AI-Native Revenue Intelligence for Hospital Operations and Claims AccuracyAI-Native Revenue Intelligence for Hospital Operations and Claims Accuracy

The client is a US-based healthcare technology company that helps hospitals and health systems improve safety, efficiency, and accountability through real-time visibility. Their platform combines real-time location systems (RTLS), clinical workflow data, and operational analytics to give providers objective insight into staff activity, asset utilization, and care delivery across their facilities.

Hospitals in the US face a structural challenge: the care that gets delivered is not always the care that gets billed. Despite the widespread adoption of Electronic Health Records, real-world staff activity and equipment usage remain largely under-documented. The gap between what actually happens at the bedside and what ends up in a submitted claim creates revenue leakage, compliance exposure, and audit risk that most providers have no systematic way to address.

Key Challenges

The client's platform captured rich RTLS and clinical workflow data, but lacked the intelligence layer needed to connect that data to billing outcomes and compliance requirements.

• Clinical staff time at the bedside and equipment deployment across patient areas were not tracked or were poorly documented, resulting in revenue leakage in time-based billing models.
• RTLS systems could not reliably distinguish between passive asset presence and active clinical use, making billing justification difficult and audit defense challenging.
• Without timestamped, objective interaction records, providers faced significant exposure under HIPAA and CMS scrutiny, and internal audits depended heavily on manual logging.
• Under bundled payment models like BPCI and CJR, high-acuity cases risked underpayment when coders lacked the operational telemetry to justify DRG upgrades.

The client needed a partner who understood both the operational complexity of hospital billing workflows and the engineering depth required to build a production-grade intelligence layer atop a live RTLS and EHR data environment. Coditas brought together AI platform engineering, healthcare data integration experience, and knowledge of claims reconciliation, compliance requirements, and clinical documentation workflows.

We applied our Agentic System capabilities to design an AI-driven reconciliation agent that could reason over structured RTLS telemetry, clinical notes, and billing data simultaneously to surface revenue gaps and compliance risks that no manual process could reliably detect.

Coditas engineered a unified intelligence platform that integrated RTLS telemetry, ADT feeds, and claims data into a single governed system, with an AI reconciliation layer on top. The foundation was a real-time resource utilization engine that used custom geofencing logic to map staff and asset locations against patient occupancy, converting entry and exit events into timestamped logs of every physician, nurse, and tagged piece of equipment. Accelerometer data was used to distinguish passive presence from active clinical use, providing the system with a reliable basis for billing validation.

We built an AI billing reconciliation agent on top of that data platform to evaluate discrepancies between actual care delivery and submitted claims. The agent cross-referenced RTLS logs against CPT and DRG codes, flagged mismatches, and used AI to match free-text clinical notes with presence data to validate care intent and surface documentation gaps.

The system also extended into staff utilization modeling, bundled payment integrity checks, and narrative-aware auditing, supporting both revenue capture and broader operational transparency. Deployment followed a phased rollout from single-unit pilots to enterprise-grade integration, with full HIPAA compliance, RBAC, audit trails, and end-to-end encryption throughout.

Technologies

RTLS, ADT/EHR via HL7/FHIR, Kafka, Python ETL, Redox, ML-based pattern recognition, GenAI narrative extraction, time-series databases, Kubernetes, S3, BigQuery, React, HIPAA-compliant architecture, RBAC, audit logging

The engagement transformed raw operational telemetry into structured financial and compliance intelligence across pilot units.

• $1.2M+ in previously unclaimed revenue identified across pilot units through flagged missed CPT codes, underbilled nursing time, and overlooked equipment charges
• 28% reduction in audit-related claim denials through timestamped logs and narrative validation
• 3x faster discrepancy resolution cycle, from 12 days to under 4
• 15% improvement in DRG assignment accuracy through bundled care analysis of high-acuity encounters
• Staff utilization benchmarks established across 10+ units, uncovering overutilization and under-allocation patterns
• Platform architected for multi-hospital scale-up without disrupting existing clinical workflows

Data is one of the leading culprits behind revenue leakage in hospital billing. The gap between what care teams deliver and what billing systems record is structural, and no amount of manual review closes it reliably at scale.

Coditas helped the client turn their existing RTLS and clinical data infrastructure into an active revenue intelligence layer, one that does not just report on what happened, but reconciles it against what was billed, flags what was missed, and builds the documentation trail that makes claims defensible. That is what it means to bring AI into a healthcare revenue workflow with the right engineering depth and the right understanding of the operational context.