Extracting data from vector figures in scholarly articles
This addresses a practical problem for researchers who need to reanalyze data from published figures, though it is incremental as it builds on existing vector format capabilities.
The researchers tackled the problem of extracting data from vector figures in scholarly articles, which is typically time-consuming and error-prone when done manually. Their alpha software correctly extracted data from 50% of funnel plots tested (12 out of 24).
It is common for authors to communicate their results in graphical figures, but those data are frequently unavailable for reanalysis. Reconstructing data points from a figure manually requires the author to measure the coordinates either on printed pages using a ruler, or from the display screen using a cursor. This is time-consuming (often hours) and error-prone, and limited by the precision of the display or ruler. What is often not realised is that the data themselves are held in the PDF document to much higher precision (usually 0.0-0.01 pixels), if the figure is stored in vector format. We developed alpha software to automatically reconstruct data from vector figures and tested it on funnel plots in the meta-analysis literature. Our results indicate that reconstructing data from vector based figures is promising, where we correctly extracted data for 12 out of 24 funnel plots with extracted data (50%). However, we observed that vector based figures are relatively sparse (15 out of 136 papers with funnel plots) and strongly insist publishers to provide more vector based data figures in the near future for the benefit of the scholarly community.