Critical Care Suite
GE HealthCare
An AI package embedded in the X-ray system itself rather than in the PACS. Analyzes chest radiographs on the device seconds after acquisition and flags suspected pneumothorax before the image reaches the reading queue.
At A Glance
- Manufacturer
- GE HealthCare
- Medical specialty
- Radiology
- Clearance route
- 510(k)
- Year authorized
- 2019
- Modality
- Chest X-ray
- Role in workflow
- Triage and notification
The interesting design decision here is location. Most imaging AI runs downstream in the PACS, which means the finding surfaces after the study has been transmitted, queued, and routed. Running the model on the X-ray unit itself collapses all of that, and in a portable ICU setting the technologist is still at the bedside when the flag appears.
Where It Is Used
- Point-of-acquisition pneumothorax flagging
- ICU and emergency portable radiography
- Automatic endotracheal tube placement assessment in later versions
Limitations Worth Knowing
On-device inference constrains model size and update cadence: the algorithm ships with the equipment and does not improve between service updates. The indication is also narrow. This is not a general chest X-ray reader, and the flag applies to a single finding class.
Regulatory status, indications, and labeling change over time. Confirm current status on the FDA's AI/ML-Enabled Medical Devices list and in the manufacturer's current instructions for use before relying on anything stated here.