Data engineering for patient analytics platform

2 minutes read

Having big data is not enough. To be effective, healthcare technology companies need to be able to use it effectively. VITech set out to create an innovative patient analytics platform able to analyze 20 billions medical records of 45 millions patients, categorize the population stratification by certain criteria, and provide real-time calculations.

About the client:

World leading healthcare company, a focused leader in health technology determined to build upon our rich heritage by touching billions of lives each year with our innovative technology solutions.

Value delivered

The VITech team designed and developed a system that supports complex queries consisting of over 50 clinical rules and requesting data of different type/level, supports custom combinations of clinical rules and date ranges, helps to categorize population stratification by specific criteria, analyzes the data in the patient database faster and more reliably and helps doctors and clinics to make preventive medicine decisions.

Technologies used:

Java

Scala

Python

TypeScript

Spring-boot

Angular

React

Terraform

Ansible

MySQL

Redis

ElasticSearch

AWS

Kubernetes

Apache Spark/ Airflow/ HBase/ Kafka

Data engineering

Project overview

Challenge

Being among the top five healthcare enterprises in the world, the client had to manage information about more than 60 million patients, which is 7.5 billion medical records. It was quite a challenge for the client to make effective use of all that accumulated data and derive meaningful insight from it.

Solution

To help the client out, VITech built the architecture and implemented a system that allows:

  • Managing 7.5 billion medical records of 60 million patients effectively
  • Supporting intricate search criteria with dozens of clinical rules combined into a complex criteria tree
  • Providing near-real-time calculations for alerts
  • Covering calculations for every alert/measure, every provider/organizational unit and every date range which will result in billions of aggregate numbers and impose prohibitive storage requirements

Results

The VITech team designed and developed a system that:

  • Supports complex queries consisting of over 50 clinical rules and requesting data of different type/level
  • Supports custom combinations of clinical rules and date ranges
  • Has a friendly functionality that also supports real-time updates
  • Includes pluggable extensions with new clinical rules and aggregation types
  • Helps to categorize population stratification by specific criteria
  • Analyzes the data in the patient database faster and more reliably
  • Helps doctors and clinics to make preventive medicine decisions – an essential step toward market leadership in population analysis

Our case studies

prev
next

Our case studies

Leave us a message

You’re in a good company: