PhD Position in Multimodal Machine Learning in Healthcare

The PhD researcher will be part of the eMedia research lab under the supervision of Prof. Bart Vanrumste. The research group is embedded in the Department of Electrical Engineering (ESAT) of KU Leuven, Europe's most innovative university [Reuters]. Prof. Vanrumste’s research focuses on multimodal sensor integration and machine learning for monitoring of older persons and patients with chronic diseases.

Project

Accurately recognizing human activities is an important requirement for various healthcare applications (e.g., to enable wearable robotics to proactively respond to the user’s activities). Deep learning has enabled promising results in various applications by automatically discovering complex representations from raw input data. Taking human activity recognition as an example; video contains a rich description of activity context but provides only a rough description of biological processes and articulated movement. Fusing video with wearable sensors such as inertial measurement units and (neuro)physiological sensors can address such shortcoming and hence improve decision making. However, further basic research is required as it remains challenging to model heterogeneous data sets such that the respective advantages are exploited, and the disadvantages are suppressed. In this context, you will study, among others:

• How to obtain activity context from video data;
• How to train the heterogenous data sources cooperatively;
• How to learn optimal fusion schemes rather than hand selecting modalities;
• How to optimize fusion schemes for (real-time) healthcare use-cases;

Profile

We are looking for a dynamic and motivated PhD candidate with a strong interest in AI, who is interested in studying how advancements in multimodal learning can lead to improved monitoring of older persons and patients with chronic diseases. The candidate will be responsible for research and development of advanced multimodal AI-pipelines and will be involved in the AidWear project, where the candidate will investigate whether fusion of video and wearable sensor data can lead to improved monitoring of older persons. The candidate will also contribute to teaching activities related to machine learning or other areas depending on the candidate’s profile. Moreover, the candidate is a team player that enjoys collaborating with people within the research group, the project, and beyond, and has:

• A master's degree in Engineering with a background in mechanical engineering, electrical engineering, computer science, AI, or related field, from a reputable institute, with outstanding study results,
• Programming experience in Python, particular experience in common deep learning frameworks (e.g., PyTorch and TensorFlow) would be a benefit,
• The qualities to carry out independent research, demonstrated e.g., by the grades obtained on your MSc thesis,
• An excellent command of the English language, both in spoken and written form,
• Is comfortable assisting in data collection experiments with participants in general but older ones in particular,
• A critical mindset.

Offer

•  A doctoral scholarship of four years and, if successful, a PhD in Engineering Technology,
•  A competitive salary and additional benefits such as health insurance, access to university sports facilities, etc.
•  The opportunity to be active in an exciting and international research environment, engage in research collaborations and participate at international conferences,
•  A full-time employment for four years, with an intermediate evaluation after each year,
•  An excellent doctoral training at the Arenberg Doctoral School in an international environment at a top European university, 
•  A flexible working culture with opportunity of up to 40% remote working.

Interested?

For more information please contact Dr. Benjamin Filtjens, tel.: +32 14 74 15 99, mail: benjamin.filtjens@kuleuven.be or Prof. dr. ir. Bart Vanrumste, tel.: +32 16 32 64 07, mail: bart.vanrumste@kuleuven.be.

KU Leuven strives for an inclusive, respectful and socially safe environment. We embrace diversity among individuals and groups as an asset. Open dialogue and differences in perspective are essential for an ambitious research and educational environment. In our commitment to equal opportunity, we recognize the consequences of historical inequalities. We do not accept any form of discrimination based on, but not limited to, gender identity and expression, sexual orientation, age, ethnic or national background, skin colour, religious and philosophical diversity, neurodivergence, employment disability, health, or socioeconomic status. For questions about accessibility or support offered, we are happy to assist you at this email address.