This guide walks you through the complete Doctora encounter workflow, from selecting a patient to sending finished documentation to your EHR. If you can talk to a patient, you can use Doctora.
1. Starting an Encounter
There are two ways to begin:
From the schedule. Your daily schedule appears on the dashboard. Tap any patient tile to select them. Doctora pulls in patient demographics, encounter type, and any prior visit context automatically. The tile expands into the encounter workspace.
From a blank encounter. If a patient is not on the schedule (walk-in, added late, etc.), click the microphone button at the top of the dashboard to start a new encounter. Doctora creates a fresh encounter record and you can begin recording immediately.
Choosing a Template
Before you record, check the template shown below the patient name. The template controls which chart sections Doctora will generate---for example, a "Comprehensive Eye Exam" template includes anterior segment, posterior segment, and refraction fields, while a "Contact Lens Follow-Up" template focuses on lens evaluation and fit assessment.
To change the template, click the template name and select a different one from the dropdown. This creates a template override for this specific encounter without changing your default. If your practice has custom templates configured in settings, they will appear here alongside the built-in options.
2. Recording
Hit the large Record button to start. The first time, your browser will ask for microphone permission---grant it, and Doctora remembers the choice for future visits.
Once recording starts you will see:
- A timer showing elapsed recording time
- A live transcript that scrolls as you and the patient speak
- A pause button if you need to step away
Speaking Naturally
There is no special vocabulary or format to follow. Talk to your patient the way you normally would. Doctora's AI is trained on optometric encounters and understands clinical terminology, abbreviations, and conversational context.
What the AI listens for:
- Chief complaint and history of present illness
- Review of systems responses
- Medications, allergies, family and social history
- Examination findings you dictate (slit lamp, fundus, etc.)
- Visual acuity, refraction, and IOP readings
- Assessment, diagnosis, and plan
Speaker identification. Doctora uses real-time diarization to distinguish between you and the patient. In the transcript, each speaker is labeled (typically "Speaker 1" and "Speaker 2"). This helps the AI separate patient-reported symptoms from your clinical observations.
3. During the Exam
While recording, the transcript updates in real time. You do not need to watch it---it is there for reference if you want to confirm something was captured.
Tips for best results:
- Speak at a normal pace. You do not need to slow down or enunciate differently.
- State findings clearly: "Slit lamp shows 2+ nuclear sclerosis OD, trace cortical OS" gives the AI exactly what it needs.
- If you mention a value, say it: "IOP 16 and 18" is better than pointing at a screen.
- Patient conversation is valuable---when a patient describes symptoms, that becomes HPI and ROS data automatically.
- You can pause and resume the recording if you leave the room for testing.
What about sensitive comments? The AI extracts clinical data, not every word. Side conversations, personal remarks, or off-topic chat will not appear in the chart output. The structured extraction focuses exclusively on medically relevant content.
4. Stopping the Recording
When the exam is done, tap Finish Exam. Two things happen:
- Audio finalization. Doctora uploads any remaining audio chunks and assembles the full recording. This typically takes a few seconds.
- AI processing. The transcript is sent through the extraction pipeline. Doctora's AI reads the full conversation and generates structured chart data---not a simple copy of the transcript, but organized clinical documentation with proper field mapping.
You will see a processing indicator while the AI works. This usually takes 10--20 seconds depending on exam length.
What Happens Under the Hood
The AI extraction worker analyzes your transcript against your template's field structure. It produces structured data for each section: patient info, chief complaint, HPI, examination findings, assessment, diagnosis codes, and plan. A post-processing step then normalizes values, validates ICD-10 codes, and applies any custom instructions you have configured (see Custom Instructions for details).
5. Reviewing the AI Output
After processing completes, the Chart tab appears with your structured data organized into collapsible sections. These sections are driven by your template---you will only see fields relevant to the encounter type.
