Machine learning and artificial intelligence (AI) have been generating a lot of buzz in recent years. Due to advances in processing power, the availability of broader and deeper data, and advances in the maturity of AI tools, AI has a real opportunity to help improve the efficiency and outcomes of imaging reads. While progressive organizations have started rolling out AI tools, regulatory and legal barriers can delay implementation. However, on September 12, GE announced that its Critical Care Suite, a collection of AI algorithms which can be embedded into a mobile X-ray, was approved by FDA.
Leveraging the algorithm: How imaging leaders can benefit from early AI adoption
The system automatically scans for signs of suspected pneumothorax, a type of collapsed lung. If this condition is identified, an alert is sent along with the original chest X-ray to a radiologist via PACS. Due to the rise in urgent imaging orders, it can be challenging for radiologists to identify orders that legitimately require immediate care. The Critical Care Suite allows radiologists to immediately triage collapsed lung cases and ensure patients receive the timely care they need.
What separates the Critical Care Suite?
There are many, many AI solutions available, ranging from algorithms to improve mammography reads to those that reduce unnecessary utilization for specific conditions or gauge follow-up adherence to imaging orders. However, there are two key applications that distinguish GE's Critical Care Suite from other available AI products:
- Critical Care Suite notifies the radiologist and technologist of critical findings to ensure proper communication and care delivery. Many AI applications ignore a key player in the imaging equation: the technologist. However, the Critical Care Suite sends notifications to both the radiologist and technologist with any critical findings. Further, the system automatically runs quality algorithms to flag proper protocol and auto-rotates the images on-device for the technologist. This support provided to radiologists and technologists alike can bring care delivery to the next level.
- On-device installations allows for immediate results without dependency on connection or transfer speed. Due to on-device integration, critical findings are included when the radiologist reviews the original image, preventing processing delay. This also benefits the technologist, as quality check-ins are received immediately, helping them make the correct decision in the moment.
4 Steps for effective implementation
While there's plenty of buzz around AI's capabilities, there's less guidance on implementation. To ensure successful deployment, imaging programs must have clearly defined goals and measures of success, a deepened focus on workflow and process integration, and consistent monitoring of progress towards strategic outcomes.
Principles to ensure effective implementation
- Engage all stakeholders in the planning process. Machine learning has the potential to revolutionize medical imaging. Radiologists can use this technology to make volumes of data actionable, streamline workflow, and ultimately improve patient outcomes. However, machine-learning initiatives can fail if health care organizations do not address existing cultural resistance to new IT systems, or quell the fear that AI will make the radiologist role obsolete.
- Be mindful of your scope of application and implementation timeline. Many machine-learning algorithms are narrow in their application, working across select modalities to inform decisions on specific diseases. Although compelling cases exist in imaging, many machine-learning tools are still under development, and may take years before they are available for clinical use.
- Incorporate machine learning as a complement to the radiology staff. Even when algorithms are accurate, radiologists still need to apply their own judgment, using the algorithm as a secondary support system to optimize care.
What imaging should do
While IT may take the lead in AI initiatives, imaging needs to play its part in the planning process. Imaging leaders should consider tying AI projects to organizational strategy, involve staff in the decision making process, ensure proper governance, and play close attention to new technology. Thoughtful planning on the imaging end can ensure programs get the most out of investment and reap the rewards of AI.
Follow the path to AI at your organization
A new wave of AI-powered capabilities is likely to improve—and even—transform health care operations, but success with these new technologies requires a strong foundational analytics program.
Health care organizations must ensure they have the necessary data sources, architecture, governance, talent, leadership, and data-driven culture for their programs to deliver consistent value. With all of the necessary assets in place, health care organizations will then be ready to put analytics and AI into practice.