Critical Care Suite


    The first-of-its-kind imaging device with embedded artificial intelligence for case prioritization and quality control.

    510(k) cleared: K183812


      Critical Care Suite

      GE Healthcare has developed a revolutionary new tool that is designed to quickly identify and help prioritize critical cases such as Pneumothorax.
      • Critical Care Suite2, powered by Edison, embeds artificial intelligence for triaging in an x-ray imaging device, turning what was once a conceptual idea into reality by:

        • Detecting nearly all large pneumothoraxes (96% sensitivity)
        • Detecting 3 out of 4 small pneumothoraxes (75% sensitivity)
        • Limiting false alerts (94% specificity)
        • An Area Under Curve (AUC) of 0.96
        • Positive Predictive Value of 35% to 70% for pneumothorax prevalence of 4% to 15%

        The Critical Care Suite also provides triage notifications that:

        • Are sent to PACS upon transfer of the original diagnostic images
        • Enable PACS worklist prioritization
        • Present on-device notifications to the technologist3

      Quality Care Suite

      • Quality Care Suite operates in parallel to system image processing to provide real-time quality alerts. 

        1. Intelligent Protocol Check2 accurately conducts (AUC>0.99) an automated quality check to detect errors on the acquisition system, such as improper protocol used, thus enabling the technologist to determine if the image needs to be repeated or reprocessed before sending to PACS.

        2. Intelligent Field of View2 accurately detects (AUC>0.99) when a lung field is clipped in a frontal chest x-ray, allowing the technologist to determine if a repeat is required before sending the image to the PACS.

        • Assists the technologist in his/her quality check by monitoring the coverage of the lung field and notifies any user if a lung field is clipped.
        • Provides real-time feedback on patient positioning which is helpful for busy techs.

        3. Intelligent Auto Rotate saves the technologist 3-4 user interface clicks on >80% of mobile chest x-ray exams.

        • At a medium-to-large size hospital, that totals a savings of nearly 20 hours, or 3 working days and 70,000 "clicks" per year.4
        • With an AI algorithm being 99.4% accurate, it is estimated that 20 hours of manual "clicks" would be reduced to 7 minutes per year.4,5

      AMX™ 240 with Critical Care Suite

      • Listen to your peers

        Listen to your peers from University of California, San Francisco, Humber River Hospital, Mahajan Imaging, and St Luke's University.

        “The benefit of having the AI algorithm is ultimately better patient care. Better encompasses a number of sub-facets, the most important of which is accurate results.”

        Dr. Karl Yaeger

        Diagnostic Radiologist, St. Luke’s University Health Network

      Critical Care Suite in action

      Critical Care Suite automatically scans images immediately upon acquisition in the x-ray system for critical findings (pneumothorax) without requiring additional infrastructure or IT networking to implement the AI solution. The x-ray artificial intelligence notifications arrive in PACS at the exact same time as the original DICOM images, causing no additional delay or processing time, enabling escalation of image review and shorter turn-around times for prioritized cases. This enables case prioritization of suspected Pneumothorax (PTX) cases to the Radiologist. Designed to optimize x-ray image management and enhance triage notification procedures, Critical Care Suite utilizes on-device artificial intelligence to prioritize pneumothorax detection.

      Related products

      1. World Health Organization Report -Communicating Radiation Risks in Pediatric Imaging
      2. K183182
      3. The tech notification is generated 15 mins after exam closure. It is contextual and does not provide any diagnostic information. The on-device, tech notification is not intended to inform any clinical decision, prioritization, or action.
      4. GE Healthcare Data on File
      5. Younis, K., al. (2019). Leveraging Deep Learning Artificial Intelligence in Detecting the Orientation of Chest X-ray Images. SIIM Conference on Machine Intelligence in Medical Imaging (C-MIMI), Oral Presentation.