What makes AI tools change after hospital commissioning?

Question

Accepted Answer

AI tools can change after commissioning because updates, new data, and real-world clinical changes can shift how they perform

AI systems do not always remain unchanged after they enter hospitals. Their performance can be affected by software updates, new datasets, changing patient populations, and evolving clinical practices, meaning a tool that works well at one point may not perform the same way later.
Software updates can alter how an AI system behaves over time
Once an AI tool is deployed, changes to its software can affect its performance. The key point is that the system’s behaviour is not fixed after it is introduced, so later updates can lead to differences in how it processes information.
New datasets and patient data shifts can change results
AI performance may also be influenced by new datasets. When patient populations differ from the data used to develop or train the system, the tool may behave differently, particularly across settings with different demographics and disease patterns.
Changing patient populations can affect accuracy and reliability
Patient demographics and disease patterns can vary significantly between locations. The example given is that an AI tool trained using patient data from Europe or North America may not automatically work equally well for patients in Kuala Lumpur, Kota Bharu, Kuching, or rural Sabah.
Evolving clinical practices and workflows can require re-validation
Clinical practices and hospital workflows evolve, and that can affect how an AI tool performs in routine use. Because healthcare environments differ, safety cannot stop at initial approval and needs continuous oversight to keep accuracy, reliability, and fairness in view.

Reading history

Hide →

This history is cleared when the page is closed.

What makes AI tools change after hospital commissioning?

Loading

Was this article helpful?

Helpful

Not helpful

Report an issue

1

0

0

Share with friends