Part 5 of 5 in Circle's Coronary Plaque series. Also read:

Part 1 — How Advanced Plaque Analysis Changes the Clinical Calculus

Part 2 — The IT Infrastructure Behind CCTA Plaque Analysis

Part 3 — The Financial Case for Coronary Plaque Services

Part 4 — Delivering Plaque Analysis Without Disrupting Your Department

You have watched the trajectory. Twelve months ago, the conversation about coronary plaque analysis was happening at conferences. Six months ago, it was happening in your referring cardiologists' offices. Now it is happening in your reading room — which lesions are vulnerable, what the total plaque burden is, whether coronary plaque tells a different story than the stenosis grade.

That part is good news. Your patients are getting better assessments and the evidence base is catching up to the clinical intuition. On January 1, 2026, the financial case caught up too: the AMA retired the Category III plaque codes (0623T–0626T) and replaced them with a single Category I code, CPT 75577, for AI-enabled coronary plaque assessment (ACC Coding Corner). Plaque analysis is no longer an emerging-technology line item. It is a national fee-schedule procedure.

The harder question is operational: is your program set up to deliver it on its own — or to send the studies out and watch most of the reimbursement leave with them? For programs already running CCTA at any meaningful scale, becoming your own plaque lab is more accessible than most assume. It is a workflow choice, not a capital project.

Two Paths into Plaque

There are essentially two adoption models in market today.

The outsourced model. Your site acquires the CCTA, transmits the dataset to an external vendor, the vendor performs the quantitative plaque analysis, and a report comes back. The vendor captures most of the new service revenue. Your readers retain the professional interpretation, but the bulk of the 75577 reimbursement lives outside your service line. Turnaround depends on the vendor's queue, and the data has left your network.

The in-house model. Your readers perform AI-assisted plaque analysis themselves, on data that never leaves your environment, in the same post-processing workspace they already use for CCTA stenosis assessment. Your program bills 75577 directly, with both the technical and professional components captured under your TIN. Turnaround is whatever you make it. The plaque service is part of your program — not a vendor's.

The outsourced model has a real role for low-volume sites without cardiac imaging infrastructure. For everyone else, the math and the workflow case favor in-house — and the threshold for "everyone else" is lower than the early conversations suggested.

The Volume Math: Ten a Month Is the Threshold

If your program can perform around 10 plaque analyses a month — roughly two to three a week — you have the volume to:

• Justify the software investment against the reimbursement those reads represent, now that 75577 sits on a national Category I fee schedule rather than variable Cat III pricing
• Keep your readers in practice on the workflow so quality stays consistent across cases and shifts
• Build a reputation with referring physicians as the program that does the full read in-house, same-day

That is not a high bar. A site running ~50 CCTAs a month — well below the volume of an established cardiac imaging center — and selecting plaque analysis on even a quarter of those is already over the threshold. Confirm the current Medicare allowable for 75577 and your commercial payer coverage before locking the financial model.

What "Your Own Plaque Lab" Actually Bills

As of January 1, 2026, the coding for AI-enabled coronary plaque analysis is materially simpler than it was twelve months ago:

• 75577 (new, Category I): AI-enabled coronary plaque assessment, with interpretation and report by a physician or other qualified health care professional. This single Category I code replaces the deleted Category III codes 0623T–0626T.
• 75574: the underlying CTA of the coronary arteries with contrast — the prerequisite study. Plaque analysis is "post-procedure analysis" of an already-completed CCTA, not a substitute for it.

The shift from Category III to Category I matters. Cat III codes signal emerging technology with variable pricing and inconsistent payer coverage. Category I means a national Medicare physician fee schedule, established RVUs, and a clear path to commercial payer adoption.

A few coverage details worth knowing before you bill:

• Sequencing. The base CCTA must be acquired and interpreted before 75577 is performed. Plaque analysis is a downstream service on already-read imaging.
• Documentation. Claims should support the H&P, presenting symptoms, the CCTA findings that prompted plaque analysis, the clinical indication for the AI study, and the AI plaque report itself. The referring/ordering physician's name and NPI go on the claim.
• Physician sign-off. 75577 explicitly requires interpretation and a report by a physician or other qualified health care professional — the AI does not stand alone.

