Leiden University Medical Centre (LUMC) is going to use Philips' IntelliSpace AI Workflow Suite. This platform collects new and existing artificial intelligence (AI) models from the LUMC in one central location. Through its collaboration with Philips, LUMC aims to facilitate and improve the use of AI in clinical processes; the platform gives researchers and doctors at the hospital easier access to AI.
IntelliSpace AI Workflow Suite is a solution presented by Philips at the 2019 Radiological Society of North America Annual Meeting (RSNA). The platform is still under development and not yet officially for sale. Nevertheless, LUMC is already starting to use the platform. This makes the Leiden hospital the first party in the world to deploy the IntelliSpace AI Workflow Suite.
Retrieve AI analyses from other systems
The platform will be linked to clinical systems at LUMC. This integration allows systems to automatically request AI-driven analysis, for example of images from an MRI scan. The platform brings together both new and existing LUMC AI models on a single platform. This allows doctors and researchers to find, retrieve, adapt and use all available AI models on one central platform. It is also possible to share scientifically validated AI models with other hospitals.
Specialists from both Philips and LUMC are also joining forces to accelerate the embrace of AI. Philips supports LUMC in personalising, evaluating, validating and approving new AI models. Examples include an AI model that provides real-time feedback to lab technicians while they take X-ray images. Such a model increases the likelihood that a high-quality image will be taken of the patient immediately during the first attempt.
Tracking abnormalities in patient data
Together with Philips, LUMC reports having identified several application areas for the new platform. For example, together with the Intensive Care Department, LUMC is developing a model that identifies and reports abnormalities based on patient data, but also predicts readmissions. Philips is providing the technical implementation of this model. In the Radiology department, AI is also used to analyse radiological images, identify and report on abnormalities. This is always done under the supervision of a doctor, reports LUMC.
LUMC also expects that AI models can improve the image quality of medical images and reconstruct images. In addition, AI can take over administrative and operational work, examples of which include scheduling patient appointments and predicting waiting times. With the help of AI, LUMC aims to make better use of available scanners in the radiology department and thus be able to help more patients. Work is also under way to roll out the platform in the Intensive Care and Cardiology departments.
'Application of AI lags behind expectations'
"There is a lot of talk about the application of AI in healthcare and the expectations are tremendously high. However, the actual application of AI in day-to-day healthcare lags behind. It is high time that AI is implemented in daily practice and that we establish its added value. We are convinced that AI can demonstrably improve quality of life and quality of care," says Prof Dr Mark van Buchem, Radiologist and AI leader at LUMC. "And yet we still too often see people stop developing AI. This is happening all over the world and we need to do something about it. There is now a real need to make AI fit for practice and actually start using it with doctors and other healthcare professionals. Only then can we use the potential of AI in healthcare to our advantage and test it. By working with Philips to establish this AI platform and making AI available first to our colleagues at LUMC, and if proven successful to all healthcare professionals worldwide, we are broadening and accelerating the application of AI in healthcare."
"We are very proud that LUMC is the first in the world to launch the IntelliSpace AI Workflow Suite platform. The platform is a virtual environment in which AI models are managed. Doctors and researchers at LUMC can continue to develop and share models with each other in IntelliSpace AI Workflow Suite," says José Strijbos, Director Healthcare Informatics at Philips Benelux. "Our goal is to develop solutions to healthcare challenges together with university medical centres and hospitals. AI is a promising development to improve healthcare outcomes and the patient experience. In addition, AI can help improve the working environment of healthcare providers and help reduce healthcare costs. The introduction of IntelliSpace AI Workflow Suite is an important step to realise the potential of AI in healthcare."
Author: Wouter Hoeffnagel
Source: Philips
Source: LUMC
