An Audio Scribe Only Hears Half the Visit.

Medicine is visual. A rash, a wound, a lab report, a monitor: none of it is in the transcript. DocGlasses sees what the provider sees, so the note is complete.

Vision Changes What a Scribe Can Catch.

An independent study from Flinders University, published in npj Digital Medicine (2026), compared vision-enabled AI scribes on smart glasses with audio-only scribes across 110 simulated clinical consultations.

Omission errors across 110 consultations
0 errorsAudio-only AI
0 errorsVision AI
0×

fewer omission errors with vision-enabled AI.

0%
Vision AI accuracy
0%
Audio-only accuracy

“Vision-enabled AI scribes are transformatively more accurate at capturing what matters most in a clinical setting.”

npj Digital Medicine, 2026

From DocGlasses pilots

0%

clinical capture accuracy in pilots

~0 min

saved per visit in pilots

The study validated the vision-enabled architecture DocGlasses is built on. Pilot figures are DocGlasses' own and reported separately.

Audio Scribe or DocGlasses.

What gets capturedAudio scribeDocGlasses
What the patient saysYesYes
Skin, wounds and visible findingsNoYes
Handheld reports, labels and devicesNoYes
Photos attached to the noteNoYes
Suggestions in the field of viewNoYes

Three Inputs. One Engine. Zero Typing.

The DocGlasses AI engine fuses medical records, patient identity and the live visit, then returns clinical insight to the lens and a finished record to the EHR.

InputMedical reportsX-rays, lab results, devices
InputPatient recordIdentity, history and medications
InputThe visitReal-time voice and vision
DocGlasses Vision AIEdge and cloud
OutputEHR updateStructured note with ICD codes
OutputClinical insightLikely causes and next steps, in-lens

Try Them On.

See DocGlasses on real glasses, in your care setting. It takes about a minute to book.