Educational content on VJOncology is intended for healthcare professionals only. By visiting this website and accessing this information you confirm that you are a healthcare professional.

Share this video  

ESMO AI and Digital Oncology 2025 | Educating clinicians to use and evaluate AI tools in oncology

Anita Grigoriadis, PhD, King’s College London, London, UK, comments on the importance of education in helping clinicians understand and work with artificial intelligence (AI) tools, highlighting the need for courses and lectures on topics such as data governance, model transfer, and intellectual property licenses. Clinicians must not only learn about new AI-related terms and concepts, such as probability calibration and uncertainty quantification, but also understand their practical applications and limitations, including issues related to data diversity and bias. This interview took place at 2025 European Society for Medical Oncology (ESMO) Asia Congress in Singapore, Singapore.

These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.

Transcript

Well, I think there are educational, you know, there will be more and more educational paths. So, for example, you know, we have one that is a new one that will come up in October that teaches about governance. Yeah, because there’s a lot of, you know, data governance also involved with using AI tools. So more and more, I think there will be, you know, courses, lectures that will come out that will, you know, will be maybe open source or maybe sometimes one has to pay for it that will help our clinicians to learn, you know, what kind of, you know, what new tools are coming out, what are the new names...

Well, I think there are educational, you know, there will be more and more educational paths. So, for example, you know, we have one that is a new one that will come up in October that teaches about governance. Yeah, because there’s a lot of, you know, data governance also involved with using AI tools. So more and more, I think there will be, you know, courses, lectures that will come out that will, you know, will be maybe open source or maybe sometimes one has to pay for it that will help our clinicians to learn, you know, what kind of, you know, what new tools are coming out, what are the new names. You know, as we all know, there’s new names popping up everywhere from, you know, foundation models to AI agent to hallucination. You know, what does this all mean? You know, but equally, well, you know, we need to not only know the words, we also need to understand, you know, what they could do. You know, you hear about, you know, monitoring strategies such as probability calibration, uncertainty quantification, conformal prediction. So lots of new words are coming out. But it’s not only about the models, but it’s also about the governance, you know, and the data. So you need to know, you know, how the, you know, what is important for model transfer, you know, how IP licenses. And then at the end also where the data is coming from. So where is the data, you know, what was the data that was used for building these AI tools? You know, is it across different socioeconomic populations? You know, is there ethnic diversity? And is it actually addressing the problem when I have this patient sitting in front of me?

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.

Read more...