AgentHack submission type
Enterprise Agents
Name
Alishba Shahid
Team name
HealOps
Team members
@Nick Noel
How many agents do you use
Multiple agents
Industry category in which use case would best fit in (Select up to 2 industries)
Healthcare Pharma
Insurance
Complexity level
Advanced
Summary (abstract)
MediAuthPro – Transforming Prior Authorization with AI & Automation
MediAuthPro is an AI-powered automation agent that revolutionizes the prior authorization process in healthcare. It enables doctors to upload patient data or voice notes in realtime, automatically checks if Pre Auth is required, generates clinical justifications using LLMs, and submits them for insurance approval and maintains a dashboard—all within seconds.
Detailed problem statement
Prior authorization in healthcare is a time-consuming and manual process that delays patient care. Physicians must gather records, draft justifications, and manually submit them through complex insurance portals—often leading to errors, rejections, and delays. This inefficiency creates significant administrative burden, burnout among providers, and increased denial rates from payers. There’s a need for an intelligent, integrated solution that automates this multi-step workflow while ensuring clinical and compliance accuracy.
Detailed solution
MediAuthPro is a UiPath-powered agent that automates the end-to-end prior authorization process:
Data Ingestion: Patient records are fetched from EMRs, PDFs, or real-time FHIR feeds.
Optional Voice Capture: Physicians can dictate notes, which are transcribed and used as input.
AI Processing: An embedded LLM reviews the clinical data and checks if pre auth is required if yes, it generates a payer-specific justification.
Submission & Tracking: The agent logs into payer portals, submits requests, and tracks approval status.
This solution was built using UiPath Studio, Orchestrator, Document Understanding, Apps, and Agent Builder, delivering seamless automation with audit logs, retry mechanisms, and interactive dashboards for clinicians and admins.
Demo Video
Expected impact of this automation
90% Reduction in Manual Effort: Eliminates the need for physicians or admins to draft justifications or navigate insurer portals.
Faster Patient Care: Accelerates prior auth approvals from days to minutes, enabling timely treatment decisions.
Improved Approval Rates: Standardized, payer-aligned AI-generated justifications reduce denials and resubmissions.
Reduced Physician Burnout: Allows clinicians to focus on patient care instead of administrative tasks.
Scalable Across Specialties: The agent is adaptable for any clinical procedure or specialty (e.g., radiology, cardiology, surgery).
Future-ready: Built to integrate with FHIR APIs, voice dictation tools, and LLM advancements for long-term scalability.
UiPath products used (select up to 4 items)
UiPath AI Center
UiPath Agent Builder
UiPath Apps
UiPath Studio
Automation Applications
Insurance Portal
Integration with external technologies
Open AI
Agentic solution architecture (file size up to 4 MB)
Sample inputs and outputs for solution execution
Inputs:
Patient Identification:
Patient Name: e.g., “John Doe”
Optional File Upload: PDF of EMR or medical case summary
Procedure Details:
Procedure Name: e.g., “Knee Arthroscopy”, “CT Scan”, etc.
Diagnosis (ICD-10 Code): e.g., M17.9 (Osteoarthritis, knee)
Insurance Information:
Payer Name: e.g., Aetna, Cigna, UnitedHealthcare (selected from dropdown)
Doctor Information:
Physician Name: e.g., Dr. Jean Mathews
NPI or Provider ID: optional
Voice Note (Future Enhancement):
MP3 Audio File: Doctor’s spoken clinical note for transcription
Processed Internally via Agent:
PDF parsed using UiPath Document Understanding
EMR data converted to structured format
GPT-powered LLM generates clinical justification based on diagnosis, procedure, and payer
Prior Authorization submitted to insurance portal
Outputs:
AI-Generated Justification:
A medically sound, payer-specific clinical justification text
Submission Confirmation:
Status: Approved / Denied / Pending
Tracking Number / Reference ID: e.g., “PA12345678”
Submission Date: e.g., “2025-06-27”
Dashboard Metrics:
Total Requests, Success Rate, Average Approval Time, etc.
Status Table (Query Page):
Displays a list of past submissions with:
Patient Name
Procedure
Status
Request Date
Other resources
Epic EMR, Cerner, SAP (for hospital ERP), Salesforce Health Cloud
These are possible sources of Electronic Medical Records (EMR) or referral flows that the agent can extract or interface with.