Document archaeology
Nurses and intake coordinators scroll unindexed PDFs page by page. A complex referral can take 60 to 90 minutes.
Clinical document intelligence for inbound fax
Stop document archaeology.
Healthcare needs info, not data. InfoNotData routes every inbound fax page by confidence in milliseconds, extracts only what your specialty’s intake checklist requires, and cites every value to its page and line.
Projected Proposed single-site VHA pilot: 30–40% fewer consults returned for missing information.
- No generated text
- Every value cited to page and line
- Uncertain pages go to a person
- Built for hospital security review and the VA RMF
01 / The problem
The fact was in the packet. Nobody could find it.
Outside records arrive as faxed images: hundreds of pages, no index, no search. Somewhere inside is the one value the specialist needs before they’ll see the patient.
Consults sent back for “missing” information
The ECG, troponin trend, or ejection fraction was in the packet all along. The referral bounces anyway, and the patient waits.
Duplicate tests
When nobody can find the outside result, the ECG or lab gets reordered. The cost lands on the system and on the patient.
AI bills that scale with paper
Running a full model pass over every raw page, cover sheets included, can reach six figures a month at enterprise volume.
Stored isn’t findable. And a summary isn’t a source.
See the InfoNotData Method ↓02 / The evidence so far
Projected for a proposed single-site VHA pilot.
These figures are projections prepared for a proposed pilot at one VA medical system. The pilot hasn’t run yet, so they are not results. We label them that way everywhere, and we’ll walk you through the assumptions.
25–35%
less time on pre-visit record review
Projected · proposed single-site VHA pilot
30–40%
fewer consults returned or cancelled for missing information
Projected · proposed single-site VHA pilot
11,000+
clinician hours reclaimed
Projected · proposed single-site VHA pilot
$4.5M
in annual soft-cost savings
Projected · proposed single-site VHA pilotWorks with the fax channels you already run
- RightFax and other fax servers
- Central mail and fax gateways
- DirectTrust
- eHealth Exchange
InfoNotData doesn’t integrate directly with EHRs today, so nothing in your EHR changes. Product names are trademarks of their owners and are listed only to describe where InfoNotData fits.
Built for your security review
- HIPAA
- FIPS-validated encryption
- VA Handbook 6500
- VA RMF and ATO process
- Section 508 / WCAG 2.1 AA
“Built for” means designed to meet these requirements. Ask us for current authorization status. Security and compliance
03 / Who it’s for
Anyone who receives records at volume and has to act on them.
- Health systems Central fax queues of hundreds of thousands of pages a month, and outside records lost in the chart.
- Community Care networks Referrals returned for missing documentation, on both the sending and the receiving side.
- Specialty groups Every incoming packet checked against your intake requirements before a patient can be scheduled.
- VA and federal health 800-page community care PDFs in JLV, and consults cancelled for missing information.
- Post-acute, home health and labs Records arrive as images, and someone still has to find the facts.
04 / The InfoNotData Method
Route. Extract. Check. Cite.
Four steps, in that order. The order is the point: nothing expensive runs until a page has earned it, and nothing reaches a clinician without its source.
-
Route
Every page is sorted Easy, Medium, or Hard in milliseconds, before any expensive model runs.
Stop paying full price for cover sheets. -
Extract
Pull only the values the intake checklist asks for. No summaries, no generated text.
Clinicians get facts, not paragraphs. -
Check
Match what was found against the specialty’s requirements, like an ECG within 90 days for cardiology.
Consults go out complete the first time. -
Cite
Link every value to its page and line. Send anything uncertain to a person.
Anyone can verify it in one click.
Easy · auto-processed
7,214 pages in this batch
High-confidence pages are processed automatically on the fast path.
Medium · extraction-only
1,906 pages in this batch
Pages that need a closer read get a targeted extraction pass. Values only, each with its source.
Hard · human review
312 pages in this batch
Low-confidence pages, like handwriting or a poor-quality fax, go to a person. Never a guess.
Check and cite
The specialist’s checklist, already run.
InfoNotData maps each specialty’s intake requirements to the values it finds. A coordinator sees what’s there, exactly where it is, and what’s missing before the consult goes out.
When something isn’t in the packet, it says so. That’s the moment to request it, not three weeks later.
How source citations work →
Cardiology consult readiness
Synthetic example
- ECG within 90 days 02/11 p.412 · L8
- Troponin trend 3 values · 02/09–02/10 p.388–391
- Left ventricular ejection fraction 35% p.212 · L14
- Current medication list 12 medications p.17 · L3
- Stress test result Not in packet · routed to intake nurse
4 of 5 requirements found · 846 pages · 3.8 s
05 / In practice
Specific numbers, labeled honestly.
Projected · proposed single-site VHA pilot
A primary care team preparing cardiology consults
- Problem
- PACT nurses spend up to 30 minutes per patient finding ECGs, troponin trends, and ejection fractions in outside records before a consult can go out.
- Approach
- Route inbound community care records, extract the cardiology checklist values, and cite each one to its page and line.
- Projected outcome
-
- 25–35%less time on pre-visit record review
- 30–40%fewer consults returned or cancelled for missing information
- 11,000+clinician hours reclaimed
- $4.5Min annual soft-cost savings
All figures are projections for a proposed single-site pilot at a VA medical system. The pilot has not run yet.
Scale
~1.8M
faxes a month at one Tampa Bay health system
That’s the inbound volume of a single regional system. InfoNotData is built for queues of 100,000 to more than 1,000,000 pages a month.
Design target
Up to 80%
lower OCR overhead, by design
Advanced pre-processing and a model selector decide which pages need which pass, so you stop paying for model passes that a page doesn’t need.
What InfoNotData doesn’t do
- Write summaries or any generated text about patients
- Guess at a value on a low-confidence page
- Replace your fax server or your fax numbers, or require changes to your EHR
- Hide what it couldn’t find
Filter before you spend
Size it for your own queue.
Put in your volume and your rates. Every assumption is on the screen, so you can take the math to your CFO.
InfoNotData’s router decides page by page which pass each page needs. The share you set here is your estimate, not a promise from us.
06 / Next step
See it on a real packet.
A 30-minute walkthrough with an engineer. We’ll run a synthetic or de-identified packet through routing, extraction, and citation, and show you exactly where it hands off to a person.
What happens next
- An engineer replies to set a time that works for your team.
- A 30-minute walkthrough on a synthetic or de-identified packet.
- If it fits, we scope a pilot on your own queue.
An engineer, not a sales script. We’ll show you what it doesn’t do, too.
- Security and complianceFor your privacy and security reviewers
- How routing worksThe Easy, Medium, and Hard router, step by step
Tampa, Florida