Royal Philips launches IntelliSpace to support deployment of AI in radiology

22 November 2018 (Last Updated November 23rd, 2018 04:54)

Royal Philips has introduced a comprehensive and open platform called IntelliSpace Discovery 3.0 to enable development of artificial intelligence assets and their deployment in radiology to help teams carry out clinical and translational research.

Royal Philips launches IntelliSpace to support deployment of AI in radiology
Philips IntelliSpace Discovery 3.0. Credits Philips

Royal Philips has introduced a comprehensive and open platform called IntelliSpace Discovery 3.0 to enable development of artificial intelligence assets and their deployment in radiology to help teams carry out clinical and translational research.

Launched ahead of 2018 Radiological Society of North America (RSNA), IntelliSpace Discovery is already in use at more than 50 hospitals and academic institutions. It is being used in the development of radiology applications for rendering, segmentation, and quantification.

The open platform, powered by Philips HealthSuite, enables radiologists to have thorough data analytics in medical imaging.

It is currently used for research and clinical validation, and the tools/applications can be subsequently deployed into the radiology workflow onto the Philips IntelliSpace Portal.

Although AI solutions offer the potential to boost patient care and care delivery, there can be difficulties in launching them into healthcare clinical practice. Problems can occur in the means of collection and preparation of quality data, the approaches for training and validating the tools, and avoiding disruption during AI deployment.

Following a people-centered approach called adaptive intelligence, the firm has combined AI and other technologies with knowledge of the clinical and operational context to build integrated solutions intended to meet the requirements of healthcare providers.

The launch of this platform enables radiologists to have tools and applications to collect, normalise and anonymise data. This data enables radiologists to visualise and then interpret to ‘train’ and validate deep learning algorithms.

Radiologists can easily deploy the algorithms as plug-in apps into their research workflow. This helps them analyse new datasets to carry out clinical research in the fields of radiology, oncology, neurology and cardiology.

University Hospital Cologne Department of Radiology head David Maintz said: “We use IntelliSpace Discovery to bring our research activities to the next level. Everybody is talking about artificial intelligence. We are making our own deep learning AI algorithms.”

“AI is the connective tissue to seamlessly integrate data and technology to enable precision diagnosis.”

Philips chief innovation & strategy officer Jeroen Tas said: “Together with our customers, we’re enabling research in adaptive intelligence with the goal to create solutions that augment healthcare professionals and improve patient care and efficiencies of care delivery, both inside and outside of the hospital.

“AI is the connective tissue to seamlessly integrate data and technology to enable precision diagnosis. At RSNA 2018, we’re showing how AI is laying the foundations for solutions of the future.”

Philips IntelliSpace Discovery is intended for research purpose only and not for patient diagnosis or even for selection of treatments.

Among the building blocks of the platform include front-end application, study management, machine learning, clinical research and research services.

The front-end application feature enables integration with existing clinical infrastructure to provide seamless access to vendor agnostic and multi-modality data-sets. Furthermore, research advanced visualisation tools enable data to be prepared and then processed for AI training with existing AI tools.

Study management becomes possible as completely scalable infrastructure contains a vendor-neutral research archive for structured and unstructured data.

Machine learning is possible as the research data and deployment platform obtains batch processing in a scalable computing environment to facilitate iterative development and validation.

The platform enables clinical research as the AI assets and capabilities include multi-parametric enablement, tumor quantification and stratification, and deep-learning networks.