Home / Researcher Human Centered AI

Researcher Human Centered AI

As a research scientist, I have an education in Cognitive Science as well as Computer Science and Engineering. I started working at Philips research in 2011. In 2021 I started my PhD research within the e/MTIC framework. In 2024, I became a senior researcher at Avans Hogeschool.

I am currently working on user interaction with clinical decision support (CDS) in the intensive care unit (ICU) in the context of a PhD trajectory together with Máxima Medical Centre and TU/e.

My ambition is to come up with a design for a CDS system in the ICU that supports collaboration among physician and CDS to improve the decision-making process with special emphasis on the physician’s reasoning process as well as the fit within the ICU environment and the team dynamics in this environment.

In the past, I have worked on motivating people to make lifestyle changes through technology, helping cancer patients to understand their condition and make informed decisions together with the physician and making predictive models for treatment and diagnosis in oncology.

Past experience

From 2019 to 2023, I work as a researcher in the Cognitive Engineering team at Philips Research. I conducted research into Clinical Decision Support tools in the area of Explainable AI and Human Interaction with AI.

From 2014 to 2019, I worked as a researcher in the Oncology Solutions department at Philips Research. I conducted research into predictive models for cancer care, patient centered support tools and dashboards for decision support for physicians.

From 2011 until 2014, before I started focussing on the topic of interaction with Artificial Intelligence, I worked on user-system interaction through formal modeling and simulation of psychological processes as well as unobtrusive user profiling techniques through machine learning, reasoning, etc., as a research scientist in the Human Interaction and Experiences research group at Philips Research.

Education

In 2010 I received two Master of Science degrees after finishing a Bachelor of Science in Computer Science at the Technische Universiteit Eindhoven:

  • Master Artificial Intelligence, Vrije Universiteit Amsterdam Specialisation: Cognitive Science Thesis: A cognitive agent model for mindreading Diploma 16/06/2010 Judicium: Cum Laude Publication: Monique Hendriks and Jan Treur. 2010. Modeling super mirroring functionality in action execution, imagination, mirroring, and imitation. In Proceedings of the Second international conference on Computational collective intelligence: technologies and applications – Volume PartI (ICCCI’10), Jeng-Shyang Pan, Shyi-Ming Chen, and Ngoc Thanh Nguyen (Eds.), Vol. PartI. Springer-Verlag, Berlin, Heidelberg, 330-342.
  • Master Computer Science and Engineering, Technische Universiteit Eindhoven Specialisation: Formal Methods Thesis: A formal language for cognitive agent models Diploma 29/11/2010 Judicium: Cum Laude

As a student, I was involved in several committees of GEWIS, the study association for mathematics and computer science students of the TU/e. In 2005/2006, I was the vice-president of the board of this study association. I was involved in organising the National Computer Science Conference, a collaboration among several study associations of computer science throughout the Netherlands. I was involved in organising the introduction week for mathematics and computer science students. I was a student member of the faculty council. I wrote a column and articles for Cursor, the university news paper of the TU/e. I was involved in the development of a theatre show for promoting the computer science degree. I was a student assistant, involved in supporting several classes and informing high school students about the degree of computer science. And finally, I did scouting for recruitment agency YER.

