Project Update: Deep Learning to extract biomarkers for the diagnosis and personalized treatment of neuropsychiatric disorders

February 2020 - Q&A with dr. Guido van Wingen

When did the project officially start?

In March 2019 the first researchers started the project.

Link to project page.

How many people are working on the project?

At the moment a postdoc, a PhD student, a principal researcher and assistant professor are working on this project.

How did you get started?

A proposal has been sent from the university to Philips specifically for this research project. This has resulted in a very good collaboration, Philips is a very suitable research partner for us for this project. They have the right expertise and invest in this field .

What is the biggest challenge so far?

Drawing up the contracts between the different partners that relate to the data and IP has cost a lot of different lawyers.

Is there a lot of contact with the private partners and what does the collaboration look like?

For this research we use data from the UK Biobank, a public database. This data makes new types of analyses possible. The data researchers use insights from the scientific literature to develop code to analyse medical images. The work mainly consists of developing code and conducting experiments to determine how we can detect patients with a particular disorder. Within this collaboration, Philips brings in specialists, contacts and practical knowledge.

Are the research questions still modified during the course of the research?

No. A new research question arose from a meeting of Commit2Data, but it is separate from this project.

In which phase of the research are you now?

The research has already made steps forward and results have been presented at a scientific congress and published in a scientific journal (https://doi.org/10.1007/978-3-030-32695-1_11) We have developed a new algorithm that produces good results. In this phase we test applications of this algorithm and develop new algorithms. For example, an algorithm that searches for a new classification of patients based on the data itself.

What is your ultimate goal?

Ultimate goal is to give personal medical advice based on neuro imaging data in the future.

20 February 2020

Newsletter

Sign up for our newsletter and stay informed of the latest news Commit2data