Loading…
ARKA-CLIN: Clinical Decision Support
Evidence-based imaging appropriateness evaluation powered by the ARKA Imaging Intelligence Engine (AIIE).
What it does
ARKA-CLIN provides real-time clinical decision support for imaging appropriateness. Enter a clinical scenario or choose a demo case to receive an evidence-based appropriateness score (1–9), transparent factor breakdown (SHAP-style), alternatives, and links to peer-reviewed evidence.
Who it's for
Ordering physicians, radiologists, and healthcare organizations who want to align imaging orders with guidelines (e.g., ACR, RAND/UCLA) and reduce unnecessary imaging while supporting appropriate studies.
How it fits in the ecosystem
ARKA-CLIN is the core CDS module in the ARKA suite. It feeds into ARKA-ED (emergency department workflow) and ARKA-INS (prior auth and payer alignment). Together they support ordering, workflow, and reimbursement with one evidence base.
Loading demo…
How ARKA Works
ARKA (Imaging Intelligence Engine) helps healthcare providers decide when imaging is appropriate. You enter a clinical scenario; ARKA applies evidence-based rules and returns a score with a clear explanation. Below is the high-level flow and the FDA Non-Device CDS criteria ARKA satisfies.
Data Input
Clinical scenario (complaint, duration, red flags, proposed imaging) is entered by the provider or pulled from the EHR.
AIIE Processing
AIIE applies weighted clinical factors from peer-reviewed literature to compute an appropriateness score (1-9) and explanation.
Recommendation Output
You see a score, category (Usually Appropriate / May Be Appropriate / Usually Not Appropriate), and factor-level explanation.
FDA Non-Device CDS Criteria
Data Input Criterion
Clinical scenario and proposed imaging are the inputs; no hidden data drives the recommendation.
Medical Information Criterion
Recommendations are based on medical information (evidence, guidelines, clinical factors) you provide.
HCP Recommendations Criterion
Output is intended to support, not replace, healthcare professional decision-making.
Independent Review Criterion
Users can independently verify the basis (literature, factor weights, explanation) for each recommendation.
AIIE uses a proprietary scoring methodology based on RAND/UCLA appropriateness methods and peer-reviewed literature. Every recommendation shows exactly WHY it was made.
The AIIE Scoring Methodology
- 1
Start with Baseline Score
Every evaluation starts at 5.0 (neutral). This represents "we don't know yet" — neither appropriate nor inappropriate until we assess the clinical factors.
- 2
Apply Weighted Clinical Factors
Each clinical factor (red flags, duration, age, prior workup) has an evidence-based weight derived from peer-reviewed literature. Factors either increase (+) or decrease (-) the appropriateness score. Example: Cancer history → +3.0 points (Deyo & Diehl, JAMA 1988)
- 3
Calculate Final Score (1-9)
All factor contributions sum to produce a final score, capped between 1 and 9. 7-9: Usually Appropriate (green). 4-6: May Be Appropriate (yellow). 1-3: Usually Not Appropriate (red).
- 4
Show SHAP-Style Explanation
Unlike black-box systems, AIIE shows exactly which factors contributed to the score and by how much. This satisfies FDA Non-Device CDS Criterion 4 (independent review).