The defibrillation success rate versus energy relationship: part ii--estimation with the 'bootstrap'.
Seventy or so defibrillation trials were typically attempted to determine the relationship between defibrillation success rate and energy (DSRE). Clinically, it may be desirable to estimate the DSRE relationship with fewer trials. We used the statistical resampling technique called the 'bootstrap' t...
| Publicado en: | Pacing & Clinical Electrophysiology Vol. 13; no. 4; pp. 425 - 432 |
|---|---|
| Autores principales: | , , |
| Formato: | research Journal Article |
| Publicado: |
Wiley-Blackwell
Apr1990
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=106086717&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 106086717 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 01478389 4F8 jtl: Pacing & Clinical Electrophysiology issn: 01478389 maglogo: Y pubinfo: dt: Apr1990 vid: 13 iid: 4 pid: 480 pub: Wiley-Blackwell place: Malden, Massachusetts artinfo: ui: 106086717 106086717 2009412356 10.1111/j.1540-8159.1990.tb02057.x NLM1692126 106086717 ppf: 425 ppct: 7 formats: fmt: @attributes: type: P tig: atl: The defibrillation success rate versus energy relationship: part ii--estimation with the 'bootstrap'. aug: au: Gliner BE Murakawa Y Thakor NV sug: subj: Cardioversion Statistics and Numerical Data Animal Studies Dogs Models, Statistical Probability Random Assignment Regression ab: Seventy or so defibrillation trials were typically attempted to determine the relationship between defibrillation success rate and energy (DSRE). Clinically, it may be desirable to estimate the DSRE relationship with fewer trials. We used the statistical resampling technique called the 'bootstrap' to determine the number of defibrillation trials necessary for an accurate estimation of the DSRE relationship. The bootstrap technique assumes that the observed database is the maximum likelihood sample of the estimated population. The observed database is repeatedly resampled to produce a large bootstrap database and the bootstrap best estimate of a statistic is determined. DSRE data were obtained from ten dogs (20.5 ± 1.5 kg). We bootstrapped our experimental DSRE data by two methods: (1) randomly choosing with replacement a specified number of defibrillation trials per energy; and (2) randomly choosing with replacement a specified number of defibrillation trials per bootstrap replication. For both bootstrap techniques, 100 replications were made. We performed a linear regression analysis on the bootstrap success rates and the observed success rates determined from 71.0 ± 6.8 defibrillation attempts from each of the ten dogs. We concluded that 28 defibrillation trials are necessary to estimate the observed DSRE relationship with a correlation coefficient of 0.95. pubtype: Academic Journal doctype: research Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|