📚 Class Preparation

2025/26 Autumn Term

🗺️ Context

This week, we’ll look at our second case study!

A field where AI/data has long been touted as a solution to every problem is medicine. Supervised learning techniques, in particular, seemed as if they would bring about huge benefits when it came to diagnosing patients more accurately, tailoring treatment regimens, etc… But, there was always the promise and there was clinical reality…When the pandemic happened, the good, the bad and the ugly sides of AI in medicine came to the fore. Many uncomfortable questions were brought forth and many lessons learned.

📖 Class Preparation

Take a look at the class handout ahead of time

Imagine that your goal is to partition the naughts and the crosses. How would you approach this?

Note

You can download the handout by clicking on the button below:

Read the following articles

Mandatory readings

• Benaich, Nathan (2020) “AI has disappointed on Covid”. Financial Times – (Benaich 2020)
• Heaven, William Douglas (2021). “Hundreds of AI tools have been built to catch covid. None of them helped”. MIT Technology Review – (Heaven 2021)
• Callaway, Ewen (2023). “How AlphaFold and other AI tools could help us prepare for the next pandemic”. Nature– (Callaway 2023)
• Ono, Sachiko, and Goto, Tadahiro (2022). “Introduction to supervised machine learning in clinical epidemiology”. Annals of clinical epidemiology vol. 4,3 63-71. 1 Jul. 2022 – (Ono and Goto 2022)
• Nwanosike, Ezekwesiri Michael, Conway, Barbara R, Merchant, Hamid A and Hasan, Syed Shahzad (2022) “Potential applications and performance of machine learning techniques and algorithms in clinical practice: A systematic review”. International Journal of Medical Informatics, Volume 159, 2022, 104679, ISSN 1386-5056. –(Nwanosike et al. 2022)
• Morris, Stephen and Heikkilä, Melissa (2025). “Microsoft claims AI diagnostic tool can outperform doctors”. Financial Times –(Morris and Heikkilä 2025)

📖 Recommended readings

• Väänänen, Antti and Haataja, Keijo and Vehviläinen-Julkunen, Katri and Toivanen, Pekka. 2021. “AI in healthcare: A narrative review [version 2; peer review: 1 approved, 1 not approved]” F1000Research 10:6.

• Wynants Laure, Van Calster Ben, Collins Gary S, Riley Richard D, Heinze Georg, Schuit Ewoud et al. 2020. “Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal” BMJ 369 :m1328.

💡Tip

This lab is going to be a little on the technical side (this is a data science class after all). However, please come prepared to discuss the wider social ramifications of these topics. Overfitting is not just a technical detail, for example, it can lead to a host of negative externalities if overlooked.

References

Benaich, Nathan. 2020. “AI Has Disappointed on Covid.” Financial Times, September. https://www.ft.com/content/0aafc2de-f46d-4646-acfd-4ed7a7f6feaa.

Callaway, Ewen. 2023. “How AlphaFold and Other AI Tools Could Help Us Prepare for the Next Pandemic.” Nature 622 (7983): 440–41. https://doi.org/10.1038/d41586-023-03201-4.

Heaven, Will Douglas. 2021. “Hundreds of AI Tools Have Been Built to Catch Covid. None of Them Helped. MIT Technology Review.” July 30, 2021. https://www.technologyreview.com/2021/07/30/1030329/machine-learning-ai-failed-covid-hospital-diagnosis-pandemic/.

Morris, Stephen, and Melissa Heikkilä. 2025. “Microsoft Claims AI Diagnostic Tool Can Outperform Doctors.” Financial Times, June. https://www.ft.com/content/149296b9-41b6-4fba-b72c-c72502d01800.

Nwanosike, Ezekwesiri Michael, Barbara R Conway, Hamid A Merchant, and Syed Shahzad Hasan. 2022. “Potential Applications and Performance of Machine Learning Techniques and Algorithms in Clinical Practice: A Systematic Review.” International Journal of Medical Informatics 159: 104679. https://doi.org/https://doi.org/10.1016/j.ijmedinf.2021.104679.

Ono, Sachiko, and Tadahiro Goto. 2022. “Introduction to Supervised Machine Learning in Clinical Epidemiology.” Annals of Clinical Epidemiology 4 (3): 63–71. https://doi.org/10.37737/ace.22009.