# Medical Record Annotation Tool > Medical record annotation for highlights, notes, and tags on record pages, shared across the review team, pinned to its page and never altering the record. Canonical page: https://medrecords.ai/product/medical-record-annotation/ --- NewWe now read the actual MRI and CT imaging — not just the radiologist's 1-page report. [See how →](https://medrecords.ai/product/imaging/) [Home](https://medrecords.ai/) ›[Product](https://medrecords.ai/product/) › — Record Annotation & Collaborative Review MEDICAL RECORD ANNOTATION TOOL ## Mark up the record together — without losing where anything came from. A medical record annotation tool for marking up the file together: highlights, notes, and tags placed directly on record pages and shared across the review team. Every annotation is pinned to its page — filterable, exportable as cited work product, and kept separate from the record itself, so the record stays the record. [Test a file ](https://medrecords.ai/test-a-file/?src=product-medical-record-annotation) [Book a demo](https://medrecords.ai/demo/) Adams, Timothy — right knee · Case #IME-4812 342 pages · 2 packets Progress note · p.140 2 notes Patient reports right knee improving since 4/02, seventh follow-up visit; home exercise program continued. RK Compare against the 4/02 visit — improvement trend matters for the exam. JM Added to exam-prep list. Tagging for the chronology cross-check. #improvement #exam-prep WORK PRODUCT · SEPARATE LAYER In action — Case #IME-4812 2 reviewers, 342 pages, one annotation layer — **every note pinned to its page** and none of it written into the record. ### Pinned to the page, not to a position. Highlight a passage, drop a note, or tag a page, and the annotation anchors to that exact location in the source document. Re-sort the file, deduplicate it, rebuild the packet — the markup follows its page. Filter the layer by tag, author, or document and the whole review snaps into view. Highlights, notes, and tags anchored to exact page locations Survives re-sorting, dedup, and packet-building #exam-prep ✕ filtered view Improvement trend noted p.140 — RK MRI report — key finding highlighted p.342 — JM Dictation addendum — verify wording p.2 — RK File re-sorted by provider after these notes were made — every pin still holds. Review assignments Case #IME-4812 RK — Packet 1 — treatment records REVIEWED JM — Packet 2 — imaging & billing IN PROGRESS EX — External expert — record only NO ANNOTATION ACCESS ### One file, a whole team — with roles that mean something. Split the review by packet, provider, or tag and assign it. Threads keep the discussion attached to the evidence it's about, and visibility follows case roles: private notes stay private, team notes stay in the team, and an external reviewer sees the record without your thinking layered on it. Assignments and coverage tracking across the file Per-role visibility; annotation access is granted, never assumed ### Export the review as cited work product. Filter the annotation layer down to what matters — one tag, one author, one document range — and export it as a digest where every note carries its page reference. Exam-prep lists, deposition outlines, and case memos start from markup that already knows where its evidence lives. Filter by tag, author, date, or document before export Every exported note cites its source page Export — #exam-prep digest Word · HTML Improvement trend since 4/02 — confirm at exam; compare stated history. p.140 RK · note MRI finding highlighted for the evaluator — read alongside the ordering note. p.342 JM · highlight The digest reads like a memo and cites like the platform — page references on every line. 2 layers · one file never merged The record evidence · immutable Source pages, exactly as ingested. Producible at any time, exactly as received. The annotation layer work product · access-controlled Your team's highlights, notes, and tags — pinned to pages, visible by role, and never burned into the record. Every annotation view and edit logged in the audit trail Work product, kept apart ### The record stays the record. Your thinking stays yours. Annotations never alter a source page, so producing the file never means producing your strategy. The separation is structural, not procedural: 2 layers, distinct permissions, and an audit trail on both — audit-grade and legally defensible by design. [See Command Workspace ](https://medrecords.ai/product/command-workspace/) ### From raw file to reviewed file. 3 steps, and the whole team works one copy instead of 6. STEP 1 Upload and assign The file is read, sorted, and indexed; sections are assigned to reviewers by packet, provider, or tag. STEP 2 Annotate together Highlights, notes, tags, and threads accumulate on the pages themselves — visible by role, pinned for good. STEP 3 Filter and export Slice the annotation layer by tag or author and export a cited digest — the review becomes work product. ### Who reviews records with it. Any team where more than one person reads the same thousand pages. [ IME organizations Staff pre-mark the file; the evaluator walks in with the key pages flagged. For IME orgs ](https://medrecords.ai/solutions/medical-evaluators/) [ Law firms Paralegal, associate, and partner on one annotation layer — work product kept apart. For law firms ](https://medrecords.ai/solutions/law-firms/) [ TPAs & case management Nurse reviewers and adjusters share findings without emailing marked-up PDFs. For TPAs ](https://medrecords.ai/solutions/tpas/) [ Insurance carriers Claim teams keep review history on the file itself, with access logged per role. For carriers ](https://medrecords.ai/solutions/insurance-carriers/) FAQ ### Record annotation, answered. No. Annotations live in a separate, access-controlled layer on top of the record — the underlying pages are never altered. That separation matters legally: your team's impressions and strategy notes are work product, and keeping them out of the record itself means producing the file never means producing your thinking. Annotation visibility follows case roles and permissions. A note can be private to its author, shared with the review team, or scoped to a group — and an external reviewer given access to the record doesn't see the internal annotation layer unless you grant it. Access is managed per case, and viewing is logged like any other PHI access. Yes. Each annotation is pinned to its exact page and location in the source document, not to a position in a list. Re-sort the file, deduplicate it, or rebuild the packet, and every highlight and note stays attached to the same page it was made on. Yes. Filter the annotation layer — by tag, author, date, or document — and export the result as a cited digest: each note alongside its page reference, ready for exam prep, deposition outlines, or a case memo. The export cites pages the same way the rest of the platform does. Split the file by document range, provider, or tag and assign sections to reviewers. Threads on an annotation keep the discussion pinned to the evidence it's about, and a filterable review view shows what's been covered and what's still open — so 2 reviewers never silently duplicate the same thousand pages. ### Related capabilities. Annotation is how the team works the file — these are the surfaces it works on. [ Command Workspace Source, timeline, and report on one synced screen — annotate without leaving it. Explore ](https://medrecords.ai/product/command-workspace/) [ Smart List View Source-linked lists of every diagnosis, medication, and provider — tag from the list. Explore ](https://medrecords.ai/product/smart-lists/) [ Audit Trail & Chain of Custody Every annotation view and edit logged — the layer is governed, not informal. Explore ](https://medrecords.ai/product/audit-trail-chain-of-custody/) ### Put the whole team on one annotated file. Upload a file, invite your reviewers, and see highlights, threads, and cited exports on your own record. Handled under our BAA; never used to train a model. [Test a file ](https://medrecords.ai/test-a-file/?src=product-medical-record-annotation) [Book a demo](https://medrecords.ai/demo/) [All capabilities ](https://medrecords.ai/product/)