#Engauge digitizer segment fill how to#
Among the earliest to do so were Parmar, who described how to estimate the log of the hazard ratio, and its variance, from the survival curves themselves, rather than from numbers and summaries reported in the text. Other authors have shown that in some circumstances, and by making some assumptions, it is possible to extract additional information. Duchateau expressed caution, pointing out that the number of events should not be estimated from the Kaplan-Meier curves for meta-analytic purposes unless virtually no patients are lost to follow-up or censored and there are still many patients at risk in the two groups at the time at which the number of events is to be determined.
Some guidance on data reconstruction can be found in the meta-analysis literature, since the summaries are not always reported in the way meta-analysts would wish and since simplifying assumptions, such as a constant hazard ratio, may be inappropriate. However, our methodology is applicable to any situation where published data are in the form of cumulative incidence curves, or survival curves, of a step function form. Thus, a sequence of time-specific hazard ratios (i.e., a rate ratio ‘curve’) that accommodates this delay is more appropriate than the single-number hazard ratio typically reported by trialists. Our own experiences focus on randomized trials of cancer screening, where the mortality deficits produced by cancer screening are delayed. The researchers differ in their reasons for obtaining such data, and in the number of studies involved. Since it is not always possible to obtain the raw data directly from the authors, one is forced to make do with the information that can be recovered from the articles. Researchers often wish to carry out additional calculations or analyses using the survival data from studies of other authors. Compared with previous approaches, one advantage of ours is that there is no observer variation: there is no need to repeat the digitization process, since the extraction is completely replicable. When it is not possible to obtain the raw data from the authors, reconstruction techniques are a valuable alternative. We found that the PostScript used by Stata discloses considerably more of the data hidden behind survival curves than that generated by other statistical packages. If the original images can be obtained as a PostScript file, the data recovered from it can then be either input into these tools or processed directly. If the raster-based images are available, one can reliably recover much of the original information that seems to be ‘hidden’ beneath published survival curves. We focus on the additional precision, and elimination of observer variation, achieved by using vector-based formats rendered by PostScript, rather than the lower resolution image-based formats that have been analyzed up to now. Using examples, and a formal error analysis, we illustrate the extent to which, with what accuracy and precision, and in what circumstances, this information can be recovered from the various electronic formats in which such curves are published. Instead of using a digitizer to read in the coordinates from a raster image, we propose directly reading in the lines of the PostScript file of a vector image. Several authors have proposed methods/tools for extracting data from such curves using a digitizing software. When it is not possible to obtain the raw data directly, reconstruction techniques provide a valuable alternative. Researchers often wish to carry out additional calculations or analyses using the survival data from one or more studies of other authors.
0 kommentar(er)
