Charts, claims, intake forms and imaging reports, read and filed on your own network. Recognition runs on your own hardware by default, and the AI can too. HIPAA · On-premise · Licensed once.
A chart is never really one document. It's an intake form, an insurance card, a consent form, lab results, and a physician's notes, arriving at different times from different people. Front-desk staff end up as the safety net for whether the file is actually complete, and that's a bad job to give someone who's also answering phones.
SimpleIndex runs the recognition step instead: read what's on the page, check it against what the file is supposed to contain, and file it where your EHR or practice management system expects it. By default no scanned image leaves your own hardware.
Six setups cover most of what a practice or hospital department needs from document capture. Each is the same software, aimed at a different part of the file.
Define what a complete intake file looks like, such as an insurance card, consent form, HIPAA acknowledgment and photo ID, and every batch is checked as it processes. A chart missing a signature comes back as a list, not a surprise during an audit.
Read payer, patient, claim number and amount off EOBs and remittance advices in whatever layout each payer uses, with a language model reading fields by their labels, and export them to your billing system.
Look a patient up by the identifier on the page, confirm the match, and file the scan where your electronic health record or practice management system picks it up: a folder, a database, or an OnBase or Laserfiche import.
Every processed page gets a text layer, so a chart pulled up six months later is searchable by content, not just by the folder someone filed it in.
Hand-printed names, dates and boxes on intake and consent forms are read as index values, and ticked boxes come back as yes or no. Neat print reads well; flowing cursive is unreliable with any method.
A language model reads the fields a template can't describe, such as "Patient Date of Birth" wherever it sits on the form. Run it on a model on your own server and no patient record leaves your network.
The rules that shape a patient record today, and where this software sits against each one.
That's the normal case. Everything above is the same software configured differently for a particular kind of chart or claim.
Send us a handful of de-identified charts or claims forms and we'll build a working job from them, before the call, on your own documents.
We ask what you're processing because the answer changes what we'd show you.
HIPAA compliance is about how your organization handles patient information, so no software is compliant on its own. SimpleIndex helps by keeping recognition and filing on your own hardware by default, so no outside service reads the pages unless you choose one.
No, unless you choose a cloud engine. SimpleOCR, Tesseract and ABBYY FineReader run on your own computer. Textract and cloud models such as ChatGPT are optional; if you use one, check your agreement with the provider first.
Yes. Point the job at a model on your own server, through Ollama, LM Studio or vLLM. It runs with no internet connection at all.
Neat hand print, yes, and how well depends on the writing and the engine. Cloud engines and language models read free handwriting best, but no method reads flowing cursive reliably. See handwriting recognition.
Yes, in Professional. Give an index field a pattern, turn on Redact Matches, and each matching value is blacked out in the exported PDF. Check redacted files before you share them.
It files scans where your EHR or practice management system picks them up: a watched folder, a database through ODBC, or an import module such as OnBase or Laserfiche. It doesn't need an API from the EHR.
SimpleIndex reads claim fields like any other form. If you process CMS-1500 or UB-04 claims at high volume, we also offer a dedicated claims solution, HealthClaimsOCR, on its own site.
Professional covers everything on this page, including the AI engines, handwriting and redaction. Searchable PDFs also work in Standard.
The free trial runs the full software on your own files, and our team in Knoxville can help you set up the first job.
