Medical Research
Big data brings with it significant challenges to medical research. Large healthcare and life science organizations need to mine billions of data points representing millions of unique patients. This can be a time consuming, costly, and frustrating process. Teams of researchers may spend weeks formulating complex questions and queries necessary for accurate cohort identification and analysis. Researchers may have to wait hours or days, in some cases weeks, before they have access to their results. Time and costs matter in the medical research world, where resources and budgets are limited. Affordable scalability, combined with outstanding abilities to ingest incompatible data and then quickly identify and present results, is a must for the medical research professional.
With IMAT Solutions, researchers can produce and run sophisticated queries and reports in real time.
IMAT: Designed for Big Data Problems
IMAT is a new, disruptive technology to the field of data management and decision support. Our design isn’t based on an early- or late-binding approach. It is a unique architecture, offering a flexible, non-rigid, pattern-based, indexing system that delivers disruptive speed, accepts data from incompatible sources, delivers superior accuracy during data normalization and validation processes, and provides a user interface that non-technical medical professionals can learn to use.
Through the six-step IMAT data ingestion process – Connect, Collect, Validate, Normalize, Index, and Codify – all data, no matter the source, the format or data type can be aggregated into one common data store. This includes narrative information captured as unstructured data. The IMAT platform can capture, standardize and make available unstructured and structured data for real-time search and reporting.
Processing time for complex queries across millions or billions of records has been compressed for IMAT users. In performance and accuracy benchmarking, the IMAT system identified complex cohort sets in 2 to 20 seconds, across 2 billion records, representing 19 million patients from 20 data sources. Competing systems took 20 hours to execute the same queries. Imagine running and reviewing your cohort analysis results during your lunch hour.
Other Benefits Include:
Accuracy – The IMAT platform ensures that all data is indexed at its most granular level. Working with our clients, they pre-define a set of concept recognition, normalization, validation and code mapping rules, which are run against the data during intake to ensure a high degree of data accuracy. During a monitored ETL (extract, transform, load) process, the IMAT platform was able to take in 2,000 records per second, which included running 15 validation rules against each record, while running on a single server. Accurate data leads to more accurate analysis and decision making.
Comprehensive – Leveraging a natural language processor and a core indexing engine, the IMAT platform is able to work with both the structured and previously untapped unstructured data found within all healthcare organizations.
Big Data – Billions of rows of data can be stored on a single server running the IMAT platform. This capability dramatically reduces the strain placed on an organization in terms of budget and efficient resource utilization. Even more impressive, IMAT utilizes approximately 10% of CPU capacity when running even the most complex queries and reports associated with medical research.
IMAT provides medical researchers with disruptive technology. Complex query and deep data mining speed, accuracy, and scalability deliver unparalleled insight into critical data with unmatched efficiencies – propelling healthcare and life sciences research into the next generation.