Projects

  • Advancing Cancer Care and Cardiac Care Through Interpretable AI (ACACIA): The ACACIA project aims to advance cancer and cardiac care with AI, taking a patient-centered and physician-centered approach to address unresolved challenges that hamper adoption of AI in clinical practice. This project is a collaboration between Philips, Máxima Medical Centre and TU/e. Within this project, I fulfill a PhD position, working together with 2 other PhD’s on introducing AI driven Clinical Decision Support (CDS) in the Intensive Care Unit (ICU). I will be focusing on how to design CDS such that it fosters collaboration between AI and medical staff, taking into account the reasoning processes of the medical staff as well as the workflow and team dynamics. Furthermore, I will be developing a measure for evaluating the impact of such CDS on the decision-making. A more detailed account of my research proposal can be found in this Frontiers publication.
  • Hemodynamic solutions: The hemodynamic solutions project within Philips Research aims at providing integrated solutions for decision support in the ICU throughout the care path. By integrating data sources, we provide better overviews of the meriad of data points in the ICU and we create opportunities for data analysis to generate predictions to support the monitoring, diagnosis and treatment of the hemodynamic status of patients. Withing this project, I conduct qualitative user research including observation, facilitation of simulation studies and interviews. This qualitative user research should lead to models of the reasoning process, interactions with the environment and team dynamics. These models should inform the design of our solutions.
  • IP Generation on interaction with AI: In this project, we use story telling methodology to create rapid prototypes of methods for interacting with AI including advanced user interaction methodology such as gaze tracking, gesture control, voice control, AR/VR and virtual agents. As a creative writing expert, I facilitate workshops and contribute to the IP generation.
  • The future of UI: In our Philips internal innovation brief on the future of UI, I contributed to providing an overview of research challenges, state of the art and trends in human-centered AI.
  • Radiotherapy treatment decision support: Decision support for the radiotherapy department. Combining dose information with patient and disease information to support the complex process of decision-making in radiotherapy involving trade-offs among survival, side effects, cost, patient preferences and quality of life. My tasks: cognitive task analysis, workflow analysis, interviewing users. Results: mock-ups of a dashboard for decision support.
  • Oncology Analytics: Decision support for the tumor board (multi-disciplinary team meeting). Harvesting insights from local data in the hospital through clustering. My tasks: data analysis. Results: algorithms implemented in the R programming language.
  • Patient engagement: Decision support for prostate cancer patients. Shared decision-making, taking patient preferences into consideration. My tasks: implementing prediction models from literature in Python, research into preference elicitation and user profiling w.r.t. information needs and (shared) decision-making.
  • EURECA: Enabling information re-Use by linking clinical REsearch and Care: an FP7-ICT co-funded program. My tasks: developing a model for predicting Serious Adverse Events within the EURECA framework, working together with a pediatric oncologist at the Saarland University Hospital. Results: Framework for Serious Adverse Events prediction implemented in WPF.
  • Situated Coaching: Development of an automated coaching system for physical activity/healthy lifestlye coaching. My tasks: development of a formal language for specifying the reasoning processes of coaches, research into motivation and behavior change. Results: Specification of the formal language.
  • SmarcoS: Artemis funded European collaboration focussed on interusability (usability aspects in a multi-device, interconnected system). My tasks: leading a work package.
  • Ambient Intelligence: extensive user profiling through machine learning techniques. My tasks: specification of a reinforcement learning algorithm.

Publications

  • Talgorn, Elise, et al. “A Storytelling Methodology to Facilitate User-Centered Co-Ideation between Scientists and Designers.” Sustainability 14.7 (2022): 4132.
  • Hendriks, Monique, et al. “Respecting Human Autonomy in Critical Care Clinical Decision Support.” Frontiers in Computer Science (2021): 72
  • Hendriks, Monique, et al. “Data Visualization in Clinical Practice.” Data Science for Healthcare. Springer, Cham, 2019. 289-304.
  • Consoli, Sergio, et al. “Improving Clinical Subjects Clustering by Learning and Optimizing Feature Weights.” International Conference on Machine Learning, Optimization, and Data Science. Springer, Cham, 2018.
  • Consoli, Sergio, et al. “Improving Support Vector Machines Performance Using Local Search.” International Workshop on Machine Learning, Optimization, and Big Data. Springer, Cham, 2017.
  • Consoli, Sergio, et al. “Towards an automated method based on Iterated Local Search optimization for tuning the parameters of Support Vector Machines.” arXiv preprint arXiv:1707.03191 (2017).
  • Hendriks M. (2016). “Support for the Inclusion of Domain Knowledge in Prediction Models – User Evaluations of a Tool for Generating Prediction Models for Serious Adverse Events in Oncology”. In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies ISBN 978-989-758-170-0, pages 183-188. DOI: 10.5220/0005656201830188
  • Monique Hendriks, Norbert Graf, and Njin-Zu Chen, “A Framework for the Creation of Prediction Models for Serious Adverse Events” 2014 IEEE International Conference on Bioinformatics and Biomedicine, Page 17-23
  • Hekler, E.; Klasnja, P.; Traver, V.; Hendriks, M., “Realizing Effective Behavioral Management of Health: The Metamorphosis of Behavioral Science Methods,” Pulse, IEEE , vol.4, no.5, pp.29,34, Sept. 2013, http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6603401&isnumber=6603365
  • M. Hendriks, V. Antila and T. Lavrysen, 2012, Inter-usability and Intelligent Communication: Usability Aspects in a Multidevice Personal Attentive System, Proceedings of the 2012 Workshop on Ambient Intelligence Infrastructures (WAmIi), p15-18, http://alexandria.tue.nl/repository/books/755360.pdf – Monique Hendriks and Jan Treur. 2010. Modeling super mirroring functionality in action execution, imagination, mirroring, and imitation. In Proceedings of the Second international conference on Computational collective intelligence: technologies and applications – Volume PartI (ICCCI’10), Jeng-Shyang Pan, Shyi-Ming Chen, and Ngoc Thanh Nguyen (Eds.), Vol. PartI. Springer-Verlag, Berlin, Heidelberg, 330-342.

