This graduate course follows a research-based teaching approach and discusses methods for supporting human intelligence with artificial intelligence to augment a human-in-the-loop to solve problems from health informatics. For practical applications we focus on the industrial standard Python. Students should get acquainted with the specific features of medical data and basic examples of automatic and interactive machine learning, spedifically being able to use Python for solving typical problems in medical informatics. A special focus of the class of 2019 is on explainable AI and ethical, legal and social issues of machine learning in health informatics. Please consult the course homepage for updted information!
Please refer to the course Webpage for updated information:https://hci-kdd.org/machine-learning-for-health-informatics-class-2019
ML meets health informatics, introduction to the health domain, challenges and future direction; fundamentals and specifics of biomedical data, information and knowledge; knowledge, decision, cognition, reasoning, probability, uncertainty, Bayesian statistics, Gaussian processes; iML - Interactive Machine Learning with the human-in-the-loop: protein folding, crowdsourcing, gamification and ML; iML: towards open medical data: k-anonymization, privacy preserving ML; iML: intelligent, interactive visualization and visual analytics, subspace clustering; iML: interactive tumor growth simulation; outlook and future challenges.
The 2019 course has a special focus on privacy, security, data protection safety, ethical and social issues and particularly on explainable AI.
Due to raising legal and privacy issues in the European Union glass box approaches will become important in the future to be able to make decisions transparent, understandable and re-traceable.
Our aim is to explain why a machine decision has been made, paving the way towards explainable-AI.
ECTS-Breakdown (sum=75h, corresponds with 3 ECTS, where 1 ECTS = 25 h students workload):
a) Presence during the lecture 8 x 3 h = 24 hb) Preparation before and after the lecture 8 x 1 h = 8 hc) Preparation of assignments and presentation 28 h + 2 h = 30 hd) Written exam including exam preparation 1 h + 12 h = 13 h
Sum TOTAL student's workload = 75 h
Will be adapted to the previous knowledge of the students and announced once the course has started.
Entry in the course list on Tuesday, 12 March 2019, 17:30 hrs
Holzinger, A. 2014. Biomedical Informatics: Discovering Knowledge in Big Data, New York, Springer, doi:10.1007/978-3-319-04528-3.
Holzinger, A. (ed.) 2016. Machine Learning for Health Informatics: State-of-the-Art and Future Challenges, Lecture Notes in Artificial Intelligence LNAI 9605, Cham: Springer International, doi:10.1007/978-3-319-50478-0.
Holzinger, A. 2016. Interactive Machine Learning for Health Informatics: When do we need the human-in-the-loop? Brain Informatics, 3, (2), 119-131, doi:10.1007/s40708-016-0042-6.
Interest in machine learning with application to health informatics with a special focus on privacy, security, data protection safety, ethical and social issues and in explainable AI [3] and the doctor-in-the-loop [4]
[3] Andreas Holzinger (2018): From Machine Learning to Explainable AI. 2018 World Symposium on Digital Intelligence for Systems and Machines (DISA), 23-25 Aug. 2018 2018. 55-66, doi:10.1109/DISA.2018.8490530.
[4] Andreas Holzinger (2016): Interactive Machine Learning for Health Informatics: When do we need the human-in-the-loop? Brain Informatics, 3, (2), 119-131, doi:10.1007/s40708-016-0042-6.