How it works

How does DocVoice fill in EHR fields automatically?

DocVoice listens to the visit, turns it into a complete, structured note and enters it straight into the fields of your EHR. Everything runs locally, on the clinic's own computers.

Three steps per visit

  1. The physician starts DocVoice with a right-click or a keyboard shortcut.
  2. DocVoice transcribes the conversation with the patient, or a summary the physician dictates, in Hebrew, English or a mix of the two.
  3. The physician right-clicks inside the open chart, and DocVoice finds the open fields and fills each one in with the right part of the conversation.

Any field in the chart

DocVoice isn't limited to a fixed set of fields. A local AI model reads the open fields in the chart, works out what each one is asking for and fills it in from the conversation.

The visit or a dictated summary

DocVoice transcribes the conversation with the patient during the visit, and a summary the physician dictates afterward. Either way, the result is the same: a structured note entered into the chart.

Built for clinical language

The models are built for the way physicians talk: medical terminology, drug names, dosages and common abbreviations, even when the physician switches between Hebrew, English and Latin mid-sentence.

The physician signs off

DocVoice fills in the fields of the open chart but never saves the note itself. The physician sees what went into each field, edits it as usual and saves the note the way they always do.

Giving physicians their time back

A time-and-motion study of office-based physicians found that they spent 49.2% of their day on the EHR and desk work, and only 27% face to face with patients. On top of that, they reported one to two hours of after-hours EHR work every night.

Speaking is also faster than typing: in a Stanford study, voice input was 2.9 times faster than typing on a phone keyboard (153 vs. 52 words per minute in English).

Sources

  1. Sinsky C. et al., Allocation of Physician Time in Ambulatory Practice, Annals of Internal Medicine, 2016
  2. Ruan S. et al. (Stanford), Comparing Speech and Keyboard Text Entry for Short Messages in Two Languages on Touchscreen Phones, 2016