Applications using artificial intelligence (AI) continue to be integrated to support everyday consumer and business activities. The healthcare field is gaining ground with rapid development and adoption of AI applications across many different areas. In addition to automating tasks and saving time, healthcare applications that integrate clinical data and AI have proven to be advantageous in areas such as radiology, where they can help streamline workflows and help clinicians manage the large amounts of data produced by imaging exams.
AI can identify patterns in imaging data with high reliability. It’s a tool that can assist clinicians in reducing their workloads, such as providing workflow automation and suggesting critical findings. AI solutions for X-rays are embedded within the imaging system itself and help detect subtle or complex patterns within the images, moving critical cases to the top of the priority radiologist reading worklist in real time. According to clinicians, using these AI tools has resulted in improvements across the board in efficiency, quality, and clinical accuracy.[1]
Leveraging AI in X-ray to help with growing demand
X-ray is a workhorse modality in radiology and is likely to hold its position in the years ahead. Demand for imaging services continues to rise as does the prevalence of certain diseases and the growing aging population.[2] According to market research trend reports, the global X-ray market is estimated to grow at a 5.14 percent compound annual growth rate from a market size of $10.793 billion in 2019 to $14.580 billion by 2025.[3] Leaders in the development of AI solutions in radiology, such as GE HealthCare, are committed to supporting clinicians with AI tools in X-ray that integrate clinical data and AI to improve efficiencies in image acquisition, reading, triage, and clinical decision support. GE HealthCare’s suite of AI algorithms built into its X-ray systems has successfully improved workflow efficiency and clinical outcomes through its accuracy and ability to reduce the cognitive workload of radiologists and technologists.
“We’re committed to developing innovative solutions to facilitate the radiology workflow, integrating the work of technologists, radiologists and clinicians with AI assistance to impact patient care,” explained Katelyn Nye, General Manager, Mobile X-ray & Artificial Intelligence.
“Our on-device AI algorithms in X-ray impact workflow by improving efficiencies, imaging quality, and consistency and enabling timely treatment with triage solutions for suspicious findings such as pneumothorax.”
Optimize image quality and consistency with on-device AI in X-ray
Understanding the clinical workflow in X-ray is key to developing effective AI solutions, such as embedding AI algorithms in the imaging system to assist imaging technologists in image acquisition with real-time quality control alerts and image rotation automation.
Innovative on-device AI solutions such as GE HealthCare’s Quality Care Suite of X-ray AI solutions include AI algorithms that operate in parallel to image acquisition and help technologists reduce image quality errors and improve efficiency during the exam. The AI conducts an automated protocol check to ensure that the correct one is used. It also automatically rotates the images for 85% of mobile X-ray exams* and can detect if the field of view is correct or if, for example, a lung field is clipped in a frontal chest X-ray. This real-time alert helps the technologist determine if the correct image is ready for radiologist review or if it needs to be retaken before the patient leaves the exam room.
“The value of these tools is especially evident to our technologists,” explains Dr. Amit Gupta, the Modality Director of Diagnostic Radiography at University Hospitals Cleveland Medical Center and Assistant Professor of Radiology at Case Western Reserve University, Cleveland, Ohio, who uses GE HealthCare’s suite of X-ray AI solutions.. “Regarding quality and consistency, we don’t realize how important it is to rotate the images and how many extra clicks it takes to rotate the image. But your technologist can tell you. This on-device quality control mechanism improves the efficiency of the entire system.”
Integrate X-ray AI into routine clinical workflow for improved care delivery
AI healthcare solutions need to be carefully vetted, tested, and secured to minimize risk. They also should provide support with minimal effort and without adding to healthcare providers’ workload burden.