The point of the in-house model is straightforward: when the plaque analysis happens at your site, the 75577 reimbursement stays at your site. When it is performed off-site by a vendor, the bulk of the new revenue leaves with the data.

The Concrete Checklist: Becoming Your Own Plaque Lab

Six steps to stand the service up. Most programs running CCTA today already have four of them.

1. Confirm clinical readership. At least one cardiologist or radiologist on staff with SCCT Level II or III verification or equivalent cardiac CT credentialing. They are the responsible physician for plaque interpretation and reporting.

2. Validate your CCTA acquisition protocols. Plaque quantification depends on a clean contrast-enhanced CCTA — appropriate heart rate control, contrast timing, gating, and slice thickness. The SCCT acquisition guidelines are the reference standard. If you are already reading CCTAs against guideline targets, this step is already done.

3. Choose the platform: native or bolt-on. Standalone plaque tools require new DICOM routes, new logins, separate report assembly, and a second vendor relationship. A unified post-processing platform runs plaque analysis inside the same workspace your readers already use — same DICOM path, same SSO, same audit trail, same report. cvi42 | Plaque is built on that architecture.

4. Set your reporting template. Plaque-capable reporting should include total plaque volume, low-attenuation plaque (LAP) volume, percent atheroma volume, high-risk plaque feature scoring, and segment-level summaries that referring cardiologists can act on. cvi42's plaque outputs are designed to flow into PowerScribe One and standard structured-reporting workflows for supported environments.

5. Configure billing from day one. Load CPT 75577 in your charge master alongside the underlying CCTA code (75574). Verify your MAC's coverage article, ICD-10 coverage list (R93.1 is the published anchor under WPS Article A59813), modifier requirements, and any prior-authorization rules with your revenue cycle team. Confirm that your reporting workflow captures the documentation the article specifies — the indication, the CCTA findings, the AI plaque report, and the ordering physician's NPI on the claim.

6. Pilot, measure, scale. Run the first 20–30 plaque analyses with a defined reader pool. Track per-study analysis time, report turnaround time, and inter-reader consistency. Once those metrics are stable, open the service to your full reader pool and communicate the new capability to your referring physicians.

That is the sequence. There is no separate lab to build, no new credentialing pathway, no major capital project. You already have the patients, the scanners, the readers, and the PACS. The question is which platform handles the analysis — and whether the technical component stays with your program or leaves it.

Why It Matters Clinically

For your patients, the difference between in-house and outsourced is not abstract. Same-day reads change downstream care decisions. Continuity of interpretation — the same physicians who saw the patient in clinic reading the plaque study — supports better preventive management. And serial follow-up by the same readers on the same AI-assisted algorithm produces the comparable measurements that progression tracking depends on.

Guideline-concordant plaque assessment, delivered locally, is the direction the field is moving. The programs that own that workflow shape the referral patterns in their market — for years, not quarters.

Why It Matters Operationally

For the program, the in-house plaque lab is a service-line decision with three downstream effects:

• CPT code 75577 reimbursement capture. Now that plaque assessment is a Category I service on a national fee schedule, the program that performs the work captures the reimbursement. The bulk of the reimbursement doesn't go to an outsourced service provider.
• Throughput control. Same-day turnaround is a referral-source advantage and a patient-experience advantage.
• Platform leverage. When plaque analysis runs natively in the same software you use for CMR, CT function, and structural heart — one contract, one support relationship, one training pathway — the marginal cost of adding the next CCTA capability is minimal.

A Reasonable Question

If you are reading 10 plaque-eligible CCTAs a month, the question is no longer whether plaque analysis fits in your program. It is whether the program does the work itself or sends it out. The economics, the workflow case, and the clinical case all point the same direction — and the path to standing up an in-house lab is shorter than most teams expect.

cvi42 | Plaque is built for that path. AI-assisted, on-premise, native to the platform your readers already use, and structured to grow with the volume rather than fight it.

Book a cvi42 | Plaque demo to see the in-house lab workflow →