When will your EHR take dictation

Many physicians hate the idea of entering data on a computer. They yearn for a program that can understand their dictation and drop patient data into the correct fields in an EHR. Speech recognition technology is a step in this direction, but it can only create blocks of text or navigate fields using speech-triggered macros. The next step is what is known as "natural language processing," which pulls information from dictated text and distributes it into categories.

Philips, the Netherlands-based electronics giant, demonstrated a form of natural language processing at the recent HIMSS conference in New Orleans. Developed in collaboration with Health Language Inc. (HLI), a company based in Aurora, CO, the prototype software extracts findings, diagnoses, drugs, allergies, and other relevant information from dictation so the data can be slotted into EHR fields. It also recognizes different terms and phrases for the same medical concept and maps them to standard terminologies like SNOMED and ICD.

The technology is still under development, notes Nick van Terheyden, MD, chief medical office and director of business development for Philips Speech Recognition Systems. But he's hopeful that EHR vendors will be interested in working with Philips to integrate natural language processing into their products. "The future is tied to extracting clinical data from everyday speech," declares van Terheyden. Meanwhile, Philips will keep selling its SpeechMagic voice recognition system to hospitals, but will leave the physician office market to Dragon NaturallySpeaking.