Overview

For physicians, understanding a complex patient’s history can mean going through years of clinical information before a single appointment. Mediora Clinical Partners, a US-based multi-specialty outpatient network, faced exactly this challenge. Patient information was spread across visit notes, lab reports, specialist referrals, and discharge summaries, often stored in different formats and written by different clinicians. Before a complex patient visit, physicians and care coordinators had to manually review a dozen or more documents to connect this information and understand the patient’s complete clinical history. This took valuable time and made it harder to quickly identify relevant details. Bacancy Technology addressed this challenge by building a generative AI-powered clinical data analysis solution that brings a patient’s longitudinal medical history together and generates a citation-linked pre-visit summary.

Technologies Used

python
PostgreSQL with pgvecto
LangGraph
GPT-4
FHIR APIs
Docker

Project Highlights

checkmark

AI-generated pre-visit summaries built from a patient's full longitudinal clinical history

checkmark

Negation-aware clinical language processing to avoid misreading patient history

checkmark

Every AI-generated statement links back to its exact source note for physician verification

checkmark

Physician review and edit workflow before any summary is saved to the patient chart

The Challenges

1

Before a complex visit, care teams had to manually review years of records across multiple specialists. The challenge was not simply finding a particular note. It was connecting information from different points in a patient's medical history to understand the complete picture.

2

Patient information came from multiple EHR modules and external specialist systems. Records included clinical notes, scanned referral letters, discharge summaries, and laboratory data. Bringing these sources together required significant data processing and normalization.

3

Traditional keyword search could find symptom mentions but could not reliably understand the clinical context. For example, “denies chest pain” should not be interpreted as evidence of chest pain. This meant physicians still had to manually verify search results.

4

Any AI solution handling clinical data had to meet strict requirements for PHI protection, auditability, and healthcare compliance. More importantly, the system had to support physicians rather than make clinical decisions on their behalf.

Solutions by Bacancy

1

Bacancy Technology’s team began with discovery sessions involving physicians and care coordinators to understand where the most time was being spent. The main challenge was connecting information from multiple sources to understand the patient’s overall history. Based on this finding, the team designed the solution to synthesize the patient’s clinical story rather than provide another document search tool.

2

Through our generative AI expertise we built an ingestion pipeline that collects structured and unstructured clinical data, including visit notes, lab PDFs, referral letters, and discharge summaries, directly from the EHR. The pipeline normalizes this information into a consistent format and applies clinical language processing to handle negation correctly. This ensures denied symptoms are not treated as reported symptoms.

3

Our RAG development team indexed each patient’s longitudinal record using encounter-aware chunking instead of arbitrary text splitting. The retrieval layer then identifies the most relevant clinical history based on the specific visit. The retrieved information is passed to a grounded language model, which generates a pre-visit summary with citations linking each statement back to the original clinical documentation.

4

The system does not make clinical decisions. Each AI-generated summary includes citations to the source documentation so physicians can verify the information. Before the summary is added to the patient chart, a physician must review and can edit it. PHI access controls, encryption, and audit logging are applied throughout the pipeline to support healthcare compliance requirements.

Core Features

checkmark

AI-generated pre-visit patient summaries pulling from years of scattered clinical notes

checkmark

Negation-aware clinical language processing to avoid misreading patient history

checkmark

Citation-linked summaries so every AI-generated statement traces back to its source note

checkmark

Physician review and edit workflow required before any summary reaches the patient chart

No. of Resource

06

No. of Resource

Time Frame

Jan 2026 - Apr 2026

Time Frame

Project Snapshot

We Integrated GenAI Into Clinical Data Analysis for Mediora Clinical Partners

Outcomes

Pre-visit chart review time dropped from roughly 18 minutes to under 5 minutes for complex patients.

Physicians reported catching previously overlooked history details in a meaningfully higher share of complex cases.

Physician adoption reached most of the target care team within the first quarter after rollout.

Follow-up questions between visits caused by missing context dropped noticeably across the care team.

Care coordinators redirected the reclaimed time toward outreach for higher-need patients instead of manual chart review.

Care coordinators redirected the reclaimed time toward higher-priority patient follow-ups instead of manual chart review.

Technical Stack

Programming Language Python
Workflow Orchestration LangGraph
Language Model GPT-4
Retrieval PostgreSQLEmbedding Search
Clinical Data Integration FHIR APIsEpic EHR Integration
Validation Rule-Based Clinical Validation Layer
Database PostgreSQLRedis
Deployment DockerKubernetes

Experience With Bacancy

Optimize Your E-Commerce Workflow for Maximum Efficiency

iconWatch the video

Fintech Platform Scalability & Growth Optimization

iconWatch the video

Legacy Code Upgrades for Modern User Experience and Enhanced Data Analytics

iconWatch the video

Empowering Retail Marketing With Seamless User Experiences

iconWatch the video

Migrating To Laravel From Core PHP With Twilio Integration

iconWatch the video

Upgrading PHP Laravel version while scaling the Admin Dashboard

iconWatch the video

Optimizing CRM Performance With Data Management and Salesforce Integration

iconWatch the video

AI Solution Engine for Improved Performance and Error Handling

iconWatch the video

Enhancing Managed IT Services With Streamline Automation Process

iconWatch the video

SEO and Bulk Data Upload Solutions For An E-Commerce Healthcare

iconWatch the video

Elevate Your E-Commerce with Scalable UI and Enhanced Calculators

iconWatch the video

Simplifying Annuity Rate Management for Complex Calculations

iconWatch the video

Addressing Null Safety, Code Structure Issues, and Manual Deployment Issues

iconWatch the video

Optimize Task Management With Advanced Tracking

iconWatch the video

Revolutionize Applicant Tracking: Advanced Image Optimization Techniques

iconWatch the video

Cost-Effective Media Management Solution and Resolving Data Upload Issues

iconWatch the video

Enhance Your Management Service with User-Centric Innovations

iconWatch the video

2500+ Projects Experienced Innovation with Bacancy!

Get access to an experienced team of developers and engineers from Bacancy, handpicked to ace your goals. Kickstart within 48 hours, no-risk trial.

Book a 30 min call

14+

Years of Business Experience

1458+

Happy Customers

12+

Countries with Happy Customers

1050+

Agile Enabled Employees