Best student paper award to PhD student at DSV
Mahbub Ul Alam, PhD student at the Department of Computer and Systems Sciences, DSV, at Stockholm University, participated in an international conference in July. The paper he presented was awarded for being one of this year’s best student papers.
Congratulations Mahbub, tell us about the paper!
– Thank you! The title of my paper is “Exploring LRP and Grad-CAM visualization to interpret multi-label-multi-class pathology prediction using chest radiography”. My co-authors are Jón Rúnar Baldvinsson at Skatturinn in Iceland, and Yuxia Wang at Qamcom Research and Technology in Sweden. I would like to express my gratitude to Jón Rúnar and Yuxia for their contribution.
– In this paper, we tried to explain the decision process of deep neural networks to predict pathology – abnormality – in chest X-ray data using two popular interpretable methods. We investigated whether this explanation matches the clinical diagnosis or not. Interpretability is very crucial and it is emphasized in the recent European Union Artificial Intelligence Act. We hope that this paper will create a positive impact in this aspect.
Where did you present your paper and how did it go?
– We presented at CBMS 2022, the full name of the conference is the IEEE 35th International Symposium on Computer Based Medical Systems. It was organized in Shenzhen, China, as an online event on July 21–23, 2022. I would say that this is the premier conference for computer-based medical systems, and one of the main conferences within the fields of medical informatics and biomedical informatics. The paper was well received during our presentation, and we were very delighted and honored to receive the Best student paper award.
How is this paper connected to your PhD thesis?
– The primary focus of my thesis is improving the clinical decision support system with the help of machine learning and the internet of medical things (IoMT). One crucial aspect in this regard is the acceptance of such support systems in the healthcare community. Without a proper “explanation” of the decision and diagnosis, it would be challenging to incorporate such systems into the traditional healthcare systems. We already have several state-of-the-art interpretable algorithms. Therefore, my initial focus was to investigate how these algorithms' results resemble the actual clinical diagnosis.
– In this paper, we tried to find the answer through pathology diagnosis prediction using chest X-rays and machine learning. The insights and findings of this paper will help to create an interpretable machine learning-based IoMT application in my future work. This application will be more effective and acceptable.
How far along are you in the thesis work – and how will you move forward?
– I am at the beginning of the end part of my PhD journey. Now I am focusing on closing up, and then I will start working on my final work – writing the thesis report. Currently, I am elaborating the work based on this paper with a real-time approach. If everything goes well, I will be able to provide exciting results in the future.
– I believe in diversity and love to explore new and fresh technological innovations. I like to think that every aspect of my previous experiences is helping me to move forward in my future career. This PhD journey is a beautiful opportunity for me to learn from different sources. DSV allows me to introspect my knowledge from different viewpoints: the social aspect, the technological aspect, the ethical aspect, and the sustainability aspect, to name a few.
What are you looking forward to?
– I have a central principle in life: “Let all of us prosper together”. I genuinely believe that only through collaboration, I can move forward in a positive direction. The world is experiencing a difficult time now. The only thing I am looking forward to is stability for all. If my work can provide a minuscule level of help, that would be a bonus.
Last updated: August 3, 2022
Source: Department of Computer and Systems Sciences, DSV