Idea 3: Text to EHR

Hospital information systems(HIS) act as information repositories in hospitals. However most of this information is in the form of unstructured text. This text is generated by transcription of dictated reports or feeds from other information systems such as a Radiology information system(RIS).
Currently when a patient requests his/her medical record, the hospital needs to sift through his/her record in the HIS and print out the reports for the patient in the entirety, time spent in this activity translates to a sunk cost for the hospital. If the patient is interested in maintaining a concise health record online, he/she needs to understand the report, extract useful information and enter it manually into the health record. People who want to start maintaining an electronic health record are turned off by this.
The idea is to come up with a text parser that parses unstructured reports and extracts relevant information and updates the EHR. Some examples of relevant information would be Medications, Allergies reported, Height, Weight, Blood pressure and other measurements, Attending physician's name.
The parser would reside as a service that polls the HIS for new text reports.