Are you fascinated by signal processing and artificial intelligence (AI) for healthcare? Are you interested in developing smart algorithms for new sleep monitoring technologies and exploring how AI can be used to go from sensor-specific to sensor-agnostic models? Join us to work on reliable and sensor-agnostic algorithms for long-term unobtrusive monitoring of sleep and sleep-disordered breathing conditions.

Job Description

Sleep disorders are highly prevalent and have severe consequences for health and wellbeing. Prominent examples are sleep-disordered breathing conditions, most notably obstructive sleep apnea (OSA), in which sleep is disrupted through repetitive reductions in respiratory flow. Due to insufficient tools for case finding and diagnosis, an OSA diagnosis may come late however. Moreover, current sleep diagnostic tools are too obtrusive for long term monitoring of treatment effect and outcome.

The present gold standard for sleep diagnostics is polysomnography (PSG), which combines multiple electrodes to assess various physiological parameters during sleep. In recent years, advances have been made in the analysis of cardiorespiratory signals such as heart rate and breathing rate to monitor sleep, which can be obtained in a much less intrusive way, for example with wrist-worn photoplethysmography (PPG). This sensor modality can also be used to estimate OSA-parameters such as the apnea-hypopnea index (AHI), but performance is not optimal yet. In addition, AI models developed for one sensor often need to be adapted or retrained for a different sensor modality.

This project aims to make significant improvements in AI models to perform automatic sleep staging and OSA monitoring using advanced unobtrusive sensors including, but not limited to, wrist-worn PPG (wPPG) and remote PPG (rPPG) with camera devices. The focus of the project is on improving the accuracy, efficiency, and generalizability of AI models across sensor modalities and/or datasets. Starting from models previously developed by Philips and the Advanced Sleep Monitoring group at the TU/e, the PhD will further improve the diagnostic performance of PPG-based sleep and OSA monitoring, leveraging the large SOMNIA dataset at the Sleep Center Kempenhaeghe, comprising over 1,500 annotated gold standard PSG recordings together with high quality wPPG.

Key tasks include: verification of PPG-based sleep staging algorithms and software as the backbone for OSA monitoring. Improving unobtrusive OSA monitoring and characterization using advanced reliable AI algorithms such as knowledge transfer, contrastive representation learning or active learning. Multi-task modelling for sleep staging and OSA assessment; and sensor-agnostic analysis and modelling with measurements from various input devices.

The PhD trajectory is part of the 'Medical Innovation and Research Advancing Clinical Learning and Excellence (MIRACLE)' project, a large research effort from the Eindhoven MedTech Innovation Center (e/MTIC), including 11 parallel projects. e/MTIC combines an academic partner (TU Eindhoven) with industrial partners and semi-academic hospitals: Máxima Medical Center, Catharina Hospital, and Kempenhaeghe. In particular, this position will be embedded in the Advanced Sleep Monitoring group (Biomedical Diagnostics lab, Signal Processing Systems, Department of Electrical Engineering, TU/e) in close collaboration with the Sleep Medicine Center Kempenhaeghe and Philips. As a result, temporary relocation at the partners' sites (Kempenhaeghe in Heeze and Philips Integrated Technology Solutions in High Tech Campus Eindhoven) will be considered to facilitate the project progress in its different phases.

The envisioned research will not only aim at furthering the knowledge in the field, but it is expected to make an impact in the clinic during the duration of the project. To this end, engineers, data scientist and clinicians will be part of the supervisory team and work closely with the student to support the development of new models and support the fast translation of the envisioned research to a clinical setting.

• A master's degree (or an equivalent university degree) in Electrical Engineering, Computer Science, Biomedical Engineering or related fields.
• Strong background or experience in signal processing and machine learning (deep learning).
• Excellent skills in programming languages such as Python.
• A research-oriented attitude with proactive mindset.
• Strong ability to work and interact in an interdisciplinary team and interested in collaborating with clinical and industrial stakeholders.
• Motivated to develop your teaching skills and coach students.
• Fluent in spoken and written English (C1 level).
• Ability to convey results in a clear and rigorous way, and interest in publishing in top international journals and/or conference proceedings.
• Affinity with clinical topics.

A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you:

• Full-time employment for four years, with an intermediate evaluation (go/no-go) after nine months. You will spend 10% of your employment on teaching tasks.
• Salary and benefits (such as a pension scheme, paid pregnancy and maternity leave, partially paid parental leave) in accordance with the Collective Labour Agreement for Dutch Universities: €2,901 (in year 1) up to €3,707 (in year 4).
• A year-end bonus of 8.3% and annual vacation pay of 8%.
• High-quality training programs and other support to grow into a self-aware, autonomous scientific researcher. At TU/e we challenge you to take charge of your own learning process.
• An excellent technical infrastructure, on-campus children's day care and sports facilities.
• An allowance for commuting, working from home and internet costs.
• A Staff Immigration Team and a tax compensation scheme (the 30% facility) for international candidates.