More than 500 AI-enabled devices have been approved for clinical use by the Food and Drug Administration (FDA)—91 in 2022 alone.[4] GE HealthCare is committed to facilitating the adoption of AI-enabled technologies into the radiology workflow. A leader in AI innovation in medical devices, the company already has 42 AI-enabled offerings authorized by the FDA, including its suite of on-device X-ray AI solutions.[5]
Gupta and his radiology team were one of the first to implement GE HealthCare’s Critical Care Suite of X-ray AI solutions in their clinical workflow. They published their use-case observations and discussed how AI affected their care delivery when detecting pneumothorax, or collapsed lung, on X-ray exams.
“On-device AI is the key to bringing AI to the bedside,” Dr. Gupta explains. “A very important merit of this system is that the data is on the device itself and then on to the PACS [picture archiving and communication system]. That is the real world we live in, and that’s where the value goes to the next level.”
Dr. Gupta’s team collaborated with GE HealthCare to integrate the AI alerts on suspicious findings for pneumothorax directly into the PACs environment. The AI uses pattern recognition to review images and creates alerts in the PACS so that reading radiologists have a list of scans automatically prioritized by clinical urgency.
“A single click in PACS prioritizes the workflow, and I get to the scans I need to read quickly,” notes Dr. Gupta. “I can toggle between the parent image and the secondary AI, and even though the AI may be better than human-level performance, it’s always the radiologist who is reading the scan and making the diagnosis. After all, the AI is not a calculator; it’s an application built to assist you. It’s an excellent triage tool, and we’ve really successfully integrated it into our workflow for the benefit of radiology, our clinical team, and especially the patients.”
The effectiveness of the integrated AI workflow reported by Dr. Gupta’s team included significant reductions in reporting time for STAT and routine exams for patients presenting with possible pneumothorax. It was also notable how effective the AI was in detecting cases of pneumothorax during overnight hours that would not otherwise have been reported—and leaving patients untreated—until the following day.
Intelligently efficient future in X-ray
As the demand for medical imaging grows, X-ray technology will remain fundamental to diagnostic imaging. On-device AI solutions from industry partners such as GE HealthCare will continue to support radiologists and clinicians in care delivery while reducing their heavy workload. The seamless integration of these solutions—facilitated by an open collaboration between innovators and practitioners—is the key to helping clinicians achieve workflow efficiency and deliver better care.
RELATED CONTENT
View the on-demand presentation: AI-enabled detection of pneumothorax on chest X-rays
Learn more about GE HealthCare’s X-ray AI on-device solutions:
Visit the GE HealthCare X-ray AI Portal for live demonstration of our AI solutions
DISCLAIMERS
Not all products or features are available in all geographies. Check with your local GE HealthCare representative for availability in your country.
*GE HealthCare data on file
REFERENCES
[1] Pierce JD, Rosipko B, Youngblood L, et al. Seamless integration of artificial intelligence into the clinical environment: Our experience with a novel pneumothorax detection artificial intelligence algorithm. Journal of the American College of Radiology. 2021 Nov;18(11):1497-1505. doi: 10.1016/j.jacr.2021.08.023..
[2] Smith-Bindman R, Miglioretti DL, Larson EB. Rising use of diagnostic medical imaging in a large integrated health system. Health Affairs. 2008 Nov-Dec;27(6):1491-1502. doi: 10.1377/hlthaff.27.6.1491.
[3] Global X-ray device market expected to grow to $4.580 billion by 2025. Imaging Technology News. Published August 7, 2020. https://www.itnonline.com/content/global-x-ray-device-market-expected-grow-4580-billion-2025. Accessed February 1, 2023.
[4] Reuter E. 5 takeaways from the FDA’s list of AI-enabled medical devices. MedTech Dive. Published November 7, 2022. https://www.medtechdive.com/news/FDA-AI-ML-medical-devices-5-takeaways/635908/. Accessed February 1, 2023.
[5] Artificial intelligence and machine learning (AI/ML)-enabled medical devices. U.S. Food and Drug Administration. Last updated October 5, 2022. https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-aiml-enabled-medical-devices. Accessed February 1, 2023.
