Artificial intelligence and organ transplantation: Beyond the hype | Webinars

Artificial intelligence seems to be everywhere, but what really is AI, and what isn't it? How do we ensure that we use it as a force for good?

On 10th November 2025, Georgios Kourounis, a researcher at Newcastle University, joined us to talk about the use of AI in transplantation.

We looked beyond the buzzwords to explore how these new technologies can help expand access to life-saving transplants, and to reflect on the opportunities and challenges that lie ahead.

The research

Hundreds of people are currently waiting for a liver transplant, but it can be difficult to judge whether or not a donated organ is suitable for use. Surgeons use as much information as they can to make that judgement, but they will generally be cautious. This can lead to some transplant centres rejecting more donor organs than others.

Researchers from Newcastle University and Bradford University have developed an Artificial Intelligence (AI) app called Organ Quality Assessment, or OrQA for short. It can use images of organs to provide a score that can help surgeons to make their decision about whether an organ is suitable to use. They are currently testing the app to see if it is helpful to liver transplant surgeons.

The webinar

Watch the webinar here 

Your submitted questions

Below are your submitted questions from the webinar, with answers provided by the speaker Georgios Kourounis.

How do we prevent AI exacerbating existing inadvertent biases and harming equity of access?

To address the potential biases and avoid them in our training, we need to identify them. Then, we can implement techniques and solutions that make the AI algorithms learn to not replicate these biases.

Could this be used for other organs, such as lungs?

Yes, work is currently being done with kidneys and pancreases. There is also a plan to do work with lungs and hearts too. In other organs, such as hearts and lungs, it can be more useful to look at videos rather than images to see how the organ contracts.

What impact could this have on communication?

This concept could potentially help patients to have a more active role in decision making, as clinicians could share the information that has been provided by the tool. It could also make communication between clinicians more straightforward. One person's idea of 'moderate' could be different to someone else's, however an OrQA score of two will always be a score of two.

There may be some variability in biopsy and histopathology results. Can you trust validating the OrQA against these findings? 

There are ways of putting them up against each other. Part of the speaker's PhD is to compare their predictive ability against outcomes and in doing that we'll be able to see whether or not OrQA is inferior to biopsy. However, doing this requires large projects with lots of participants.

Will AI be able to tell if organs that aren't healthy enough now could recover enough to become viable if put on a perfusion machine?

It is possible that these AI tools could help triage organs. For example, it could identify which organs are ready to be transplanted straight away, which need to go through viability assessment, and which are very unlikely to succeed.

Can we as human beings ever trust AI recommendations? Who will be the final decision maker?

AI is intended to be just one of the tools that is used to make a decision. It can help in making some of the very difficult and subjective evaluations more objective and give us more evidence, but we need to be responsible for our own actions.

Has your research involved any social scientists?

Yes, as part of the wider orQA team we have worked with other allied health professionals such as psychologists and ethics experts.

When you were building a machine learning model to predict graft outcomes before transplantation, especially with new treatments, are you training the model on both the discarded and accepted organs?

For these livers, the algorithm doesn't incorporate outcomes of the transplant, and as such we have used both organs that were transplanted and not transplanted.

Are patients going to have a choice to opt out of whether AI has played a part in selecting their organ?

Patients should have the right to do that, which depends on the communication that patients have with their clinicians.