Patent applications

  • Granted: WE FR 21739693.6, DE 4181789 “1D representation of 2D movements for fine guidance”
  • Granted: 2017P02263WOUS (2016ID01658 , “A system and method for identifying prognostic biomarkers in Oncology using Deep Learning”
  • Granted: 2017P02259WOUS (2017ID03045 , “An automatic system for similar patients clustering by learning feature weights”
  • Granted: 2017P01825WOJP (2016ID01662 , “Communication quality assessment tool for Tumor Board meetings”
  • Granted: 2017P02263WOJP (2016ID01658 , “A system and method for identifying prognostic biomarkers in Oncology using Deep Learning”
  • Hendriks, Monique, Jacek Lukasz Kustra, and Sandra Vosbergen. “System and method for providing model-based predictions of quality of life implications of a treatment via individual-specific machine learning models.” U.S. Patent Application No. 17/053,954.
  • Consoli, S., Hendriks, M., Vos, P.C., Kustra, J.L., Mavroeidis, D., Hoffmann, R. (2019) EP 3 460 807 A1 “Subject clustering method and apparatus.”
  • Mavroeidis, D., Hendriks, M. Vos, P.C., Consoli, S., Kustra, J.L., Janssen, J., Hoffmann, R. (2019) EP 3 460 723 A1 “Evaluating input using a deep learning algorithm.”
  • Kustra, J.L., Hendriks, M., Vos, P.C., Consoli, S., Mavroeidis, D., Van Wissen, A., Van Halteren, A.T. (2019) EP 3 407 226 A1 “An apparatus and method for providing feedback to a participant of a communication.”
  • Vos, P.C., Hoffmann, R., Hendriks, M. (2019) EP 3 573 072 A1 “Performing a prognostic evaluation.”
  • Hendriks, M., De Pee, J., Dadlani, P. (2019) EP 3 584 796 A1 “Determining medical treatment preferences.”
  • Hendriks. M. (2018) WO/2018/154128 “SELECTING A CRITERION FOR DETERMINING WHICH SUBJECTS TO INCLUDE IN A MEDICAL TRIAL”
  • Riistama, J.M., Hendriks, M., Korst, J.H.M. (2017) US 2017 / 0344713 A1 “Device, system and method for assessing information needs of a person”
  • Hendriks, M., Van Halteren, A., Wang, L. (2014) US 2014/0272844 A1, “METHOD FOR INCREASING THE LIKELIHOOD TO INDUCE BEHAVOR CHANGE IN A LIFESTYLE MANAGEMENT PROGRAM”
  • Van Dantzig, S., Barbieri, M., Speelpenning, T., Hendriks, M. (2014) US 2014/0276243 A1 “Behavioral risk analyzer and application that estimates the risk of performing undesired behavior”

Current projects

As of February 2024, I am senior researcher at Avans Hogeschool. My personal expertise and ambition can be found in the area of interaction with Artificial Intelligence, more specifically in the area of clinical decision support. I have built up knowledge about human judgment, naturalistic decision-making and reasoning under uncertainty and use that knowledge to develop decision support tooling tailored to the ways in which humans take decisions.I am doing this research as part of a PhD trajectory, collaborating with Máxima Medical Centre and the Technical University of Eindhoven (TU/e).

Projects

  • No projects started yet

Publications

  • No publications yet

Ambition

In my research into user interaction with AI in the intensive care unit, my ambition is to come up with a concrete concept for Clinical Decision Support that supports human-machine collaboration by facilitating the incorporation of facts and data-driven learnings provided by an AI algorithm into the human reasoning process. The idea is to provide support in the reasoning where machine reasoning surpasses human reasoning, while facilitating exploitation of human reasoning strengths such as holistic reasoning, flexibility and empathy.
The CDS concept should be value-aware, actively supporting the medical staff in considering tradeoffs in human values in the care of their patients. The envisioned concept uses a conversational AI approach to enable the CDS to become value-aware and story structures to help the user integrate facts and data-driven learnings provided by the CDS with their own value judgements in a natural way.
An important question that needs to be answered during the trajectory of this research is: How will we measure the success of a value-aware, conversational and narrative CDS? In order to know whether the decision-making process has improved, we need to understand the values that are involved, both the professional values of the physician as well as the personal values of the patient and to balance them with society’s values. We need to define some measure of balance among the utilitarian and the deontological perspective in medical decision-making.
A more in depth decription of my research proposal can be found in this Frontiers publication

Topics

  • Explainable AI
  • Moral technology
  • Clinical Decision Support
  • Human judgement and decision-making
  • Measuring the quality of the decision-making process
  • Qualitative research
  • Human technology interaction