An AI audit agent re-reads the de-identified chart against the codes your coder selected, scores the coding and proposes what to add, remove or change - each with its reason. Your auditor accepts or rejects every proposal, and every decision is recorded.
The coding agent suggests. The audit agent challenges. Your auditor decides.
The audit agent receives the chart's de-identified documents and the codes the coder selected - the same evidence, a second set of eyes.
A summary, a score out of 100 and proposals to add, remove or change a code. A proposal for a code outside the chart's scheme is never shown.
Accept a proposal and it is applied to the codes. Reject it, and the reason is recorded. Then the auditor marks the chart audited.
Six steps from the doctor's words to a coded, audited and grouped chart, with people reviewing the results along the way.
Dictate or record the consultation with consent; the doctor approves the draft.
Exports PDF, JSON or FHIR R4Scans become text, and identifiers are masked before any AI sees them.
Suggested codes arrive with the passage behind them, for the coder to accept.
A second AI pass scores the coding and proposes changes for the auditor.
Documentation gaps become neutral, non-leading physician queries.
A rules-based grouper assigns the DRG, final when the chart is locked.
Auditing starts by itself once a chart is coded, and finishes with a person's sign-off.
Whether it was marked coded by a person, coded automatically or coded earlier, the chart is queued for audit without anyone having to ask.
The audit agent compares the selected codes with the documentation and returns a summary, a score out of 100 and proposals to add, remove or change a code. Each run is kept, so you can see how the audit changed after the documentation did.
On the Audit tab, beside the document, the auditor works through the proposals. An accepted proposal is applied to the selected codes; a rejected one needs a reason. Then the auditor marks the chart audited.
Once audited, the chart's clinical facts and code attributes are protected. Locking the chart for billing makes its codes - and its DRG, when grouped - final.
An auditor should never have to reverse-engineer a machine. Every proposal says what it would change and why, and the decision - with its reason when rejected - stays on the chart.
Illustration - a fictional chart.
Because every audit and every decision is recorded, coding quality becomes something you can measure, not guess.
The person who uploads a chart and the person who codes it are kept apart, so every audit is credited to the right coder.
A reporting view over every audit - scores and proposals per coder and per chart - for feedback and education.
Each auditor sees the charts assigned to them, with filters, bulk reassignment and a notification when a chart becomes theirs.
Every code change, status change, proposal and review is written to the chart's log, with who did it and when.
The audit never waits for the DRG, and the DRG never reopens the audit. They simply stay in step.
Claims Validator™ checks billed services against the clinical evidence for insurers and revenue cycle teams, with risk scores and audit-ready reports.
Explore Claims Validator™A demo account opens the whole platform - code a chart, watch the audit agent score it, and accept or reject its proposals.
Start a Free Demo Talk to us first