HP Autonomy unveils healthcare analytics system for textual data

HP has unveiled a new healthcare analytics product that, the company says, will enable medical staff to interrogate text as well as data in relational databases.

Alan Stein, vice-president of medical technology at HP Autonomy, said the platform, which is built on the company's Intelligent Data Operating Layer (Idol) software, provides "a way to connect the users more directly with their data." The supplier has wrapped healthcare-specific online information around the tool.

The platform is designed to reduce costs and improve medical care, he said. Clinicians, and other healthcare workers, can use the tool without having to know about data types or specialised data management techniques. "They just need their domain knowledge."

The company said the "platform provides healthcare workers with insights that reduce errors and unnecessary treatments, impact how delivery and treatment affects outcomes, and enable preventive measures that decrease the rate of avoidable diseases. The platform also helps to reduce costs by identifying and eliminating waste, streamlining workflow and elevating staff productivity by removing the skill barrier for accessing clinical intelligence."

Stein is a physician based in the U.S., though an HP employee. The product, called HP Healthcare Analytics, was developed in collaboration with Stanford Children's Health and Lucile Packard Children's Hospital Stanford, which specialises in paediatric and obstetric medicine, based in Palo Alto.

The supplier and Stanford Children's Health have worked together for more than a year to develop software that enables hospitals to extract information from structured and unstructured clinical data.

Stein said the system will be rolled out in the Americas first, but will come to the UK, where Idol is in place in NHS hospitals.

The software integrates with healthcare information systems, data warehouses, and other related clinical, operational and financial data sources. It brings in some 400 data connectors and includes support of more than 150 data types.

It also integrates the Systematised Nomenclature of Medicine (Snomed) taxonomy system, which contains more than 350,000 healthcare concepts and more than two million clinical concepts. It also incorporates the International Classification for Disease (ICD) coding system.

Stein concluded: "Until now, healthcare providers primarily used traditional analytics tools focused on the structured data within their healthcare systems, failing to analyse up to 90% of their information. This helps organisations unlock information in everything from clinical notes to laboratory results.

"With this platform you get a comprehensive view of the data, regardless of where it resides or whether it is structured or unstructured data. Everyone is talking about big data, but this gives real-world applicability, and it is very tangible."