The candidate will be developing new Visual Analytical methods for the understanding of medical image analysis models for accurate segmentation and 3D reconstruction of coronary vessels and stents based on 2D X-rays. The work of the PhD should result in better models that will be more easily deployed in the clinic.
Artificial Intelligence (AI) has the potential to benefit Percutaneous Coronary Interventions (PCI) .
The accurate segmentation of coronary vessels from 2D X-ray runs is needed to quantitatively characterize the degree of stenosis, and selecting the appropriate coronary stents. Furthermore, it is important to examine the final result after coronary stent implantation. Recurrent symptoms and complications after PCI are often related to inadequate sizing and expansion of stents, leading to under-expansion and/or mal-apposition. Proper characterization of the stenosis, and its 3D reconstruction rather than using 2D views can provide relevant information. Similarly, a 3D model of the stent after deployment is essential to assess its success.
Despite data based models exist and commonly outperform traditional approaches in segmentation and 3D reconstruction tasks, they are prone to poor generalization and susceptibility to domain shifts. The black-box nature of the models makes model development inefficient and mainly based on a trial-and-error approach. Understanding of the internal workings of the model is missing also making the interpretation of the results difficult and potentially misleading. Visual analytics techniques that support the understanding, interpretation and presentation of the models are an essential component to provide the effective development and deployment of the models. Getting the necessary insight into the models may, therefore, become a key enabler for development and acceptance of coronary segmentation, and 3D reconstructions from 2D X-ray angiographic in clinical practice.
The project will be developed within the visualization cluster under the supervision of Prof. Anna Vilanova and Dr. Nicola Pezzotti (Philips- part time TU/e). This position is part of a MEDUSA project with collaborators from Philips, cardiology department of Catherina Hospital, Biomedical Technology and Electrical Engineering from TU/e.
This project is part of The Eindhoven MedTech Innovation Center ( e/MTIC). A public-private partnership that aims to create a fast track to high-tech-health innovations in the perinatal, cardiovascular and sleep fields maximizing value for patients.
The visualization cluster (https://research.tue.nl/en/organisations/visualization) at TU/e has a strong track record in visualization and visual analytics for ML models and high-dimensional data. It has generated several award winning contributions at major visualization conferences (IEEE VIS, IEEE InfoVis, IEEE VAST, EuroVis); several successful start-up companies (MagnaView, Process Gold and SynerScope); and a number of techniques that are used on a large scale world-wide.
We are looking for a candidate who meets the following requirements:
• You are enthusiastic about research in visual analytics, medical image analysis, and Explainable AI
• You have experience with or a strong background in medical visualization, visual analytics, medical image analysis and/or machine learning. Preferably you finished a master in Computer Science, (Applied) Mathematics, Biomedical Engineering or Electrical Engineering.
• Expertise in the fields of explainable AI is a plus but not mandatory.
• You have good communication skills and are able to work in a multidisciplinary team.
• You have strong programming skills.
• You are creative, critical, analytical, hardworking and persistent.
• You have a good command of the English language (knowledge of Dutch is not required).
Your application materials should include:
Please apply via the 'Apply now' button on this page. You can upload a max. of 5 documents of max. 2 MB.
For further information, please contact or consult: