关 键 词：

脉搏 设计 CAD 程序

资源描述：

血氧脉搏仪设计含4张CAD图及程序,脉搏,设计,CAD,程序

内容简介：

ORIGINAL ARTICLEWavelet analysis of pulse oximeter waveform permitsidentification of unwell childrenP Leonard, T F Beattie, P S Addison, J N Watson. .See end of article forauthors affiliations.Correspondence to:Dr P A Leonard,Department of Accidentand Emergency Medicine,The Royal Hospital for SickChildren, Edinburgh,Scotland; paul_leonardblueyonder.co.ukAccepted for publication5 May 2003.Emerg Med J 2004;21:5960Background: Children who are unwell often display signs of circulatory compromise. It has been observedthat pronounced changes occur in the appearance of the photoplethysmogram (pulse oximeter tracing) inthese children. The aim of the study was to discover if wavelet transforms can identify more subtle changesin the photoplethysmogram of children who are unwell.Methods: Photoplethysmograms were obtained from children attending a paediatric accident andemergency department with clinical features suggestive of significant bacterial illness or circulatorycompromise. Photoplethysmograms were also obtained from a control group of well children. Wavelettransforms were applied to the traces in an attempt to separate the two groups.Results: 20 traces were obtained from unwell children and 12 from controls. Analysis of the entropy of thewavelet transform of the photoplethysmogram allows the differentiation of unwell children from controls(p=0.00002).Conclusions: Wavelet transform of the photoplethysmogram offers the possibility of a rapid non-invasivemethod of screening children for significant illness.General practitioners and accident and emergencystaff are often presented with children who arefeverish. Most will have benign, self limiting viralillnesses that need no treatment other than parental adviceand antipyretic drugs. However, a small number will haveseriousbacterialillnesses(forexample,meningococcaldisease).Itisrecognisedthatitcanbedifficulttodifferentiate these two groups of children and variousguidelines using combinations of clinical signs and bloodtests have been proposed.15These guidelines are oftendifficult to institute in the community where access tolaboratory tests is limited, and even in the hospital setting aretime consuming and cause distress because of the need forphlebotomy.Children with significant bacterial illnesses often displaysigns of circulatory compromise in the form of tachycardia,prolonged capillary refill time and increased core-peripheraltemperature difference. We have noted that the macroscopicvisual appearance of the pulse oximeter tracing (photo-plethysmogram) changes in these children, from the wellrecognised two humped waveform to a flatter one humpedappearance, or a completely flat trace if pulsatile capillaryflow is absent at the site of monitoring.We wondered if wavelet transform based analysis techni-ques6could be used to identify more subtle changes in thephotoplethysmogram of children who are unwell. This wouldpotentially provide a rapid non-invasive method of screeningchildren for significant illness.METHODSChildren attending the accident and emergency departmentof a busy paediatric hospital were eligible for inclusion in thestudyiftheydisplayedclinicalfeaturessuggestiveofsignificant bacterial illness or circulatory compromise (box).Prior ethical approval had been obtained from the localregionalethicscommittee,andinformedconsentwasobtained from the parents of all enrolled children, and whereapplicable from the child themselves.All children had pulse oximetry performed simultaneouslyat two sites (lobe of ear and either a finger or a toe astolerated by the child) using paired Nelcor 100 pulseoximeters,whichwereconnectedtoastandardDellPentiumPCviaaComputerBoardsPCI-DAS1602/16analoguetodigitalconverterboard.Thisenabledtheplethysmogram to be captured and stored for analysis.Monitoring took place over a period of an hour or untilthe child left the department. Clinical care for the childwas not interrupted and continued as per departmentalprotocol.A control group of well children was recruited from thesiblings of patients attending the department with minortrauma. These children were monitored for a period of fiveminutes using the same equipment in the same environmentas the study group. This shorter monitoring period for thecontrol group was chosen to enable capture of an adequatelength of plethysmogram for analysis without causing unduedistress to the child.All traces were analysed using wavelet techniques, first toremove the effects of noise and drift, and then to determinethe wavelet based features that permit separation of thecontrol group from the unwell children. Maximal separationwas achieved when the wavelet power across the 1 Hzfrequency band was plotted against