OCR, or optical character recognition, is a process whereby a paper document, or fax, is turned back into a near digital representation of the original. This allows data to be extracted from source, and possibly inserted into the EMR as discrete data. Typically though the process of OCR requires the user to select fields to scan, to check and correct the text, and then add it to the correct patient.
Synapse has a new optional utility that automates much of this by automatically OCR any documents of type TIFF or PDF, and associating them with the scanned image. If the image is of sufficiently high quality, it may be possible to take results from the OCR and then inject them into the database. You might want to check them first though to make sure that the accuracy is good enough.
In the first run, it took about 12 hours to automatically OCR several thousand of my own scans. Thereafter it sits waiting for more scans to arrive. Because the processing is done as a web service, there is very little CPU usage on the server.
We may expand this facility to automatically scan patient intake forms and enter patients into Synapse based upon these forms. Note that OCR web services can be purchased for about $250 - $300 per year, and that does not include any integration with Synapse. Our price will be about a third of this 