For teams · Radiology over read services

Radiology over read services compared: overread providers, X-ray overread software for urgent care, and which fracture detection AI holds FDA clearance

The short answer

A radiology over read is a formal interpretation of an imaging study by a board certified radiologist, produced after a non radiologist has already looked at the film and treated the patient. Urgent care centers, chiropractic clinics, orthopedic practices and mobile X-ray providers buy overreads because the clinician who treated the patient is usually not the person qualified to sign the diagnostic report. The purchase has two parts that buyers routinely confuse: the human read, bought from a teleradiology group on a per study rate, and the software layer that flags suspected findings the moment the image is acquired, so the treating clinician is not making a disposition decision on their own for the next several hours. Radiological.ai is the second part. It flags suspected findings on the X-ray at the point of care and drafts the structured report, and the radiologist still reviews and signs every read.

An urgent care that shoots its own films has a gap measured in hours. The patient is standing in front of you, the X-ray is on the screen, and the person qualified to render the diagnostic interpretation is a contracted radiologist who will get to it sometime tonight. You splint, you discharge, and the overread arrives later. Most of the time it agrees with you. The times it does not are the reason the contract exists.

Radiological.ai closes that gap on the software side. It reviews the study as it is acquired, flags regions worth a second look before the patient is discharged, and drafts the structured report so the radiologist doing the formal overread starts from a populated template rather than a blank one. It is decision support, not a replacement for the overread: the radiologist reviews and signs every read, and the discrepancy rate is exactly why you keep them.

Last updated September 2026

The Reading Station

Assistant ready

Worklist

SERIES 1 · AX

SLICE 24/64

W 80 · L 40

NOT FOR DIAGNOSTIC USE

Structured report

Draft

Run the assistant to draft this report for review.

You review & sign

Illustrative study. Not for diagnostic use.

Drafted in · you review & sign Worklist re-prioritized

Decision support for qualified clinicians. Radiological.ai does not provide a diagnosis and is not a substitute for professional judgment.

Run the assistant

Flag · prioritize · draft · you review and sign

X-RAY CT MRI BUILT WITH RADIOLOGISTS

Decision support not a diagnosis

You review & sign

If the question is who reads the studies rather than what flags them, our page on teleradiology companies covers provider selection, contract shape and state licensure, and teleradiology software covers the platform your contracted radiologists actually work in. AI fracture detection goes deeper on the extremity films that make up most of an urgent care X-ray volume.

On the money side, teleradiology pricing breaks down per study read rates and contract minimums against federal award records, and radiology AI CPT codes and reimbursement covers whether the professional component and the algorithm itself are separately billable. FDA approved radiology AI lists the cleared products by product code so you can verify a K number before you sign.

Primary source

Every AI product FDA cleared to detect a fracture, and the six companies that hold them

Fracture detection AI clears under product code QBS, which the FDA defines as radiological computer assisted detection and diagnosis software for fracture, regulated at 21 CFR 892.2090 as a class II device. There are 11 such records in the entire openFDA database and they belong to six companies. If a vendor is selling you AI fracture detection and is not on this list, ask which K number it clears under.

Company and product	Fracture (QBS) records	Most recent record	What the record covers
AZmed (Rayvolve)	4	K261378, August 28, 2026	Rayvolve AZchest, the most recent fracture clearance on file
Gleamer (BoneView)	2	K222176, March 2, 2023	BoneView. Gleamer also holds ChestView under MYN and BoneMetrics under QIH
Imagen Technologies (FractureDetect, OsteoDetect)	2	K193417, July 30, 2020	Includes DEN180005, the May 2018 De Novo grant that created this category
Milvue (TechCare Trauma)	1	K242171, January 17, 2025	Trauma reading support
Icometrix (icobrain aria)	1	K240712, November 7, 2024	Cleared under the same fracture code
GE Medical Systems (Critical Care Suite)	1	K223491, May 25, 2023	Critical Care Suite with pneumothorax detection
Experity and other overread service providers	0	No 510(k) on record	A teleradiology service is not a device. Read this zero correctly, it is expected

Source: openFDA 510(k) database, queried by product code QBS on September 9, 2026, and the FDA device classification endpoint for the code definition. A count of zero against a service provider is not a warning sign: a company that contracts radiologists to read studies is selling professional services, not software that makes a diagnostic claim, so it has nothing to clear. The zero matters only when a company markets an algorithm.

Why it works

What your group gets with radiology over read services

The overread and the AI are two different purchases, and conflating them is how clinics overpay

A teleradiology overread contract buys you a physician signature and the liability position that comes with it. Software buys you speed at the point of care. They are priced differently, they are procured differently, and one does not substitute for the other. Clinics that treat AI as a cheaper overread end up with neither a defensible report nor a billable professional component. Clinics that treat the overread as sufficient accept a multi hour window in which the treating clinician has no second opinion at all. The workable arrangement is both: flags at acquisition, signature within the contracted turnaround.

