1. Vacatures
  2. Technische Universiteit Eindhoven
  3. POSTDOC POSITION Self-driving ultrasound imaging

Helaas, deze vacature staat inmiddels niet meer online

Kijk gerust verder naar andere vacatures.

POSTDOC POSITION Self-driving ultrasound imaging

PostDoc positionSelf-driving ultrasound imaging for adaptive intra-cardiac tissue characterization.Keywords: Ultrasound imaging, reinforcement learning, …

3 maanden geleden


de Rondom, Eindhoven, Noord-Brabant
Tijdelijk contract / Tijdelijke opdracht
Uren per week:
38 uur


PostDoc position

Self-driving ultrasound imaging for adaptive intra-cardiac tissue characterization.

Keywords: Ultrasound imaging, reinforcement learning, beamforming, deep learning, cardiac ablation.

Research Summary

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in humans, causing several complications, including stroke and death. Treatment costs are estimated at 26 B$/year in the USA and over 637 M€/year in the Netherlands alone. Catheter ablation to create pulmonary vein isolation has become the standard therapy for AF. However, optimal and safe ablation is hampered by the lack of tools for detecting fibrosis or scar tissue, and for assessing the quality of the ablation lesions. As a result, one-year single procedural success rates are disappointingly low (60%). Parallel to the growing need for a safe and cost-effective tool to monitor and assist cardiac ablation, ultrasound transducer technology has advanced into miniaturized implementations, opening up new diagnostic imaging opportunities that are cost-effective and perfectly suitable for intra-procedural use.

In this project we will devise a new imaging paradigm, termed 'self-driving ultrasound', in which the ultrasound transmit driving schemes and beamforming algorithms are continuously adapted in a reinforcement learning framework to provide at any time the optimal tradeoff between various imaging properties (e.g. frame rate and resolution), disturbing artefacts, and system-level constraints on power and data-rates. Novel multi-parametric receive signal-processing approaches will be employed to enable quantification of local tissue properties relevant for ablation monitoring (e.g. viscoelasticity and vascular perfusion). The technology will be implemented on existing 2D intra-cardiac probes with a versatile ultrasound system.

Research embedding

The project is a collaboration between the Biomedical Diagnostics lab (BM/d) of the Eindhoven University of Technology (TU/e), Philips Research, and the Catharina Hospital Eindhoven. The project team is designed to combine extensive knowledge and expertise in ultrasound system design, signal analysis, modelling, implementation and interventional cardiology, with an elaborate clinical infrastructure. There is a long-standing collaboration between these partners, as part of a research consortium called the Eindhoven MedTech Innovation Center (E/MTIC), focusing on topics such as peri-operative monitoring and care, ultrasound-guided interventions, and physiology- and data-driven decision support models.

The Signal Processing Systems (SPS) group of the Electrical Engineering Department of the TU/e, chaired by Prof. Jan W.M. Bergmans, has a strong and longstanding international academic reputation in physiological monitoring of cardiac activity and ultrasound imaging as evidenced by the numerous international publications and awards. Prof. Massimo Mischi, coordinator of the Biomedical Diagnostic (BM/d) Lab within the SPS Group (www.bmdresearch.nl), has a strong track record in quantitative ultrasound imaging and biomedical signal processing.


Job description and requirements

In this context, we are seeking a highly motivated post-doctoral researcher with knowledge and interest in image acquisition physics (preferably ultrasound) and deep learning (specifically reinforcement learning). The position is available within the BM/d research lab, part of the Signal Processing Systems (SPS) group (Electrical Engineering department, TU/e), and it involves a strong collaboration within the E/MTIC consortium, i.e. with Philips Research and the Catharina Hospital.

We are looking for candidates that meet the following requirements:
  • PhD in Physics, Electrical Engineering, or Computer Science;
  • Experience with medical image acquisition (preferably ultrasound beamforming);
  • Experience with machine/deep learning (preferably reinforcement learning);
  • Strong background in (adaptive) signal processing and optimization;
  • Good communicative skills in English, both in speaking and in writing;

  • The Postdoc is expected to:
  • Perform scientific research in the domain described;
  • Collaborate with other researchers in the project team (industry, hospital, PhDs);
  • Co-supervise PhD candidates in the project team;
  • Publish results in leading scientific journals and conferences in the field;

  • Arbeidsvoorwaarden

    We offer:
  • full-time employment as a Postdoc for a period of 4 years, with an intermediate evaluation after one year.
  • annually 8% holiday allowance and 8.3% end of year allowance;
  • support with your personal development and career planning including courses, conference visits etc.;
  • a broad package of fringe benefits (including an excellent technical infrastructure, child care, moving expenses, savings schemes and excellent sports facilities).

  • More information on employment conditions can be found here: www.tue.nl/en/university/working-at-tue/working-conditions/ .

    Additionele informatie


    For more information about this position contact dr. ir. Ruud van Sloun, e-mail: r.j.g.v.sloun[at]tue.nl .

    More information on employment conditions can be found here: www.tue.nl/en/university/working-at-tue/working-conditions/ .


    If interested, please use the 'apply now'-button. You should upload the following:

    The application should consist of the following parts:
  • Cover letter explaining your motivation and qualifications for the position, as well as your current experience with and understanding of the topic
  • Detailed Curriculum Vitae;
  • List of key publications;
  • A copy or a link to your PhD thesis. If you have not completed it yet, please explain your current situation.
  • Names of at least two references.

  • The selection process will start in March 2019 and will continue until the position gets filled. The position is fully funded and immediately available. The successful candidates are expected to start ASAP.