Portrait photo of Zhendong Wang, PhD student at the Department of Computer and Systems Sciences, SU
Zhendong Wang. Photo: Private.

In June 2021, Zhendong Wang presented a paper at the International Conference on Artificial Intelligence in Medicine (AIME) – and won the best student paper award. The paper is titled “Counterfactual Explanations for Survival Prediction of Cardiovascular ICU Patients” and investigates how explainable machine learning methods can be applied in the medical domain. It is co-written with DSV researchers Isak Samsten and Panagiotis Papapetrou.

Congratulations, Zhendong! What was it like to receive the award?

– Thank you! There were three candidates for the best student paper at AIME, and it was really surprising and exciting to get the award. It happened right before midsummer, so it was the perfect time to celebrate. This was the first time I presented a paper at a scientific conference. The conference was supposed to take place in Portugal and I was looking forward to go there to meet different researchers, but because of the pandemic it was held online. It was still a good experience to present online, and hopefully I’ll be able to go to other conferences in the future.

What’s the topic of your paper, and what are the implications of your research?

– The paper is focused on explainable machine learning, but we’re also experimenting on data from real patients in the US who have been in hospital with severe heart conditions. The paper deals with so-called counterfactual explanations, and how machine learning can help doctors understand how to alter a prediction outcome with reasonable modifications. By using historical, cardiovascular records and combining machine learning with human intelligence, diagnoses will be more trustworthy. Medical doctors can make better decisions concerning treatments and drugs, and the goal is of course to give patients a better chance at surviving. I think that we will see more and more machine learning applications in hospitals in the future.

How did you end up doing your PhD here at DSV?

– I moved to Sweden five years ago. At first, I studied my Master programme at Linköping University for two years. Then I went to Stockholm, to work in data engineering for two years. After that, I decided to go back to academia. I’ve been a doctoral student at the Department of Computer and Systems Sciences – DSV – for one year now. The topic of my thesis is not yet decided, but I will continue to work in this direction – with explainable machine learning and time series analysis. I also want to put my research work to practical usages in healthcare.

– The research topic I’m dealing with now is what I’ve always wanted to work with. If I can push boundaries in this area, I will not only enhance my knowledge and expand the research field. I will also help medicial practitioners, both researchers in their work and doctors who can make better diagnosis decisions with the help of machine learning, says Zhendong Wang.

The awarded paper is published online

See Zhendong Wang’s presentation at the AIME conference

More information can be found on the conference website

Contact zhendong.wang@dsv.su.se