthcare nation‐wide screening programs.The first example is embedded within the setting of a nat...This article explains how two AI systems have been incorporated into the everyday operations of two Singapore public healthcare nation‐wide screening programs.The first example is embedded within the setting of a national level population health screening program for diabetes related eye diseases,targeting the rapidly increasing number of adults in the country with diabetes.In the second example,the AI assisted screening is done shortly after a person is admitted to one of the public hospitals to identify which inpatients—especially which elderly patients with complex conditions—have a high risk of being readmitted as an inpatient multiple times in the months following discharge.Ways in which healthcare needs and the clinical operations context influenced the approach to designing or deploying the AI systems are highlighted,illustrating the multiplicity of factors that shape the requirements for successful large‐scale deployments of AI systems that are deeply embedded within clinical workflows.In the first example,the choice was made to use the system in a semi‐automated(vs.fully automated)mode as this was assessed to be more cost‐effective,though still offering substantial productivity improvement.In the second example,machine learning algorithm design and model execution trade-offs were made that prioritized key aspects of patient engagement and inclusion over higher levels of predictive accuracy.The article concludes with several lessons learned related to deploying AI systems within healthcare settings,and also lists several other AI efforts already in deployment and in the pipeline for Singapore's public healthcare system.展开更多
Background Biliary atresia(BA)is a rare fatal liver disease in children,and the aim of this study was to develop a method to diagnose BA early.Methods We determined serum levels of matrix metalloproteinase-7(MMP-7),th...Background Biliary atresia(BA)is a rare fatal liver disease in children,and the aim of this study was to develop a method to diagnose BA early.Methods We determined serum levels of matrix metalloproteinase-7(MMP-7),the results of 13 liver tests,and the levels of 20 bile acids,and integrated computational models were constructed to diagnose BA.Results Our findings demonstrated that MMP-7 expression levels,as well as the results of four liver tests and levels of ten bile acids,were significantly different between 86 BA and 59 non-BA patients(P<0.05).The computational prediction model revealed that MMP-7 levels alone had a higher predictive accuracy[area under the receiver operating characteristic curve(AUC)=0.966,95%confidence interval(CI):0.942,0.989]than liver test results and bile acid levels.The AUC was 0.890(95%CI 0.837,0.943)for liver test results and 0.825(95%CI 0.758,0.892)for bile acid levels.Furthermore,bile levels had a higher contribution to enhancing the predictive accuracy of MMP-7 levels(AUC=0.976,95%CI 0.953,1.000)than liver test results.The AUC was 0.983(95%CI 0.962,1.000)for MMP-7 levels combined with liver test results and bile acid levels.In addition,we found that MMP-7 levels were highly correlated with gamma-glutamyl transferase levels and the liver fibrosis score.Conclusion The innovative integrated models based on a large number of indicators provide a noninvasive and cost-effective approach for accurately diagnosing BA in children.展开更多
Conventional private data publication mechanisms aim to retain as much data utility as possible while ensuring sufficient privacy protection on sensitive data.Such data publication schemes implicitly assume that all d...Conventional private data publication mechanisms aim to retain as much data utility as possible while ensuring sufficient privacy protection on sensitive data.Such data publication schemes implicitly assume that all data analysts and users have the same data access privilege levels.However,it is not applicable for the scenario that data users often have different levels of access to the same data,or different requirements of data utility.The multi-level privacy requirements for different authorization levels pose new challenges for private data publication.Traditional PPDP mechanisms only publish one perturbed and private data copy satisfying some privacy guarantee to provide relatively accurate analysis results.To find a good tradeoffbetween privacy preservation level and data utility itself is a hard problem,let alone achieving multi-level data utility on this basis.In this paper,we address this challenge in proposing a novel framework of data publication with compressive sensing supporting multi-level utility-privacy tradeoffs,which provides differential privacy.Specifically,we resort to compressive sensing(CS)method to project a n-dimensional vector representation of users’data to a lower m-dimensional space,and then add deliberately designed noise to satisfy differential privacy.Then,we selectively obfuscate the measurement vector under compressive sensing by adding linearly encoded noise,and provide different data reconstruction algorithms for users with different authorization levels.Extensive experimental results demonstrate that ML-DPCS yields multi-level of data utility for specific users at different authorization levels.展开更多
文摘BACKGROUND A well-recognized class effect of immune checkpoint inhibitors(ICI)is immune-related adverse events(IrAEs)ranging from low grade toxicities to life-threatening end organ damage requiring permanent discontinuation of ICI.Deaths are reported in<5%of patients treated with ICI.There are,however,no reliable markers to predict the onset and severity of IrAEs.We tested the association between neutrophil-lymphocyte ratio(NLR)and platelet-lymphocyte ratio(PLR)at baseline with development of clinically significant IrAEs(grade≥2)in hepatocellular carcinoma(HCC)patients treated with ICI.AIM To test the association between NLR and PLR at baseline with development of clinically significant IrAEs(grade≥2)in HCC patients treated with ICI.METHODS Data was extracted from an international database from a consortium of 11 tertiary-care referral centers.NLR=absolute neutrophil count/absolute lymphocyte count(ALC)and PLR=platelet count/ALC.Cutoff of 5 was used for NLR and 300 for PLR based on literature.We also tested the association between RESULTS Data was collected from 361 patients treated between 2016-2020 across the United States(67%),Asia(14%)and Europe(19%).Most patients received Nivolumab(n=255,71%).One hundred sixty-seven(46%)patients developed at least one IrAE,highest grade 1 in 80(48%),grade≥2 in 87(52%)patients.In a univariable regression model PLR>300 was significantly associated with a lower incidence of grade≥2 IrAEs(OR=0.40;P=0.044).Similarly,a trend was observed between NLR>5 and lower incidence of grade≥2 IrAEs(OR=0.58;P=0.097).Multivariate analyses confirmed PLR>300 as an independent predictive marker of grade≥2 IrAEs(OR=0.26;P=0.011),in addition to treatment with programmed cell death ligand 1(PD-1)/cytotoxic T lymphocyte-associated protein-4(OR=2.57;P=0.037)and PD-1/tyrosine kinase inhibitor(OR=3.39;P=0.01)combinations.Antibiotic use was not associated with IrAE incidence(OR=1.02;P=0.954).Patients treated with steroids had a>2-fold higher incidence of grade≥2 IrAEs(OR=2.74;P<0.001),although 74%were prescribed steroids for the treatment of IrAEs.CONCLUSION Given that high baseline NLR and PLR are associated with a decreased incidence of IrAEs,lower baseline NLR and PLR may be predictive biomarkers for the appearance of IrAEs in HCC treated with ICI.This finding is in keeping with several studies in solid tumors that have shown that baseline NLR and PLR appear predictive of IrAEs.
