imx.ai

What It Is

IMX-Ray™ Fraud Detection DSS is a fraud detection decision support system from IMX Data built for U.S. Medicare/Medicaid use cases. Focused on healthcare program integrity, it uses statistical analysis, machine learning, and real-time data integration to identify anomalous billing, high-risk providers, and potential fraud, waste, and abuse (FWA). The page states that its data foundation includes 100 billion+ medical claims and adjudication events, 300 million+ patient records, and integrations with data sources such as CMS, HHS, OIG, and NPPES.

Core Capabilities

The system is designed more like an “investigation workbench” than a general-purpose AI assistant. It offers a command center, Provider Search, risk watchlists, geographic risk heatmaps, network/collusion analysis, prescription monitoring, and PDF report export. Its detection methods include 30 statistical tests covering abnormal billing volume, cost outliers, single-code concentration, Benford’s Law, billing spikes, OIG exclusion list matching, controlled substance/opioid prescribing, as well as ERA/835 payment adjudication signals such as denials, resubmissions, payment velocity, DRG upcoding, reversals, and corrections. Each provider receives a 0–100 risk score and is categorized as Critical, High, Elevated, or Standard. Notably, the page explicitly states that the risk score is only an investigative lead, not a determination of fraud, and still requires human review.

Pricing and Deployment

The main content does not disclose pricing, plans, free trials, or enterprise contract costs. It only provides an approximately 10-minute, 16-step guided demo and an ROI calculator. Deployment options are relatively comprehensive, including multi-tenant SaaS, dedicated SaaS, on-premises deployment, AWS GovCloud, or Azure Government. On the security side, it lists SOC 2 Type II, AES-256 encryption, and NIST 800-53 alignment, while FedRAMP is still marked as on the roadmap/in progress.

Pros and Cons

Its strengths are a tightly focused use case, rich data sources, a detailed detection framework, explainable outputs, and support for investigative reports and cross-state/cross-payer analysis. The limitations are also clear: in the demo data, multiple flags are 0 and synthetic data is included, making it impossible to assess real-world recall and false-positive rates; APIs, SDKs, pricing, and implementation timelines are not disclosed; and it is primarily tailored to the U.S. healthcare reimbursement system, with no information on Chinese-language support or China-specific compliance.

Who It’s For and Access from China

It is better suited to U.S. federal/state Medicaid agencies, health plans, healthcare anti-fraud teams, and large research institutions. It is not a fit for individual users or general enterprise AI office scenarios. Access from China is unknown; even if it is accessible, its core data, procurement/payment processes, and compliance context are highly U.S.-centric. For alternatives in China, organizations would typically look for local healthcare cost-control systems, commercial insurance anti-fraud platforms, knowledge-graph-based risk control tools, or big data audit platforms rather than adopting this system directly.