Informed Machine Learning in Healthcare
- Type:Bachelor, Master
Artificial Intelligence (AI) and Machine Learning (ML) are among today's most disruptive technologies. Despite their success across different applications a general adoption fails to establish in the health sector. This has a multitude of reasons, including but not limited to restricted data availability, lack of explainability or everchanging best practices in patient treatment. The inclusion of available expert knowledge might mitigate many of these problems. The combination of prior available expert knowledge and ML approaches falls under the term of Informed Machine Learning.
In your thesis, you can work on several aspects of Informed Machine Learning for health care. Possible topics include, but are not limited to:
- Research on the transformation of available expert knowledge into an ML accessible format
- Conceptualization and/or Implementation of Informed Machine Learning solutions
- Adaptation of Informed Machine Learning solutions to Federated Learning
- Explainability of Informed Machine Learning solutions
This is an umbrella topic since topics of interest change rapidly. A specific topic will be selected during a first meeting.
- L. von Rueden et al., "Informed Machine Learning - A Taxonomy and Survey of Integrating Prior Knowledge into Learning Systems," in IEEE Transactions on Knowledge and Data Engineering, doi: 10.1109/TKDE.2021.3079836.
- Vapnik, V., & Vashist, A. (2009). A new learning paradigm: learning using privileged information. Neural networks : the official journal of the International Neural Network Society, 22(5-6), 544–557. https://doi.org/10.1016/j.neunet.2009.06.042
- Xie, Xiaozheng, et al. "A survey on incorporating domain knowledge into deep learning for medical image analysis." Medical Image Analysis 69 (2021): 101985.
- Chew, H., & Achananuparp, P. (2022). Perceptions and Needs of Artificial Intelligence in Health Care to Increase Adoption: Scoping Review. Journal of medical Internet research, 24(1), e32939. https://doi.org/10.2196/32939