Turnaround is the only service level worth negotiating hard, and it is measured in two numbers

Ask any prospective provider for STAT and routine turnaround separately, as measured medians rather than targets, and ask what percentage of studies breach each. A provider quoting one blended average is hiding the tail, and the tail is what generates the callbacks. Experity, the largest provider in the urgent care segment, publishes averages of 10.78 minutes for STAT reads and 22.31 minutes for routine interpretations across more than five million studies a year, which is a useful public benchmark to hold a smaller competitor against.

Ask which product code the AI clears under, not whether it is FDA cleared

Every vendor in this space says FDA cleared, and almost all of them are telling the truth about something. The question that separates them is what. QBS is fracture detection. MYN is a medical image analyzer. QIH is automated image processing. QAS is triage and notification. QFM is prioritization for lesions. A product cleared under QIH has not been evaluated for its ability to call a fracture, and in an urgent care that reads mostly extremity films, that distinction is the whole purchase. The table above gives you the six companies whose records actually say fracture.

What it handles

Flagged, prioritized and drafted for your review

The assistant pre-reads each study, surfaces a region of interest for review, re-prioritizes the worklist, and drafts the structured report in your template. You confirm, edit and sign.

Flags suspected findings on the X-ray at the point of care, before the patient is discharged
Drafts the structured report so the overreading radiologist starts from a populated template
Triages the study queue so suspected time critical findings surface ahead of routine films
Hands a consistent report format back to the referring clinician and into the chart

RADIOLOGY OVER READ SERVICES STAT

Region of interest flagged for review

A focal region is surfaced on the sample study for the radiologist to review. The assistant does not characterize it as a diagnosis.

Draft impression

Suspected finding flagged for radiologist review. Correlate clinically and confirm. Draft for review and sign-off.

Illustrative sample · not for diagnostic use You review & sign

Why Radiological.ai

One assistant across the whole read

Not three vendors stitched together. Flag, prioritize and draft in one calm pane, on X-ray, CT and MRI, with the radiologist signing every study.

Flags suspected findings

A second set of eyes surfaces regions of interest for review on every study, so a suspected finding is less likely to slip past late in a shift.

Prioritizes the worklist

Suspected-critical studies move to the top, so urgent reads surface ahead of routine follow-ups across your sites and shifts.

Drafts the report

A structured draft arrives in your template, ready to edit and sign. The draft saves the typing and the measuring, never the judgment.

Good questions

Questions about radiology over read services

A radiology overread is a formal diagnostic interpretation of an imaging study by a board certified radiologist, performed after someone else has already reviewed the image clinically. In urgent care and chiropractic settings the treating clinician looks at the film to make an immediate disposition decision, and the radiologist then produces the signed report for the record. The overread is the interpretation of record, and it is what supports the professional component of the claim.

It means a contracted radiologist reviews every X-ray your clinic takes and issues the diagnostic report, usually within a contracted turnaround window. Practically it means three things to buy: a provider, a delivery route into your chart, and a policy for what happens when the overread disagrees with the treating clinician. That third one is where the value sits, because discrepancies are found often enough to change follow up and treatment.

Overreads are priced per study, and the rate depends on modality, volume commitment and turnaround. Plain film reads sit at the low end of the teleradiology range and cross sectional studies well above it, with published vendor per study rates across teleradiology generally running from roughly $8 to $80 or more. Volume minimums, after hours premiums and interface build fees are the three costs that most often fail to appear on the rate card you are shown first.

There is no CPT code called overread. The radiologist bills the professional component of the appropriate diagnostic radiology code with modifier 26, while the entity that owns the equipment bills the technical component with modifier TC. When one entity owns both, the global code is billed without a modifier. Which side your clinic bills is a contract question you should settle before the first study, not after.

Chiropractors are licensed to take and interpret radiographs in every US state, so an overread is generally not a legal requirement for chiropractic practice. Clinics buy it anyway for two reasons: a board certified interpretation strengthens the record if a case is reviewed, and some payers and personal injury workflows expect a radiologist report. Several teleradiology groups staff chiropractic radiologists specifically, because the clinical questions differ from an emergency read.

No, and any vendor implying otherwise is describing a product that does not exist. FDA cleared fracture detection software is decision support: it flags regions for a human to review and does not render an interpretation of record. It changes when the clinic gets a second opinion, moving it from hours after discharge to seconds after acquisition, but the radiologist still reviews and signs. Treat AI as compression of the risk window, not as removal of the signature.

Eleven records exist under product code QBS, the FDA classification for computer assisted fracture detection software, and they belong to six companies: AZmed with four, Gleamer with two, Imagen Technologies with two including the 2018 De Novo that created the category, and one each from Milvue, Icometrix and GE Medical Systems. The full list with K numbers and dates is in the table above, pulled from openFDA on September 9, 2026.

An overread is a first formal interpretation by a radiologist of a study someone else has already acted on clinically, and it becomes the report of record. A second opinion, sometimes called a reread or outside read, is a fresh interpretation of a study that already has a radiologist report, usually requested because a patient is transferring care or a finding is being disputed. They are billed and documented differently, so the distinction matters to your coder.

Read every AI radiology question

Explore more