HITInfrastructure

Networking News

GPUs Enhance IT Infrastructure for Healthcare Analytics

Organizations need the advanced processing power of GPUs to embrace real-time healthcare analytics.

GPUs support real-time healthcare analytics

Source: Thinkstock

By Elizabeth O'Dowd

- Healthcare organizations are digitally transforming their IT infrastructure to support healthcare analytics, which requires more advanced processing power. Graphics processing units (GPUs) are becoming more popular as organizations search for tools that allow them to process massive amounts of data.

GPUs have recently gained popularity in healthcare because of precision medicine and value-based care initiatives. The healthcare industry experiences one of the highest volumes of data for analytics over other industries. The high volume of data requires advanced processing power to handle and analyze the data in near real-time.

The central processing unit (CPU) is where most calculations take place and is often referred to as the brain of a computer. GPU is part of the CPU and was originally created for 3D game rendering, but its capabilities extend beyond image rendering.

It was discovered that GPUs are more efficient than CPUs because they can process large blocks of data. GPUs use around 1/10 the processing power of traditional CPU.

GPUs are utilized to visualize data for real-time analytics. This benefits healthcare because it will allow clinicians to get the analytic data they need at the point of care to make a more accurate diagnosis.

Kinetica recently announced its latest GPU-accelerated database that delivers geospatial capabilities and enterprise-grade features to increase performance for better efficiency and to drive real-time, location-based analytics and analytic processing.

“Enterprises have spent the last decade deploying systems to store and manage massive data sets,” 451 Research Senior Analyst of Data Platforms and Analytics James Curtis said in a statement. “Now these enterprises are looking to extract business value from those data sets at real-time speeds and at an affordable price point.”

Earlier this year, Kinetica also announced a partnership with Fuzzy Logix to release a joint solution allowing customers to leverage high performing advanced database analytics. The tool will target the most time sensitive and compute-heavy applications in healthcare, financial services, and retail.

The tool extended Kinetica’s in-database analytic capabilities by adding hundreds of additional GPU accelerated machine learning and predictive analytics algorithms from Fuzzy Logix. The analytic functions utilize Kinetica’s distributed GPU pipeline through its User Defined Functions (UDFs).

The joint solution has a library of algorithms on a SQL-compliant, in-memory database that leverages the GPUs parallelization and powerful real-time analytics capabilities. Use cases include computing portfolio risk management, options and equity pricing, product-based inventory optimization, next-likely purchase, prescribing habits of physicians and care gap analysis.

Real-time data has become more vital to healthcare organizations as patients and clinicians seek more accurate information for diagnosing patients during the initial visit.

survey conducted last year by OpsClarity stated that healthcare providers and life science companies are among the 92 percent of cross-industry organizations that plan to invest in near real-time big data analytics applications as soon as possible.

“The ability to harness the power of real-time data analysis gives businesses a competitive edge in today’s digital economy by enabling them to become more agile and rapidly innovative,” OpsClarity CEO and Co-Founder Dhruv Jain said in a statement.

Vendors and organizations are also working to leverage GPUs for real-time big data analytics by forming groups to create frameworks.

Back in May, Continuum Analytics, H2O.ai, and MapD Technologies formed the GPU Open Analytics Initiative (GOAI) to establish common data frameworks to support data analytics and the use of GPUs across all major industries, including healthcare.

“The data science and analytics communities are rapidly adopting GPU computing for machine learning and deep learning. However, CPU-based systems still handle tasks like subsetting and preprocessing training data, which creates a significant bottleneck,” MapD Technologies CEO and Co-Founder Todd Mostak said in a statement.

An open source framework for GPU processing can make it easier for organizations to implement and deploy their analytics solution and aid in the overall improvement of and development of GPU-based applications.

Fast streaming and near real-time analytics help clinicians make highly informed decisions while treating the patient. This cuts back on costs by reducing the amount of patient visits for the same condition.

Healthcare organizations are in need of faster processing power to offer near real-time analytics for patients and to offer quick and accurate diagnoses for patients.