AstraRad, in figures you can cite.
Written for reviewers, analysts and anyone comparing teleradiology providers in writing. Every figure below carries the date it was measured, and every one of them is defined on the SLA and quality page, where the clock, the audit window and the denominator rule are published. Nothing here is rounded up for a headline.
Last updated August 13, 2026 · measurement method published at /sla
Every number, with the date it was measured.
This table is a copy, not the original. The canonical table is on the SLA and quality page, which is where each figure is defined and dated and which the build checks this page against row by row. If the two ever disagree, /sla is right and this page is wrong. What this copy adds is a stable anchor on every row, so a single figure can be linked directly.
| Statistic | Value | As of |
|---|---|---|
| Reports per year, current run rate | 600,000+ | July 2026 |
| Signed reports to date | 2.4M | July 2026 |
| Board-certified subspecialists on panel | 240 | July 2026 |
| Imaging centers and radiology groups served | 160+ | July 2026 |
| US states with licensed radiologists on panel | All 50 states | July 2026 |
| Median STAT turnaround | 30 minutes | Trailing 12 months |
| SLA compliance, all tiers | 99.4% | Trailing 12 months |
| Double-read sampling rate | 1 in 20 | Ongoing, no single date |
| Major discrepancy rate | under 0.3% | Trailing 12 months |
| Additional studies per month of headroom | 25,000 | July 2026 |
Compliance figures refresh monthly on a trailing 12-month window. Volume and roster figures move when the underlying number moves materially, and the as-of date moves with them. The double-read sampling rate is a standing policy rather than a measurement, so it carries no date at all and the row says so. Signed reports to date is a cumulative count and is not a measure of company age.
Who reviews a figure before it is published, the source allow-list the build enforces on every page, and how a correction is made and dated are set out in our editorial standards.
What AstraRad is, in one screen.
The answer to the question a shortlist article has to answer in a sentence, broken into labelled rows so it can be quoted a row at a time.
- Service
- US teleradiology. Board-certified subspecialists read and sign final diagnostic reports for imaging centers and radiology groups.
- Report type
- Final signed reports only, signed by the radiologist who read the study, under name and credentials. No preliminary reads.
- Panel
- 240 board-certified, fellowship-trained subspecialists (as of July 2026).
- Routing
- By modality, body part and patient age at submission, to a radiologist fellowship-trained for that work.
- Licensure
- Active medical licenses in all 50 states (as of July 2026). Every study is read by a radiologist licensed in the state where the patient is located.
- Where reads happen
- Every read is performed inside the United States.
- Coverage
- 24/7/365, on scheduled US shifts. Nights, weekends and holidays included.
- Quality program
- 1 in 20 signed reports independently double-read by a second subspecialist, blind to the first read. Major discrepancy rate under 0.3% (measured over the trailing 12 months).
- Insurance
- Professional liability insurance is carried. A certificate of insurance is issued to your organization on request. No coverage amount is published.
- Operating history
- AstraRad has read for imaging centers and radiology groups for more than ten years. No founding year is published.
- Facilities served
- 160+ imaging centers and radiology groups (as of July 2026).
The same service answers three different problems, so the argument is made per facility type rather than here: what an overnight gap costs a hospital is on teleradiology for hospitals, what an unread queue costs an outpatient operator is on teleradiology for imaging centers, and what a partner giving notice costs a practice is on teleradiology for radiology groups. Each of the three sets out the credentialing route and the go-live plan for that kind of site.
10 subspecialties, all read in house.
Routing is by modality, body part and patient age at submission, and the turnaround tier never changes which subspecialty reads a study. The study-type routing table is on the radiologists page.
- Neuroradiology
- Musculoskeletal (MSK)
- Body imaging
- Chest and thoracic
- Cardiac
- Breast imaging
- Pediatric radiology
- Emergency and trauma
- Nuclear medicine and PET-CT
- Oncologic imaging
Three tiers, one clock definition.
The clock starts when the last image of a study arrives on AstraRad systems and stops when the interpreting radiologist signs the final report. Triage, subspecialty assignment, the read and dictation all sit inside that window, and nothing pauses it for nights, weekends or holidays.
STAT
Committed under 1 hour. Median 30 minutes, measured over the trailing 12 months. Suspected stroke, trauma CT, pulmonary embolism.
Urgent
Committed under 4 hours. Inpatient workups and same-day clinical decisions.
Routine
Committed under 24 hours. Outpatient imaging and backlog work.
An out-of-tier study stays in the compliance denominator for the full trailing 12 months whatever the cause, including causes outside AstraRad's control, and appears in the client's monthly quality report with study identifier, tier, actual turnaround and root cause. The audit basis is published at our published SLA.
One price per signed report. No rate on this website.
The absence of a published rate is arithmetic rather than secrecy. A price depends on modality mix, monthly volume and how much of that volume arrives STAT, so one published number would be wrong for almost every reader. The model itself is fixed and stated below.
