Makes CT, MRI, ultrasound, and X-ray machines whose 4 million connected devices continuously train the Edison AI platform.
At a glance
- Depends onDownstream position: depends on 8 industries, supplies 3
- Scale
Makes CT, MRI, ultrasound, and X-ray machines whose 4 million connected devices continuously train the Edison AI platform.
Latest report · July 7, 2026
Read the full structural reportWhat this company is and how it runs — written from structure, not news.
GE HealthCare makes CT, MRI, ultrasound, and X-ray machines, and every scan those devices perform feeds imaging data into its Edison AI platform, which uses that data to make its diagnostic algorithms more accurate. Because Edison's training set is drawn from over 4 million devices spanning all four imaging types — a fleet assembled over decades — a competitor would have to place a comparable fleet and then wait years for enough studies to accumulate before the dataset could even exist, which means no amount of new investment can shortcut the process. The same MRI machines that add to this installed base require liquid helium to keep their superconducting magnets running, and industrial gas suppliers are already rationing during a global shortage, so the rate at which new devices can be manufactured and connected to Edison is capped by a raw material that GE HealthCare cannot control. The scale that makes Edison's algorithms harder to replicate also concentrates risk: if the platform were breached, the same connectivity linking all 4 million endpoints would expose patient imaging data across every one of them at once, in a way a smaller competitor with fewer connected devices simply would not face.
How does this company make money?
GE HealthCare collects a large upfront payment each time a CT scanner, MRI machine, or ultrasound system is sold to a hospital. After the sale, it charges recurring fees for service contracts that cover maintenance and repairs. Separately, hospitals using Edison pay a per-study licensing fee each time the AI software is used to help analyze an imaging scan.
What makes this company hard to replace?
Switching away from GE HealthCare equipment means reconfiguring how the hospital's EPIC or Cerner electronic records system connects to new imaging devices — a process that typically takes 6 to 12 months and that hospital IT departments are reluctant to repeat. On top of that, radiologists learn to work with GE HealthCare's AI protocols as part of their daily routine, and moving to a competitor's system would require retraining every radiologist in the department from scratch.
What limits this company?
Industrial gas suppliers are already rationing liquid helium during a global shortage, and helium is required to manufacture every new MRI unit. At the same time, the rare earth elements from China used in MRI permanent magnets face their own supply limits. Neither constraint can be fixed by spending more money — the materials simply are not available in sufficient quantities. Together, these two input ceilings set a hard cap on how many MRI machines can be built and added to the installed base, which is the same base that feeds Edison's training pipeline.
What does this company depend on?
GE HealthCare cannot operate without liquid helium from industrial gas suppliers to keep MRI magnets running, rare earth elements from China to build those magnets, iodinated contrast agent manufacturing capacity to supply Omnipaque and Visipaque, FDA 510(k) clearances before any changes to imaging devices can be sold, and working integration with EPIC and Cerner EMR systems so that Edison can connect to hospital workflows.
Who depends on this company?
Hospital radiology departments rely on GE HealthCare CT and MRI equipment for daily diagnostic imaging — if those machines went down, procedures would be delayed or cancelled. Interventional cardiologists depend on iodinated contrast agents like Omnipaque and Visipaque to see inside blood vessels during catheterizations; without supply, those procedures stop. Radiopharmacy operations at hospitals depend on technetium-99m generator supply for nuclear imaging; without it, those departments shut down.
How does this company scale?
As more hospitals use GE HealthCare devices, more imaging studies flow into Edison, and the AI algorithms get more accurate across all four modalities — each new device makes the platform slightly better for every other connected device. However, manufacturing more devices, especially MRI machines, does not get easier as the company grows: helium availability and rare earth element supply are fixed external limits that more factories or more investment cannot resolve.
What external forces can significantly affect this company?
U.S.-China trade tensions put the supply of rare earth elements from China at risk, which would directly affect MRI magnet production. Medicare reimbursement rate cuts for diagnostic imaging reduce the budgets hospitals have available to buy new capital equipment. The European Union Medical Device Regulation now requires new clinical evidence for imaging systems that were previously approved, adding cost and time before products can be sold in Europe.
Where is this company structurally vulnerable?
If a cybersecurity attack compromised the Edison platform, all 4 million connected devices would be exposed at once — meaning patient imaging data from every connected hospital could be affected in a single incident. That concentration of exposure could trigger mandatory regulatory action requiring the devices to be disconnected from the network. Disconnecting them would shut down the data pipeline that makes Edison's algorithms work, breaking the loop that the entire platform depends on.
Price is read as structure — trend, levels, range, peak and volatility drawn on the chart. It does not predict where price goes next.
Sign in to view price data.
Sign inWhat the company actually pays, and whether its own cash supports it.
The reported statements, read against the company's own industry.
3 interpretations currently present — each is a set of fired observations whose alignment reads as one structural pattern. Click an observation to see the numbers behind it.
Screen for these patternsIs this company growing?
Three multi-year observations co-occur: revenue increased year-over-year in each of the last three fiscal years, gross profit (absolute level) increased year-over-year in each of the last four fiscal years, and net income was positive in each of the last five fiscal years. The configuration describes growth-and-profitability persistence across three different windows.
How is this stock valued?
Three observations describe the present configuration: the current close sits below the 40-week SMA (the conventional 'below 200-day SMA'), the company has reported positive net income in each of the last three annual periods, and operating cash flow exceeded net income in the most recent annual period.
Where is this company structurally exposed?
Three leverage observations have converged at elevated readings: debt is large relative to equity, large relative to total assets, and large relative to trailing operating cash flow. The capital structure is leveraged on three different denominators at once.
An interpretation is present only while every observation it reads stays fired (score ≥ 70). It describes what the aligned readings show — never a verdict, never a prediction.
Shared structure with peers — never a ranking.
Structural observations derived from financial data, industry benchmarks, and supply chain position.
Companies that share the same coordination system — how they create, deliver, or capture value.
Companies that share active interpretations — structural patterns currently present in both stocks.
GE HealthCare Technologies Inc.
July 7, 2026 · CompanyGraph · GEHC
GE HealthCare reported positive net income in every fiscal year from FY2021 through FY2025, with operating cash flow exceeding net income in the most recent year — profitability that the cash largely kept pace with (verified from its own figures, as of December 2025). The sharper structural idea — that a large installed base of imaging machines feeds a data advantage that rivals can't shortcut — is the most interesting thing here, but CompanyGraph can only read it as a hypothesis: the held data doesn't split out software revenue, so the advantage that would make it true can't yet be seen.
Read the full report
CompanyGraph