the entropy at the 7 Hzband. Entropy is a measure of data spread or clumpiness,with the lowest entropy having all the signal in one place andthe highest having an even spread. Statistical analysis wasperformed using a Wilcoxon rank sum test.RESULTSPhotoplethysmograms were obtained from 20 children ful-filling the study entry criteria. Twelve control traces wereobtained.Figure 1 shows that it is possible to differentiate the controlgroup from the unwell children using wavelet analysis. Theprobability that the two samples come from populations thatare identical is p=0.00002.There was some suggestion that these features can furtherstratify the children into groups based on illness sever-ity, although numbers are too small to reach statisticalsignificance.59DISCUSSIONPulse oximeters use the differential absorption of red andinfraredlightbyhaemoglobininitsoxygenatedanddeoxygenated forms to calculate the ratio of these twomolecules in pulsatile capillary flow. The photoplethysmo-gram is a plot of the absorption of light at one of thesewavelengths (usually the infrared) against time. In the pastthe photoplethysmogram has traditionally been thought to beof no value except to determine if a pulse oximeter is pickingup a good signal and thus giving an accurate reading.7However, recent work has shown that other cardiorespiratoryvariables influence the waveform.8 9It is currently unclearwhich variable, or combination of variables gives rise to ourfindings and further work is required to investigate this.The traditional method of analysing waveforms is Fourieranalysis. However, this technique produces only globallyaveragedspectral-onlyinformation,leadingtolocationspecific features in the signal being lost. The completeanalysis of a signal requires the deduction of both thefrequency make up and temporal location of the signalcomponents. This limitation can be partly overcome byintroducing a sliding time window, which localises theanalysis in time (short time Fourier transform (STFT) andprovides a degree of temporal resolution by highlightingchanges in spectral response with respect to time. However,this method is always a compromise between temporal andfrequency resolution, which is set by the pre-defined fixedwindow width. Wavelet transforms differ from the STFT asthey permit arbitrarily high localisation in time of highfrequency signal features. They do this by having a variablewindow width that is related to the scale of observation. Thisflexibility allows for the isolation of the high frequencyfeatures obscured by STFT analysis and make them a usefultool for the analysis of biological signals as demonstrated inthis and other recent studies.1012In conclusion, our study shows that wavelet transformscan be applied to the photoplethysmogram and permit theidentification of features that differentiate control childrenfrom children with clinical markers of significant illness.Further work with larger numbers of children needs to beundertaken to determine the physiological basis and clinicalapplication of this technique. However, the potential exists toimprove the identification of children with serious diseaseusing a simple, real time, non-invasive monitoring systemthat would also provide standard pulse oximetry.Authors affiliations.P Leonard, T F Beattie, Department of Accident and EmergencyMedicine, The Royal Hospital for Sick Children, Edinburgh, ScotlandP S Addison, Faculty of Engineering and Computing, Napier University,Merchiston Campus, Edinburgh, ScotlandJ N Watson, Cardiodigital Ltd, Edinburgh, ScotlandREFERENCES1 Baraff LJ, Bass JW, Fleisher GR, et al. Practice guideline for the managementof infants and children 0 to 36 months of age with fever without source.Pediatrics 1993;92:112.2 Baraff LJ. Management of fever without source in infants and children. AnnEmerg Med 2000;36:60214.3 Ghosh S, Mittal M, Jaganathan G. Early diagnosis of neonatal sepsis using ahematological scoring system. Indian J Med Sci 2001;55:495500.4 Fleisher GR. Management of children with occult bacteremia who are treatedin the emergency department. Reviews of Infectious Diseases1991;13:S1569.5 Lieu TA, Schwartz JS, Jaffe DM, et al. Strategies for diagnosis and treatment ofchildren at risk for occult bacteremia: clinical effectiveness and cost-effectiveness. J Pediatr 1991;118:219.6 Addison PS. The illustrated wavelet transform handbook. Bristol: Institute ofPhysics Publishing, 2002.7 Moyle JT. Pulse oximetry. London: BMJ Publishing Group, 1994.8 Vegfors M, Lindberg L, Pettersson H, et al. Presentation and evaluation of anew optical sensor for respiratory rate monitoring. Int J Clin Monit Comput1994;11:1516.9 Middleton PM, Retter AJ, Henry JA. Pulse oximeter waveforms in healthyhumans. Abstract. J Accid Emerg Med