arrow_backAll modules
switch_access_2

RadMatch

AI

Study-description matching for teleradiology & priors

A fast ML engine that maps free-text study descriptions to your exam codes — a lifesaver for teleradiology groups and for normalizing priors brought in from outside facilities. LLM fallback for hard cases, review queue for anything low-confidence.

RadMatchMatching
4,127
Matched today
0
In review queue
CT CHEST W CONTRASTOutside CD
arrow_forward71260CT Chest w/ contrast98%
MRI BRAIN WO/WTelerad
arrow_forward70553MR Brain w/o + w/96%
XR CHEST PA/LATOutside CD
arrow_forward71046XR Chest 2-view99%

RadMatch learns from your own code sets to translate inconsistent, free-text procedure descriptions into the correct standardized exam codes — the reconciliation headache every teleradiology group knows, and the mess that comes with priors pulled from outside facilities. Upload a code set, train a model, and let RadMatch resolve incoming descriptions in milliseconds, falling back to an LLM only when the statistical match is uncertain. Anything below a confidence threshold lands in a review queue so a human stays in the loop.

  • check_circleNormalize priors brought in from outside facilities
  • check_circleBuilt for teleradiology and multi-site reads
  • check_circleTrain models directly from your uploaded code sets (CSV)
  • check_circleDescription → exam-code matching in milliseconds, LLM fallback
  • check_circleConfigurable confidence thresholds and review queue
  • check_circleTerm overrides and custom mappings that persist across retrains
switch_access_2RadMatch field guideStudy Description Normalization for PACS Migrations and Outside Imaging ImportsNormalize outside study names to local terminology during PACS migrations, patient-disc imports, and teleradiology workflows—at the HL7 level, the DICOM level, or both.Read the articlearrow_forward

See RadMatch in your workflow

Request a walkthrough and pricing tailored to your environment.

mailRequest a Demo