# Medical Record Sorting Services > Upload the raw production (mixed PDFs, faxes, discs); sorting, deduplication, wrong-patient quarantine, and packet building happen at upload, audited. Canonical page: https://medrecords.ai/solutions/medical-record-sorting/ --- 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/) ›[Solutions](https://medrecords.ai/solutions/) › — Medical Record Sorting MEDICAL RECORD SORTING SERVICES ## Scattered records in. A sorted, deduplicated, Bates-ready case file out. Medical record sorting that happens at upload: send in the raw production (mixed PDFs, fax runs, disc images) and sorting, deduplication, wrong-patient quarantine, and packet building run automatically. Every page kept, merged, or flagged lands in a per-page audit log, and the case file exports Bates-ready, ordered by encounter and provider. [Test a file ](https://medrecords.ai/test-a-file/?src=medical-record-sorting) [Book a demo](https://medrecords.ai/demo/) Click any chip → the source page it cites Sorted case file · Case #IME-4812 at upload Adams, Timothy · right knee 342 pages sorted into 2 packets sorted 11 duplicate pages removed deduplicated Wrong-patient page p.140 quarantined quarantined Pages 342 Packets 2 Dupes 11 ### The review can't start until the sorting ends. Most productions land as a pile: mixed PDFs, fax runs, disc images, the same page 3 times. So firms burn days of paralegal time, or ship the pile to a sorting vendor billed by the page. Keep it in-house instead: the sort happens at upload, and every kept, merged, or flagged page is logged. The same deliverable, 2 ways Turnaround Days of paralegal time At upload Pricing $0.20–$1.50 / page outsourced Included Duplicates Eyeballed Detected + logged Data control Records at a vendor Never leave your control Sort manifest · Case #IME-4812 342 pp in Packets built — 2 · chronological — sorted Duplicates — 11 pages removed — logged Wrong patient — p.140 quarantined — flagged Bates numbering — Ready on export — ready Page accounting — 342 of 342 in the log — 100% Every page kept, merged, or flagged · nothing silently destroyed How it works ### 3 steps between the pile and the packet. 01 #### Upload the raw production Mixed PDFs, fax runs, disc images, portal exports: as they arrived, in any order. In the demo case, 342 pages in one unsorted upload. 02 #### The AI sorts, dedupes, and quarantines Pages are ordered by encounter and provider, duplicates detected and removed, wrong-patient pages quarantined before analysis. Every decision lands in the per-page log. 03 #### You review and export Spot-check the manifest, restore any removed page with a click, and export Bates-ready packets with the audit log alongside. Your file, your control, minutes after upload. Nothing silently destroyed ### Every page accounted for, including the ones removed. A sorted file is only trustworthy if you can say what happened to every page in the original production. Deduplication here is not deletion: each removed page is logged with a pointer to the copy that stayed, and can be restored with a click. In the demo case, page 140 belonged to a different patient. It was quarantined before analysis and logged, not quietly dropped or quietly used. When opposing counsel asks how the file was assembled, the audit trail answers for you. Audit trail · Case #IME-4812 exportable 08:12 — Raw production received · 342 pp — system 08:19 — Sorted into 2 packets · chronological — system 08:24 — 11 duplicate pages removed · logged — system 08:31 — p.140 quarantined — wrong patient — system 08:47 — Manifest reviewed · nothing restored — paralegal 08:52 — Bates-ready packets exported — paralegal Every access logged · file deleted 30 days after delivery Why Medrecords AI ### The rules the platform never breaks. EVERY LINE CITED CASE #IME-4812 · ADAMS, T. 342 pp 2/14 — ER visit, right knee p.4 4/18 — arthroscopic surgery p.61 p.140 — wrong patient quarantined Medrecords AI Read every page · cite every line #### HIPAA, under a signed BAA Every file is handled under our Business Associate Agreement, from the first byte. #### Never trains a model Your records are never used to train any AI model — ours or anyone else's. #### Every line cited If we can't cite it, we don't say it. Every sentence links to its source page. #### Deleted after delivery Files are deleted 30 days after delivery, with a full audit log of every access. Powered by the platform ### 4 capabilities behind every clean file. The sorted case file is the first deliverable of the same platform that reads and cites the whole record. [ #### Document sorting & packet builder Mixed uploads become ordered packets: by provider, by encounter, in chronological order, ready for Bates numbering on export. IN ACTION · INPUT 342 pp raw → OUTPUT 2 packets · chronological ](https://medrecords.ai/product/document-sorting-packet-builder/) [ #### Medical record deduplication Exact and near-duplicate pages are detected, removed, and logged — with a pointer to the copy that stayed and a 1-click restore. IN ACTION · INPUT 342 pp raw → OUTPUT 11 dupes removed · each one logged ](https://medrecords.ai/product/medical-record-deduplication/) [ #### Co-mingled records detection Pages that belong to a different patient are caught and quarantined before analysis, so they never contaminate the file or the review. IN ACTION · INPUT 342 pp raw → OUTPUT 1 page quarantined · p.140, wrong patient ](https://medrecords.ai/product/co-mingled-records-detection/) [ #### Missing records identification Gaps in the treatment thread are surfaced: the visits and reports the production implies but does not contain, listed for follow-up. IN ACTION · INPUT 7-visit treatment thread → OUTPUT gaps flagged for follow-up ](https://medrecords.ai/product/missing-records-identification/) FAQ ### Medical record sorting, answered. Yes. Sorted packets export as bookmarked, Bates-ready PDFs in chronological order, with a page index, and the per-page audit log exports alongside them. In the demo case, 342 raw pages became 2 clean packets. They are removed from the working packet, but never silently destroyed. Every removed page is logged with a pointer to the copy that stayed, and you can restore any of them with a click. In the demo case, 11 duplicate pages were removed and logged. They are quarantined before analysis, so they never contaminate the chronology, the packets, or the review. In the demo case, page 140 belonged to a different patient; it was quarantined and logged, not folded into the file. No. Medrecords AI does not retrieve records from providers or facilities. You bring the records you already have — review starts in minutes from upload. Retrieval vendors take days; you can keep yours and still cut the review to minutes. Upload them to the same case. They are sorted into the existing packets, deduplicated against what is already there, and the manifest and audit log update to match. No re-sorting from scratch, and no second vendor invoice. ### Send the mess. Get back a case file. Sorted, deduplicated, quarantined — and logged page by page. [Test a file ](https://medrecords.ai/test-a-file/?src=medical-record-sorting) [Book a demo](https://medrecords.ai/demo/) [See all specialties →](https://medrecords.ai/solutions/medical-record-review/)