Download PDF
Case Studies > AZ Delta: Leveraging Data Analytics for Personalized Medicine

AZ Delta: Leveraging Data Analytics for Personalized Medicine

Technology Category
  • Analytics & Modeling - Machine Learning
  • Infrastructure as a Service (IaaS) - Cloud Computing
Applicable Industries
  • Education
  • Healthcare & Hospitals
Applicable Functions
  • Product Research & Development
Use Cases
  • Predictive Maintenance
  • Tamper Detection
Services
  • Cloud Planning, Design & Implementation Services
  • Training
The Challenge

AZ Delta, one of Belgium’s largest hospitals, was faced with the challenge of leveraging its vast amount of medical data to generate insights and improve healthcare. The data was digitally available but not in a single location and was difficult to work with at scale. The hospital's on-premises IT infrastructure was not capable of powering large scale data analytics. The data was complex and varied, with a single patient’s Electronic Medical Record (EMR) containing thousands of data points. With over 650,000 patient consultations a year, storing, processing, and analyzing this data was a significant challenge. The data needed to be processed and normalized before it could be mined. Manual queries took up to 15 minutes to run due to the scale of the project, limiting the hospital's ability to efficiently utilize the data.

About The Customer

AZ Delta is one of Belgium’s largest hospitals, with 1,400 beds and around 650,000 annual patient consultations. The hospital aims to be a leader in high-quality care, prioritizing innovation and continuous dialogue with patients. As a research-driven hospital, AZ Delta is focused on pushing medicine forward. In 2020, the hospital set up a department for innovation to leverage technology to improve healthcare. The hospital had been using Electronic Medical Records (EMR) for years, generating a vast amount of data with every new patient interaction. However, the hospital faced challenges in storing, processing, and analyzing this data due to its scale and complexity.

The Solution

AZ Delta turned to Google Cloud to overcome these challenges. The hospital partnered with Google Cloud partner ML6 to build a comprehensive medical data analytics platform that could handle the scale and complexity of the data. The platform was built around BigQuery, which replaced the previous on-premises database manager. BigQuery allowed the hospital to work at the scale and speed it needed, reducing query times from 15-20 minutes to 15-20 seconds. The platform began with an on-premises gateway that allowed staff to access the EMR system. The data was then transformed into a parquet file, uploaded to Cloud Storage buckets, and processed using Apache Airflow. The data was then loaded into a BigQuery table, combined with a reference table for consistency, and exported to a final table for analysis. With the data cleaned and normalized, AZ Delta began training machine learning algorithms with TensorFlow to provide physicians with relevant information to aid in decision making.

Operational Impact
  • The implementation of the data analytics platform has significantly improved AZ Delta's ability to leverage its vast amount of medical data. The platform has enabled the hospital to work at the scale and speed it needs, reducing query times and allowing staff to get answers back within seconds. The hospital has been able to clean and code data for 50,000 patients, handling around half a million data points with ease and speed. The use of machine learning algorithms has empowered physicians to find the best course of action faster, improving patient care. The hospital is now planning to extend the analytics platform to other patients and add other kinds of data for greater analytical insight. The implementation of the platform has opened up new possibilities for the hospital, allowing it to do things that were previously beyond the scope of human ability.

Quantitative Benefit
  • Reduced data query run time from 15 minutes to 15 seconds

  • Cleaned and coded data for 50,000 patients, handling around half a million data points with ease and speed

  • Automated much of the query building and optimization, allowing staff to interrogate the data they need and get answers back within seconds

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.