One workflow from claim packet to evidence report.
ClaimClean brings three audit-support capabilities into one workflow: Anonymization Service (ANS) protects sensitive patient information, Medical Fraud Filter (MF2) screens claims against configurable rules and policy checks, and Audit Insights Assist (AIA) generates evidence-rich reports for human reviewers. Together, they form an end-to-end review layer for U.S. health insurers — designed to reduce audit turnaround time by automating triage, policy checks, evidence extraction, and report generation.
Six stages. Human control where it counts.
Every claim moves left to right through the same auditable path. PHI is minimized before analysis, and qualified reviewers hold the decision points.
Claim packet intake
Claim documents, medical records, and attachments arrive through a structured, secure intake. Out: a prepared case ready for protection.
Human control pointAnonymization Service
ANS detects PHI in the prepared case and applies minimization and de-identification workflows. Out: protected data for downstream analysis.
Medical Fraud Filter
MF2 screens the protected claim against configurable rules, policies, and guideline checks. Out: flags and a prioritized review queue.
Audit Insights Assist
AIA analyzes records on prioritized claims, extracts evidence, and highlights potential review signals. Out: draft insights linked to source.
Auditor review
A qualified reviewer validates, dismisses, or annotates each signal against the linked evidence. Out: confirmed findings and next actions.
Human control pointEvidence report
Auditor decisions are assembled into an evidence-ready report for adjudication support, appeals, or escalation. Out: exportable findings.
Human control pointThree capabilities, in workflow order.
Each module solves a distinct problem — data protection, screening, and insight generation — and each hands clean output to the next.
Anonymization Service
Protects sensitive patient information before claim data is used for analytics, review, or AI-assisted workflows.
- Detect and reduce PHI exposure
- Support de-identification and redaction workflows
- Help teams avoid unsafe consumer-AI data sharing
- Enable privacy-aware analytics and audit review
Medical Fraud Filter
Screens claims using configurable policy and guideline checks aligned with Centers for Medicare & Medicaid Services (CMS) guidance, payer policies, and internal audit priorities.
- Apply predefined and configurable rule sets
- Flag claims for deeper review
- Support consistent policy application
- Prioritize audit queues with explainable findings
Audit Insights Assist
Helps reviewers analyze medical records, surface relevant evidence, and generate graphical reports that highlight potential review signals.
- Analyze claim documents and medical records
- Highlight potential fraud, waste, abuse, and inconsistency indicators
- Generate visual summaries and evidence-backed reports
- Keep human auditors in the final decision loop
PHI minimization happens before analysis, not after.
Protected Health Information is detected and minimized at the front of the workflow, so screening and record analysis operate on protected data. Each pilot is deployed with clear data-retention, access-control, and audit-log policies.
- Secure intakeClaim packets and records enter through a structured, controlled workflow — never a public upload.
- PHI minimizationANS detects sensitive fields and applies de-identification and redaction before anything else runs.
- Controlled analysisMF2 and AIA operate on protected data under configured policy and guideline checks.
- Auditor validationReviewers confirm, dismiss, or annotate every signal against its linked evidence.
- Exportable evidenceValidated findings become evidence-ready reports with an audit trail of decisions.
What reviewers see, and what leaves the platform.
The review workspace is built for professional auditors: a prioritized queue in, an evidence-ready report out, with every insight traceable to its source along the way.
Prioritized queue
Claims flagged by MF2 arrive ordered by configured audit priority, with the rationale for each flag attached.
Linked evidence
Every review signal points to the exact passage in the source record, so findings can be validated or challenged directly.
Report drafts
AIA assembles claim summaries, signal breakdowns, and visual summaries into drafts the auditor edits and approves.
Notes & status
Reviewer annotations and validation status are captured with each finding, forming an audit trail through the final export.
Start scoped. Deploy on your terms.
ClaimClean engagements begin with a defined pilot, so your team can evaluate the workflow on real cases with clear boundaries.
Pilot-first engagement
Every engagement starts with a scoped pilot: a defined claim set, agreed success criteria, and a joint review of results before any wider rollout.
Controlled deployment options
Deployment models are discussed during scoping to match your organization's security and data-handling requirements, with retention, access, and audit-log policies agreed up front.
Secure intake & export
Pilot data moves through structured intake and export workflows designed for sensitive claim documents. There are no public uploads — including on this website.
See the full workflow on your own claims.
Request a pilot and we'll scope a defined claim set, agree success criteria, and walk the workflow end to end — intake to evidence report.