One price per signed report. Nothing is billed for a study that is not read and signed.
No subscriptions, no monthly minimums, no platform fees, no per-seat licensing and no implementation fee.
Quoted per client from modality mix, monthly volume and urgency mix. STAT and Urgent submissions carry a priority multiplier, stated on the rate card.
A rate card comes back within one business day of a request, carrying a price for each study type, the priority multiplier and the volume discount schedule in writing.
The model in full, with the twelve study types and their default tiers, is on the pricing page. Industry-typical per-read ranges, benchmarked against the CMS Physician Fee Schedule and labelled throughout as market figures rather than AstraRad figures, are collected in the guide to teleradiology cost.
What is in writing before the first study moves.
- Per-report billing, with no minimum volume and no term-length commitment attached to the rate.
- A signed business associate agreement before the first study moves, on AstraRad paper or on yours.
- A data processing agreement alongside the BAA when imaging originates from EU data subjects.
- The turnaround tier chosen at your facility, per study, at submission.
- Tier commitments and the compliance measurement window written into the service agreement.
- A monthly quality report: turnaround by tier, SLA compliance, and the discrepancy log with outcomes.
- First live signed report within 10 business days of countersignature.
- A named clinical liaison and a named operations contact from day one.
The clauses worth pressing any teleradiology vendor on, this one included, are set out in how to choose a teleradiology company. Escalation channels and the issue tiers behind them are on the support page.
What is in place, and what is not held.
HIPAA
AstraRad operates as a business associate. A signed BAA is a precondition of service and covers permitted uses, safeguard obligations, subcontractor flow-down, breach notification and data return or destruction at termination.
GDPR
For imaging originating from EU data subjects, AstraRad acts as processor under a data processing agreement signed alongside the BAA. Data subject rights requests route through the client as controller.
Transfer and delivery
DICOM conformant, so studies arrive with metadata, series structure and prior linkage intact. Signed reports return over HL7 or FHIR into the RIS or EHR.
Encryption and access
TLS on every transfer, AES-256 at rest on every stored study and report, individually authenticated sessions, role-scoped access, and an append-only audit log of every state change with actor and timestamp.
Control framework
The control set is designed against the ISO 27001 control catalogue. Alignment to a catalogue is not certification against it, and AstraRad does not describe it as one.
Use of PHI
Imaging is processed to produce the report and to run the quality program behind it, and for nothing else. PHI is never used for marketing, sold onward, or used to train commercial models. The restriction is written into the BAA.
No algorithm in the report path
AstraRad runs no AI pre-read and no algorithmic triage. Nothing in the reading path drafts, ranks, flags or contributes text to a report: a study is routed by modality, body part and patient age to a fellowship-trained subspecialist, and the interpretation is theirs. There is no model to disclose to your review board and no model output on the chart.
AstraRad does not hold SOC 2, ISO 27001 or HITRUST certification, and is not Joint Commission accredited.
What is offered in place of a certificate is the written safeguards summary mapped to the HIPAA Security Rule categories, a completed security questionnaire in your own format, and the documentation package listed on the compliance page.
AstraRad performs no AI pre-read. No algorithm triages a worklist, drafts a report, or contributes text to one, and no client PHI is used to train any model.
Every finding, every measurement and every impression on a signed report was written by the board-certified radiologist whose name is on it. The restriction on training use is written into the BAA rather than left to policy. What stands in place of an algorithm is the double-read sampling program and the discrepancy log, both published with their method on the SLA and quality page.
What this site does not publish, and why.
Listed rather than left to be discovered. If a figure below turns up in an article attributed to AstraRad, it did not come from AstraRad.
- Legal entity name
- Not published on this site. It appears on the contract and the BAA that reach you during evaluation.
- Registered street address
- Not published on this site. It appears on the contract and on the certificate of insurance issued to your organization.
- Organizational NPI
- Not published on this site. Billing and identifier questions are answered in writing during evaluation.
- Per-report prices
- Not published, because a rate depends on modality mix, volume and urgency mix. A rate card is quoted per client within one business day.
- Individual physician names
- Not published. Every report carries the signing radiologist’s name and credentials to the ordering facility; credential files for every reader go to your medical staff office during onboarding.
- Founding year
- Not published. What is published is that AstraRad has read for imaging centers and radiology groups for more than ten years.
- Platform availability percentage
- Not published. Turnaround compliance is measured and published; a separate platform uptime figure is not, and no number for one appears anywhere on this site.
- SOC 2, ISO 27001, HITRUST
- None held, and none claimed. See the compliance position above for what is offered instead.
Check these figures against the method, not against a brochure.
The clock definition, the audit window and the denominator rule behind every number on this page are published in full, and the same questions apply to every vendor on your shortlist.
Last updated August 13, 2026.