Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data

Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data

ID:40081368

大小:610.67 KB

页数:8页

时间:2019-07-20

Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data _第1页
Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data _第2页
Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data _第3页
Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data _第4页
Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data _第5页
资源描述:

《Important 2010_6 An Online Self-Tunable Method to Denoise CGM Sensor Data 》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、634IEEETRANSACTIONSONBIOMEDICALENGINEERING,VOL.57,NO.3,MARCH2010AnOnlineSelf-TunableMethodtoDenoiseCGMSensorDataAndreaFacchinetti,GiovanniSparacino,andClaudioCobelli∗,Fellow,IEEEAbstractContinuousglucosemonitoring(CGM)devicescanbeveryusefulindiabetesmanagement.Unfor

2、tunately,theiruseinonlineapplications,e.g.,forhypo/hyperalertgeneration,ismadedifficultbyrandomnoisemeasurement.Remarkably,theSNRofCGMdatavarieswiththesensorandwiththeindividual.Asacon-sequence,approachesinwhichfilterparametersarenotallowedtoadapttothecurrentSNRarelik

3、elytobesuboptimal.Inthispa-per,wepresentanewonlinemethodologytoreducenoiseinCGMsignalsbyaKalmanfilter(KF),whoseunknownparametersareadjustedinagivenindividualbyastochasticallybasedsmoothingcriterionexploitingdataofaburn-ininterval.TheperformanceofthenewKFapproachisqua

4、ntitativelyassessedonMonteCarlosimulationsand24realCGMdatasets.Ourresultsarecomparedwiththoseobtainedbyamoving-average(MA)filteringapproachwithfixedparameterscurrentlyinuseinlikelyallcommercialCGMdevices.ResultsshowthatthenewKFapproachperformsmuchbetterthanMA.Forinsta

5、nce,onrealdata,forcomparablesignaldenoising,thedelayintroducedbyKFisabout35%lessthanthatobtainedbyMA.IndexTermsAlert,biomedicalsignalprocessing,diabetes,Kalmanfilter(KF),signaldenoising.I.INTRODUCTIONLUCOSEisthemostimportantfuelforhumanbeingsandGitslevelinthebloodist

6、ightlycontrolledbyinsulinbyaFig.1.TworepresentativeCGMtime-seriesmeasuredfor24h,at3-minsamplingrate,intwodiabeticsubjectsbyGlucoday.(Top)Low-SNRcondition.negativefeedbacksystem.Diabetesisachronicdiseasethataf-(Bottom)High-SNRcondition.fectsabout250millionpeopleinthe

7、world.Indiabeticpatients,thebodydoesnotsecreteinsulin(Type1diabetes)orderange-Inthepastfewyears,theachievementofamoreaccuratementsinbothinsulinsecretionandaction(Type2diabetes)controlseemspossibleduetothedevelopmentofcontinuousoccur.Diabetestherapyismainlybasedonins

8、ulin,diet,drugglucosemonitoring(CGM)devices[3][5].Thesedevicesal-administration,andphysicalexercise,tunedaccordingtoself-lowtomeasuregluco

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。