Medical record review AI for payers and TPAs.
YIRA MedSense reads the medical records behind every claim and pre-authorisation, from histories and discharge summaries to lab reports and itemised bills, and gives clinical reviewers a structured file with diagnoses, ICD-10 codes, medications, flagged lab values and prior-record matches. Built for health plans, insurers and TPAs worldwide, including the US.
Medically reviewed by Dr. N. S. Santosh Varma, MBBS, FIIM, FIDM, FICCM · Medical Officer, YIRA MedSense · Updated 13 October 2026
{
"documents": 14,
"diagnoses": ["…"], "icd10": ["…"],
"medications": ["…"],
"lab_flags": [ { "test": "ALT", "flag": "High" } ],
"bill_lines": "extracted",
"prior_record_match": "flagged",
"next": "clinical reviewer"
}Illustrative structure.
Clinical reviewers spend their day reading, not reviewing.
Medical record review sits behind pre-authorisation, claim review, payment integrity and audits. Most of the time goes into finding facts in long, inconsistent records.
Long records
Dozens of pages per claim, many scanned or handwritten.
Inconsistent formats
Every provider and lab formats records differently.
Hidden signals
Reused documents, abnormal values and billing gaps are buried in the pages.
Expensive reviewers
Nurses and medical officers spend hours on data entry instead of judgement.
From record pile to review-ready file.
Ingest
Records arrive by API, mobile app or bulk scanner portal: PDFs, scans, handwritten notes and photos.
Extract
Clinical NLP identifies diagnoses, ICD-10 codes, medications, procedures, lab values and bill lines.
Check
Lab values are flagged against reference ranges, documents are matched against prior records, and the member's photo can be compared with the hospital photo.
Deliver
A structured JSON file goes to your review, claims or payment-integrity system by API.
One engine, four review workflows.
| Workflow | What MedSense does |
|---|---|
| Prior authorisation / pre-auth | Summarises the clinical history and supporting reports for the requested treatment |
| Claim review | Structures the record and bill so reviewers check facts instead of finding them |
| Payment integrity | Flags reused documents, identity mismatches and bill lines that do not match agreed rates |
| Audit and SIU | Re-reads past claims at scale and surfaces records that match earlier files |
Built for claims, not generic OCR.
Clinical understanding
ICD-10 codes, medications and lab flags, not just text.
Fraud signals included
Prior-record matching and facial verification in the same file.
Human in the loop
MedSense prepares the review; your clinicians decide.
Deploy your way
Cloud API or client-managed deployment inside your own infrastructure.
In production, at scale.
MedSense runs in production at Health Assist Insurance TPA Pvt Ltd. It has parsed more than 1 lakh (100,000+) medical documents, flagged over 5,000 fraud cases for review and supported 100+ insurance operations.
Sandbox
Test MedSense on your own sample documents.
Provision
Set up document types, fields and rules.
Integrate
Connect by API to your claims or payer system.
Go live
Run in parallel, then switch over.
Timing depends on scope, data access and integration readiness. Client-managed deployment is available, so MedSense can run inside your own infrastructure.
Questions from claims teams
What is medical record review AI?
Software that reads medical records, such as histories, discharge summaries, lab reports and bills, and turns them into structured data so clinical reviewers at health plans, insurers and TPAs can make decisions faster. YIRA MedSense also adds fraud signals such as prior-record matching and facial verification.
Is MedSense available for US health plans and TPAs?
Yes. MedSense is offered to payers and TPAs worldwide, including the US. It is built to SOC 2, HIPAA and GDPR requirements and supports client-managed deployment.
Does MedSense make coverage or medical-necessity decisions?
No. MedSense prepares and structures the record and raises review signals. Clinicians and reviewers make every decision.
Which codes does MedSense extract?
MedSense identifies diagnoses and ICD-10 codes, along with medications, procedures, lab values and billing line items.
Can MedSense read handwritten and scanned records?
Yes. Scanned PDFs, photos and handwritten medical histories are converted into structured fields.
How does it connect to our systems?
Structured JSON is delivered by API to your claims, utilisation-management or payment-integrity platform.
How long does implementation take?
About 4–5 weeks from sandbox to go-live, depending on scope, data access and integration readiness.
Book your demo today.
See MedSense on the documents and fields that matter to your claims team.