# Negative Findings & Gap Flagging Software (Beta) > Negative findings and documentation-gap flagging software, in beta: surfaces explicit negatives and expected-but-undocumented facts, kept clearly distinct. Canonical page: https://medrecords.ai/product/negative-findings-detection/ --- 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/) › — Negative Findings & Documentation-Gap Flagging NEGATIVE FINDINGS & DOCUMENTATION-GAP FLAGGING ## What the record does not document can matter too. BETA Negative findings and documentation-gap flagging software from Medrecords AI surfaces explicit negatives and expected-but-undocumented facts across the file, keeping "not documented," "explicitly denied," "unknown," and "not applicable" as 4 clearly distinct states, never collapsed into one. Live in beta and testable today. [Join the beta ](https://medrecords.ai/demo/?src=beta-negative-findings-detection) [Book a demo](https://medrecords.ai/demo/) Adams, Timothy — right knee · Case #IME-4812 BETA 4 states, never collapsed Numbness — explicitly denied p.129 Prior surgery — not documented gap Substance use — unknown p.140 Pediatric history — not applicable n/a 342 pages scanned · **One documentation gap** flagged for review. IN ACTIVE BETA Refined hands-on with early customers, reading **what the file says, and what it leaves silent** . ### 4 states, never collapsed into one. The record can say a symptom is explicitly denied, say nothing about it at all, leave it genuinely unknown, or note it doesn't apply — and those are 4 different facts. Each one is tagged distinctly and cited to its source, so "not documented" never quietly reads as "did not happen." Explicit negative, not documented, unknown, and N/A stay visually distinct Every tag cited to the exact page it was drawn from The 4 states Explicitly denied — stated in record Not documented — expected, absent Unknown — stated as unknown Not applicable — doesn't apply here Gap detail Document type — Discharge summary Expected field — Follow-up plan Status — absent — flagged ### Expected fields, flagged when missing. For a given document type, certain fields are ordinarily present — a discharge summary usually documents a follow-up plan, an intake note usually documents a surgical history question. When one of those expected fields is simply absent, it's flagged as a documentation gap and cited to the document type and the expected-field basis, not asserted as a fact about the patient. Gaps shown with the document-type basis for the expectation Material conclusions require a reviewer's confirmation first Gap state NOT DOCUMENTED ≠ DID NOT HAPPEN What gets flagged · Fields expected for the document type, but absent · The document type and expected-field basis, shown What requires a human · Any material conclusion drawn from a gap · Confirming a gap before it feeds a report The boundary ### Absent from the record is not the same as absent from the patient's life. The system never turns a documentation gap into a factual negative. If a field is missing, it is shown as "not documented" — never rewritten as "did not happen," and never merged with an explicit denial that a clinician actually recorded. Every gap carries the document type and the expected-field basis that made the absence flaggable in the first place. Any material conclusion drawn from a gap — for instance, arguing that a missing note means a symptom wasn't present — requires human confirmation before it's used. The tool surfaces where the record is silent; deciding what that silence means stays a human judgment call. ### From silence in the chart to a flagged gap. 3 steps, with the reviewer as the final word on what a gap means. 01 Fields checked per document type Every relevant topic or field is checked against what the document type would ordinarily contain. 02 Outcomes tagged into 4 states Explicit negative, not documented, unknown, or not applicable — never collapsed into one. 03 Material conclusions need confirmation Anything drawn from a gap is routed to a reviewer before it's treated as settled. ### Who reads the flagged gaps. The same distinctions, useful wherever silence in the record needs a careful read. [ Defense lawyers Spot exactly which expected fields are missing from a plaintiff's own records, cited. For defense counsel ](https://medrecords.ai/solutions/legal/) [ TPAs & case management Flag documentation gaps across a claim file without re-reading it cover to cover. For TPAs ](https://medrecords.ai/solutions/tpas/) [ Medical evaluators See what the record doesn't say before treating silence as an answer. For evaluators ](https://medrecords.ai/solutions/medical-evaluators/) FAQ ### Negative findings & documentation gaps, answered. A feature that surfaces both explicit negatives a clinician recorded ("denies numbness") and expected-but-undocumented facts (a field a document type usually contains but this one doesn't), while keeping those, plus "unknown" and "not applicable," as 4 clearly distinct states. No, and the tool is built specifically so it can't be read that way. "Not documented" is shown as its own state — a gap in the paperwork, not a fact about the patient. Only an explicit denial that a clinician actually wrote down is tagged as a negative. Expectations are tied to the document type — a discharge summary, an intake note, a specialist consult, and so on each carry typical fields. When one of those fields is absent from a specific document, that absence is flagged and shown alongside the document type and the expected-field basis. Yes, in beta. Negative findings & documentation-gap flagging is live and testable now; we're refining it hands-on with early customers, and if your use case is a good fit we'll work with you directly. Not without review. Any material conclusion drawn from a gap requires human confirmation first — the feature surfaces where the record is silent, and a person decides what, if anything, that silence means. ### Related capabilities. What sits alongside gap flagging, live today or in this same beta batch. [ Missing Medical Records Identification Cross-references treatment history against what was produced and flags what should exist but doesn't. Explore ](https://medrecords.ai/product/missing-records-identification/) [ Exam Prep Pack The gaps and open questions bundled into what an evaluator needs walking into the exam. Explore ](https://medrecords.ai/product/exam-prep-pack/) [ Structured Patient History & Recovery-Factor Extraction History extracted into one cited profile — patient-reported kept separate from confirmed. Explore ](https://medrecords.ai/product/patient-history-recovery-factors/) ### See what your record leaves silent. Surface explicit negatives and expected-but-undocumented facts, with "not documented" always kept distinct from "did not happen." Join the beta on one of your own files, or book a demo first. [Join the beta ](https://medrecords.ai/demo/?src=beta-negative-findings-detection) [Book a demo](https://medrecords.ai/demo/) [All capabilities →](https://medrecords.ai/product/)