文摘This article explains how two AI systems have been incorporated into the everyday operations of two Singapore public healthcare nation‐wide screening programs.The first example is embedded within the setting of a national level population health screening program for diabetes related eye diseases,targeting the rapidly increasing number of adults in the country with diabetes.In the second example,the AI assisted screening is done shortly after a person is admitted to one of the public hospitals to identify which inpatients—especially which elderly patients with complex conditions—have a high risk of being readmitted as an inpatient multiple times in the months following discharge.Ways in which healthcare needs and the clinical operations context influenced the approach to designing or deploying the AI systems are highlighted,illustrating the multiplicity of factors that shape the requirements for successful large‐scale deployments of AI systems that are deeply embedded within clinical workflows.In the first example,the choice was made to use the system in a semi‐automated(vs.fully automated)mode as this was assessed to be more cost‐effective,though still offering substantial productivity improvement.In the second example,machine learning algorithm design and model execution trade-offs were made that prioritized key aspects of patient engagement and inclusion over higher levels of predictive accuracy.The article concludes with several lessons learned related to deploying AI systems within healthcare settings,and also lists several other AI efforts already in deployment and in the pipeline for Singapore's public healthcare system.
基金supported by a grant from the Key Program of the Independent Design Project of National Clinical Research Center for Child Health(Grant no.S20C0004)Shanghai Pujiang Program(21PJ1423100)Ministry of Industry and Information Technology Artificial Intelligence Medical Devices Innovation Program.
文摘Background Biliary atresia(BA)is a rare fatal liver disease in children,and the aim of this study was to develop a method to diagnose BA early.Methods We determined serum levels of matrix metalloproteinase-7(MMP-7),the results of 13 liver tests,and the levels of 20 bile acids,and integrated computational models were constructed to diagnose BA.Results Our findings demonstrated that MMP-7 expression levels,as well as the results of four liver tests and levels of ten bile acids,were significantly different between 86 BA and 59 non-BA patients(P<0.05).The computational prediction model revealed that MMP-7 levels alone had a higher predictive accuracy[area under the receiver operating characteristic curve(AUC)=0.966,95%confidence interval(CI):0.942,0.989]than liver test results and bile acid levels.The AUC was 0.890(95%CI 0.837,0.943)for liver test results and 0.825(95%CI 0.758,0.892)for bile acid levels.Furthermore,bile levels had a higher contribution to enhancing the predictive accuracy of MMP-7 levels(AUC=0.976,95%CI 0.953,1.000)than liver test results.The AUC was 0.983(95%CI 0.962,1.000)for MMP-7 levels combined with liver test results and bile acid levels.In addition,we found that MMP-7 levels were highly correlated with gamma-glutamyl transferase levels and the liver fibrosis score.Conclusion The innovative integrated models based on a large number of indicators provide a noninvasive and cost-effective approach for accurately diagnosing BA in children.
基金supported by the US National Science Foun-dation under grant CNS-1704397.
文摘Conventional private data publication mechanisms aim to retain as much data utility as possible while ensuring sufficient privacy protection on sensitive data.Such data publication schemes implicitly assume that all data analysts and users have the same data access privilege levels.However,it is not applicable for the scenario that data users often have different levels of access to the same data,or different requirements of data utility.The multi-level privacy requirements for different authorization levels pose new challenges for private data publication.Traditional PPDP mechanisms only publish one perturbed and private data copy satisfying some privacy guarantee to provide relatively accurate analysis results.To find a good tradeoffbetween privacy preservation level and data utility itself is a hard problem,let alone achieving multi-level data utility on this basis.In this paper,we address this challenge in proposing a novel framework of data publication with compressive sensing supporting multi-level utility-privacy tradeoffs,which provides differential privacy.Specifically,we resort to compressive sensing(CS)method to project a n-dimensional vector representation of users’data to a lower m-dimensional space,and then add deliberately designed noise to satisfy differential privacy.Then,we selectively obfuscate the measurement vector under compressive sensing by adding linearly encoded noise,and provide different data reconstruction algorithms for users with different authorization levels.Extensive experimental results demonstrate that ML-DPCS yields multi-level of data utility for specific users at different authorization levels.
