In the Information Age, Big Data rules the world. The data volumes are growing faster than ever before, and by the year 2020, IDC predicts that about 1.7 megabytes of new information will be created every second for every human being on our planet. However, Big Data is more than just loads of information; it is even more than a mere technology. It is an innovative tool for real-time analysis, planning, forecasting, and finally, instantaneous and efficient decision-making. Big Data represents a new competitive advantage for any company in any industry today.
Big Data Use in Healthcare
In recent years, Big Data has found its way into many different industries—from manufacturing and transportation to insurance and education—but it probably has the greatest potential to make a significant impact in the healthcare industry.
Massive amounts of patient data collected and deeply analyzed by healthcare professionals help to predict epidemics, track the spread of chronic diseases, increase the accuracy of diagnoses, avoid preventable conditions, and considerably reduce treatment costs, paving the way for a more efficient and personal approach to care in the future.
Indeed, Big Data is changing the way hospitals do things. The 2011 McKinsey & Co. study estimated that the effective use of Big Data could save the US healthcare industry $300 million annually; that’s equal to reducing costs by $1000 a year for every man, woman, and child. The analysis of real-time data saved one hospital $850,000 in overtime costs, with improved discharge planning, disease management, quality assurance, and performance reporting, according to an Evariant infographic.
Excited by the many opportunities Big Data offers, we at Auriga have already started working on our own Big Data solutions for the MedTech industry.
Auriga’s High-Load Cardiac Monitoring System
[pullquote]
Tools and Technologies:
- Grid Computing
- Apache Ignite
- Apache Hadoop 2.0
- Apache Kafka
- Apache Hive
- QlikView
[/pullquote]
Nearly half of patient emergency conditions are caused by cardiac activity disorder. However, serious consequences can be avoided if the patient is under continuous observation. By constantly analyzing the patient’s electrocardiogram (ECG) content, it is easy to detect an approaching critical condition at an early stage.
Auriga’s team implemented a high-load distributed scale-out cardiac monitoring system to alert users of approaching dangerous cardiac conditions. Emergency cardiac conditions are recognized with a three-layer neural network. A classical neural network was trained by reverse error propagation using data provided by the open resource, PhysioBank, and showed good results of sensitivity and specificity.
The system is a horizontally scalable service with low configuration requirements of compute nodes in heterogeneous networks. We used Apache Ignite, an open source solution, as a platform to implement the service and Apache Kafka, a distributed queue, as a buffer for reliable and high-intensity transmission of data packets. The Hadoop File System was used for streaming recordings to a persistent store.
The interoperability of the system with different ECG devices is ensured by the HL7 v3 standard. The scalability of the solution is ensured by the Grid Computing paradigm, which allows the creation of a geographically distributed infrastructure incorporating many different types of resources: CPU, ROM and RAM, storage, databases, and networks.
An in-depth statistical analysis of the huge amount of cardiac data coming from the sensors of portable ECG Holter devices, combined with weather data (e.g., atmospheric pressure), enables not only continuous monitoring of hundreds of thousands of patients simultaneously but also prevention of cardiac disorders and makes the system comparable to real-time treatment.
