PhD Machine learning on large databases for interventional cardiac PCI treatment

Project description:

Background and motivation

The time delay between medically relevant innovations and their establishment in medical practice is significant. Reducing this lag is a constant challenge for the health care system. Already today, a large amount of data is generated and stored daily in routine clinical practice and research. If we succeed in integrating and evaluating this comprehensive digital documentation of medical data, this promises a far-reaching gain in knowledge for medical science as well as an immediate benefit for those affected or at risk in the future. New technologies make it possible to link routine clinical parameters with the human genome, the activity of human genes in tissue (transcriptome and proteome), or metabolites in the bloodstream (metabolome) in functional networks. The interpretation of these data requires a reference to diseased as well as phenotypically healthy persons, whereby knowledge is used from clinical and epidemiological studies, from health insurance companies, and finally from the individual (real world data, e.g., on diet, exercise, self-perception). From a comprehensive digital view of medical and medically relevant data (Big Data), a new understanding of the development and treatment options of diseases will emerge when the data is appropriately combined and evaluated. The use of digital data and the storage of process parameters in routine clinical practice represent major and central challenges in modern medicine. Parallel to this, an innovative interpretation of coordinated changes in these parameters in the human organism as a whole, is important in order to better understand disease development or progression. This means that a first step into the medicine of the future consists of the transformation of an information-limited static (actual state: diagnostic findings) into a dynamized, directional process (causality) by means of extensive constant data collection, linkage and analysis. Digital medicine thus promises to be a key to better understanding (extended disease taxonomy to the individual) and more targeted prevention of diseases.

Methodology

To achieve linking individual omics data with medical studies, for the first time, an integrative analysis of high-throughput methods (omics) and cardiac catheter images is aimed at combining functional analyses based on genome-wide significant risk alleles and their influence on the circulating proteome level with atherosclerotic (morphological) changes in human coronary arteries. The German Heart Center Munich, which is one of our partners in this project, treats >20,000 patients annually and hosts one of the most comprehensive scientific infrastructures in the field of cardiovascular medicine worldwide. This includes, for example, more than 175,000 individual cardiac catheterizations, genetic analyses of more than 50,000 patients, and comprehensive clinical databases with standardized laboratory values, ECG and echocardiography data. Mortality follow-up was performed over a period of 17 years. This medical use case provides, by integration of large heterogeneous datasets and the use of AI based algorithms, the potential to identify disease relevant signatures (patterns) in medicine. This directly improves prediction, prevention and treatment of patients. Clustering of this data allow for example to identify patients at highest risk for secondary cardiovascular events and pave the way for AI supported decision making - aiming at personalized medicine. Further, there is also relevant economic interest in the identification of these clusters, as patients at risk come along with high economic burden for the overall health system. In summary, the models cover three major advantages. By correlating relevant clinical patient clusters with on-the-fly extractable biomarkers, risk stratification, procedure approach and follow-up regime can be determined.

Project goals

• Analyze large database (more than 100,000 individual cardiac catheterizations), angiograms, genetic analyses, and comprehensive clinical databases with standardized laboratory values, ECG and echocardiography data, using data analysis and data clustering, machine learning, unsupervised learning, etc.
• Detect key events in real-time during PCI treatment: angiogram, balloon inflation, stent deployment.
• In angiograms: automatically detect left and right tree, automatically label vessels, detect stenosis.
• For detected stenosis in angiograms: auto-determine best frame, delineate, quantify, identify calcified plaque.
• For angiograms from multiple angles: quantify and compare model to intra-vascular data (IVUS, OCT).
• Develop feature vectors describing relevant attributes extracted from raw data sequences (images, ECG, oxygen, etc).
• Develop machine learning algorithms to extract these feature vectors from unstructured real-time interventional data (e.g. image sequences).
• Develop algorithms for real-time lookup in the clustered ontology database.

Electronic Systems group at TU/e and Philips IGT-S

The Electronic Systems group consists of six full professors, three associate professors, five assistant professors, several postdocs, about 40 EngD and PhD candidates and support staff. The ES group is world-renowned for its design automation and embedded systems research. It is our ambition to provide a scientific basis for design trajectories of electronic systems, ranging from digital circuits to cyber-physical systems. The trajectories are constructive and lead to high quality, cost-effective systems with predictable properties (functionality, timing, reliability, power dissipation, and cost).

Philips Image Guided Therapy is a business category within Royal Philips. The Image Guided Therapy department focuses on developing innovative medical devices and solutions that enable minimally invasive and image-guided procedures in the field of healthcare. These technologies are designed to help healthcare professionals perform procedures such as interventional cardiology, radiology, and surgery with greater precision and efficiency. Philips Image Guided Therapy solutions often incorporate advanced imaging, data analytics, and visualization technologies to enhance patient care and improve clinical outcomes. https://www.philips.com/healthcare/e/image-guided-therapy.

Requirements:

We are looking for excellent candidates that add value to the ES group and match the following profile:

• A master's degree in Electrical Engineering, Biomedical Engineering or related disciplines with excellent grades.
• A team player that enjoys working in multicultural teams.
• Good communication and organization skills.
• Excellent English language skills (writing and presenting).

Salary Benefits:

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, scale P (min. €2,901 max. €3,707).
• 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.

38 hours per week

De Rondom 70