For the CFO
Recover revenue you already earned
ARKA documents medical necessity at the point of order, so claims go out clean. Modeled recovery: ~$3.5M/yr in avoidable imaging denials for a mid-sized system.*
Loading…
ARKA Clinical Decision Support is designed to meet all four criteria for Non-Device CDS under FD&C Act §520(o)(1)(E) and FDA's January 2026 final guidance on Clinical Decision Support Software. Recommendations support, not replace, the clinician's judgment. Every recommendation is anchored in a published guideline or peer-reviewed source, with the basis available for independent review. CLIN emphasizes imaging appropriateness at order entry.
This recommendation is intended to support, not replace, clinical judgment. It is generated by ARKA, software designed to meet the four criteria for Non-Device Clinical Decision Support under FD&C Act §520(o)(1)(E) and FDA's final guidance on Clinical Decision Support Software (January 2026). The clinician is responsible for the final decision.
One appropriateness engine · payer + provider
ARKA scores every advanced-imaging order against published guidelines in real time — auto-approving what's clearly appropriate and clearing medically-necessary studies through prior authorization with the same transparent, cited math the clinician sees. Fewer denials, fewer peer-to-peers, no black box.
Non-Device CDS · No FDA 510(k) · CMS-0057-F ready · the ordering clinician keeps the final call.
Built for
Evidence of rigor
The same engine scores cardiac, oncologic, musculoskeletal, neurologic, and pediatric imaging the moment it's ordered — escalating what's warranted and steering away from what isn't, every call tied to a published guideline.
AIIE · order-select
Real-timeImaging order
Multimodality appropriate use criteria support CMR for tissue characterization here.
Cardiac: Cardiac MRI — suspected infiltrative cardiomyopathy. Disposition: Appropriate — proceed, cited to ACC/AHA/ASNC multimodality AUC.
Every disposition is cited to a published guideline, with the basis available for independent review — the ordering clinician keeps the final call. See the evidence library for the guidelines behind every score.
Guideline names are cited for source identification only. All marks are the property of their respective owners; citation does not imply review, endorsement, or partnership.
Revenue recovery
When appropriateness is documented at the moment of order — on the clinician side and the payer side — imaging teams stop losing revenue to avoidable denials.
Imaging prior-auth denials run 20–40% — and ~86% of them were avoidable. ARKA is one engine that runs on both sides of the prior-auth wall — the doctor's and the payer's — documenting the clinical justification at the moment the order isplaced via CDS Hooks and SMART on FHIR — designed for Epic, Oracle Health (Cerner), and athenahealth after customer-specific validation — in under 800ms, without adding a single click. Clean claims go out the first time.
The clock is real
Sign by Q1 2026 to go live by Jan 2027 — health-system IT procurement typically runs 12–18 months.
161 days until CMS-0057-F — deadline January 1, 2027
161 days until CMS-0057-F — deadline January 1, 2027
Sign (by Q1 2026)
Sign by Q1 2026 to go live by Jan 2027 — health-system IT procurement typically runs 12–18 months.
Integrate via CDS Hooks (2–4 weeks)
CDS Hooks and SMART on FHIR — designed for Epic, Oracle Health (Cerner), and athenahealth after customer-specific validation; no new app, no second login.
Live before Jan 2027 (Da Vinci PAS today)
Live Da Vinci CRD/DTR/PAS endpoints in production now (discovery at /.well-known/cds-services).
The denial loop
Every imaging team knows this loop. ARKA breaks it.
The scan was justified. The auth bounced six weeks later over one line of documentation no one asked for — and now it's a write-off. Prior-auth denial rates on advanced imaging run 20–40%.
Roughly half of denied claims are never reworked — the appeal costs more staff time than the claim is worth. So the hospital eats earned revenue, and a nurse loses hours to a payer hold line.
Imaging is one of the highest-margin service lines you run. Every order stuck waiting on auth is a scan not completed this month — slow approvals are slow revenue.
The last 'AI' tool flagged everything, interrupted every order, and added five clicks. The team clicked past it until it was switched off. Sound familiar?
~86% of imaging denials are avoidable. The fix has to happen where the order is placed — not in the billing office six weeks later.
Modeled economics
Modeled for a regional hospital group running ~120,000 advanced imaging studies a year — the conservative case.
recovered/yr in avoidable imaging denials*
Modeledof imaging denials are avoidable
Measuredto score an order, in-flow, no extra click
Measuredof orders auto-clear and never hit a queue
ModeledOne guideline-redirected order: ~$1,180 avoided.* Scale that across 120,000 studies and the conservative recovery is ~$3.5M/yr — plus roughly ~$0.5M in faster throughput on your highest-margin line.
Priced at ~$0.30–$0.50 PMPM — a modeled ~2.3× first-year return.*
*Modeled, conservative estimate using published CAQH, KFF, MGMA, AMA, and ACR figures; aggressive case ~1.5×. ARKA is Non-Device CDS — figures are decision-support economics, not a guarantee of outcomes.
Your numbers
Move the sliders to model your system — conservative figures, sourced assumptions on demand.
Value proposition
Three reasons a CFO signs — and a physician never notices.
For the CFO
ARKA documents medical necessity at the point of order, so claims go out clean. Modeled recovery: ~$3.5M/yr in avoidable imaging denials for a mid-sized system.*
For compliance
Non-Device CDS under §520(o)(1)(E) — no FDA 510(k). ARKA processes structured FHIR in-context as your Business Associate; PHI is transient in the CDS Hooks / SMART request path, identifiers are hashed before persistence, and federated analytics returns aggregate query results — never row-level patient records. A CMS-0057-F Da Vinci PAS endpoint shipping in production today.
For physicians
Non-blocking, in-flow, under 800ms via CDS Hooks and SMART on FHIR — designed for Epic, Oracle Health (Cerner), and athenahealth after customer-specific validation. Silent unless a guideline fires. 35–40% of orders auto-clear and never enter a queue. No new screen, ever.
*Modeled estimate; sourced ranges in ARKA's revenue model. ARKA is decision support — the ordering clinician retains the final decision.
For your team
Five buyers, five briefings — select your role to see what matters to you.
CFO / VP of Finance
Denied imaging claims are direct write-offs — ~$4.8M at risk on 20,000 orders at a 20% denial rate.
AIIE Technology
A next-generation clinical decision support system that goes beyond static, out-of-workflow criteria tables — whether society-published, national, or payer-proprietary — patient-specific, in-workflow, and evidence-linked.
Static criteria tables — out-of-workflow lookup
Patient-specific ML at the point of order
0.0%
Voluntary static criteria CDS adoption (published literature)
0.0%
AIIE guideline concordance on signed-off scenarios (synthetic; real-world validation in progress)
0 extra clicks
In-workflow at order entry via CDS Hooks
Watch the same imaging order evaluated the traditional way versus inline with AIIE.
Criteria-table lookup
Lookup happens outside the workflow — generic, not patient-specific.
Criteria-table lookup step 1 of 3.
AIIE path: order pending. Score pending.
AIIE path
CDS Hooks · order-selectAt order entry, AIIE evaluates structured FHIR context against six weighted clinical factors — indication strength, prior imaging, red flags, guideline alignment, patient risk, and radiation burden. Each 1–9 score includes SHAP-attributed factors linked to first-party citations in the Evidence Library.
How we're different
Legacy prior-auth and radiology benefit managers review the order after it's submitted — and deny after the fact. ARKA surfaces the same cited, guideline-anchored appropriateness at the point of order, inside the EHR, in real time — so the right study clears the first time.
Compliance & workflow
Measured today: ~87.5% guideline concordance on a held-out, human-signed-off scenario cohort — not a clinical-outcome claim. Real-world outcome validation is pending pilot data. We publish concordance as concordance and never equate synthetic self-consistency with clinical validity.
Platform
ARKA is a single decision engine with a shared knowledge base. Insights from each phase inform and improve the others — the same math runs on both sides of the prior-auth wall, from the clinician's order to the payer's review. Explore the four phases below.
At ~$0.30–$0.50 PMPM, a modeled ~2.3× first-year return — and the same engine reaches into the ~$10B appropriateness layer of American medicine.
Modeled estimate; full sourced ranges in the ROI breakdown.
Outcomes
Clinician perspectives from the Stormont Vail evaluation are being finalized with the institution for publication.
Get started
ARKA recovers revenue you're already losing to denials, speeds up your highest-margin service line, and reduces the admin burden doing it — without changing how your physicians order.
remARKAbly precise — and remarkably profitable.