The chart editor is organized into logical groups:
- Patient Information -- demographics, chief complaint, reason for visit
- History -- HPI, past medical/family/social history, medications, allergies, review of systems
- Examination -- visual acuity, refraction, pupils, confrontation fields, extraocular motility, IOP, slit lamp (anterior segment), fundus (posterior segment)
- Assessment & Plan -- clinical assessment, ICD-10 diagnoses, procedures, treatment plan, follow-up
- Contact Lenses -- if applicable, lens parameters, fit evaluation, ordering details
Scroll through each section to review. Fields that the AI populated will show their extracted values. Fields the AI could not determine from the conversation will be empty---this is normal for sections you did not discuss.
6. Editing
Every field is editable inline. Click any value to modify it. Common editing scenarios:
- Correcting a value. Click the field, type the correct value, and move on. Changes save automatically.
- Adding missing data. If the AI missed something (perhaps you said "IOP 14 and 16" but only one eye populated), click the empty field and type the value.
- Removing incorrect data. Clear any field the AI hallucinated or misinterpreted.
- Structured fields. Dropdowns for fields like anterior segment findings let you select from predefined options (e.g., "clear," "trace," "1+," "2+") to match your EHR's expected values.
Adding Sections
If your template does not include a section you need for this particular encounter, use the Add Section button at the bottom of the editor. This adds the section to the current encounter only, without modifying your template.
Re-processing
If you need the AI to take another pass---for example, after adding a second recording---click Generate Chart again. The AI will reprocess all recordings for the encounter and regenerate the structured data. You can also click Add Recording to capture additional audio without losing your existing chart data.
7. Sending to Your EHR
Once you are satisfied with the chart, click Send to EHR in the bottom action bar. This is what "one-click EHR entry" means: Doctora takes every field in your chart and writes it directly into the corresponding fields in your EHR system.
How it works:
- Doctora maps each chart field to the matching field in your EHR (RevolutionEHR, Eyefinity, CrystalPM, etc.).
- The extension or desktop agent opens the encounter in your EHR and populates fields automatically.
- The process handles navigation between EHR screens, dropdown selections, and text entry.
Requirements for Send to EHR:
- Your EHR must be connected (configured in practice settings).
- The Doctora Chrome extension must be installed and active (for browser-based EHRs) or the desktop agent must be running (for on-premise EHRs like CrystalPM and OfficeMate).
- The encounter must be linked to an EHR appointment. Encounters created from synced schedule appointments are linked automatically. Manual encounters may need linking first.
The Send to EHR button will show a tooltip explaining why it is disabled if any requirement is not met.
8. After Sync
Once the EHR write completes, Doctora updates the encounter status. You should:
- Verify in your EHR. Open the patient encounter in your EHR and spot-check a few fields. The data should match what you reviewed in the Doctora chart editor.
- Handle any sync issues. If a field did not populate correctly, you can edit it directly in your EHR. Common causes: the EHR field was locked, a dropdown value did not match exactly, or the encounter was not in an editable state.
- Re-send if needed. If you make changes in Doctora after the initial sync, the Send to EHR button becomes available again so you can push updated data.
The encounter tile on your schedule will update to reflect the completed status, giving you a clear view of which patients are done for the day.
Quick Tips
- Templates save time. Set up templates for your most common encounter types (comprehensive, follow-up, contact lens, medical) so the right fields appear automatically.
- Custom instructions shape the output. If the AI consistently formats something in a way you do not prefer, add a custom instruction (e.g., "Always note IOP method as Goldmann unless otherwise stated"). See Custom Instructions.
- You can always add more audio. Click Add Recording to capture a second recording---useful if a patient returns from dilation or you need to dictate additional findings.
- PDF and text export. If you need documentation outside your EHR, use the export options in the bottom action bar to download a PDF or copy chart text to clipboard.
- Encounter time tracking. Doctora tracks recording duration and chart review time automatically, which can support medical decision-making level coding.