PhD/Postdoc positions in Big Data analysis for PRECISION PSYCHIATRY
PhD/Postdoc positions in Big Data analysis for PRECISION PSYCHIATRY: Leveraging large scale electronic health records (EHR) using multimodal learning on tabular, text and imaging data (Computer Science & Bioinformatics)
PhD/Postdoc position in machine learning on large-scale complex multimodal data to advance PRECISION PSYCHIATRY
The Copenhagen Research Center for Mental Health (CORE) is a leading Danish research center within the field of psychiatry with a remarkable research output and an extensive international network. Our main research focus is on prevention as well as improvement of symptoms and quality of life for people with severe mental disorders. The research group of Precision Psychiatry is currently innovating in the psychiatric field with quantitative approaches applying statistics and AI to medical and biological data. Thanks to the exceptional quality of nationwide Danish registers and population-based EHR, we are becoming a key reference worldwide.
About the project
The PRECISION PSYCHIATRY project brings a unique opportunity to leverage the Danish EHR, containing detailed large-scale clinical information, by utilizing multivariate statistics and machine learning including deep learning approaches. Data content includes both structured data (treatments, laboratory tests, etc.) as well as unstructured data (written clinical notes and brain scans).
This project aims to discover novel ground-breaking insights into mental disorders, identify clinical state markers and biomarkers for stratification of mental disorders and disease trajectories, as well as identify modifiable factors with prevention potential. The multivariate understanding of objective markers with clinical relevance is also needed for the clinician to get precise prediction estimates and to identify protective factors for mental disorders, which is now based on the clinicians’ intuition. All previous attempts towards more precision psychiatry have not been suited for implementation in clinical practice so far. Therefore, the field of psychiatry still has many unknowns, remaining as a challenge to which we believe this project will greatly contribute.
With the combination of exceptional large-scale data and individual AI models targeting the different data types, all later combined into a multimodal learning model, we intend to develop EHR prediction models with sufficient accuracy to warrant clinical implementation. The EHR markers can provide novel direct data-driven support for diagnostics, prevention, and treatment of mental disorders. If sufficient precision is obtained, this and potentially other models can be directly tested and implemented into the EHR system through decision support tools. And this will pave the way to actual clinical implementation that would enable the advancement of precision psychiatry to the next level.
The project is based on the EPIC EHR system, which is the most widely used EHR system in the world, securing international implementation of recent developments within precision psychiatry.
About the center and group
We offer an inspiring and informal research environment at a high international level with colleagues from different educational backgrounds, such as bioinformaticians, statisticians, doctors, project nurses, etc. The PI for the project is Professor Michael E. Benros, chief physician, PhD, and head of research for “Biological and Precision Psychiatry”, at the Mental Health Center Copenhagen, Copenhagen University Hospital and at the University of Copenhagen. Among other topics, the group has strong experience and on-going research on Big Data from the nationwide Danish registers and biobanks in addition to detailed clinical studies with multimodal data as wearables and multi-omics that we believe will be beneficial for the projects in both directions.
Moreover, the PRECISION PSYCHIATRY project will be developed with strong collaboration with co-PI Simon Rasmussen’s expert group in machine learning, Big Data and multimodal analyses for Disease Systems Biology at the University of Copenhagen (www.cpr.ku.dk), also in Copenhagen, as well as experts in AI-based analyses of brain imaging and EHR data together with other leading international groups with the same interests and private companies.
As a working culture, we are ambitious and result oriented, while also like enjoying some time together by organizing different types of social activities from time to time both at the research center and outside. You will be part of an internationally leading cross-disciplinary team working towards data-driven improvements of the current treatment of mental disorders.
About the positions
The offered positions are 3-year PhD and 2/3-year postdoc positions as part of the “Precision Psychiatry” research group in CORE that combine a strong level of technical knowledge within AI, data management and data analysis, as well as scientific interest in the clinical implementation. These are planned to be dynamic positions for applied research that can adapt to each profile.
The positions fall within a collaborative multi-disciplinary project that will involve other positions, including computer scientists, bioinformaticians, medical doctors and statisticians. It is thus required for the applicants to have the ability to work and communicate together with clinical researchers and researchers with other backgrounds along the project. Your day-to-day tasks will involve both independent work and active discussions with your colleagues.
Supervisors: Professor, Head of Research, Michael Eriksen Benros, Biological and Precision Psychiatry, Mental Health Center Copenhagen, Copenhagen University Hospital and University of Copenhagen; and Associate Professor, Group Leader, Simon Rasmussen, Disease Systems Biology, University of Copenhagen.
Start Date: As soon as possible
- A master’s degree or Ph.D. degree in Computer Science, Bioinformatics, or related fields
- Experience with big data management and data processing techniques (mainly Python)
- Strong knowledge in machine learning and deep learning
- Understanding of NLP (Natural Language Processing) & transformers, and/or image processing, for the unstructured data
- Scientific interest for the understanding and explainability of AI model predictions, to ensure a clinical implementation by providing a funded trust in the results to clinicians
- Strong communication and collaboration skills within a dynamic multi-disciplinary workspace
- Proven research expertise in NLP projects, both data preparation and model development
- Image processing experience of brain scans or similar within the medical field
- Previous experience with Azure ML or cloud computing systems
- A profound interest in our vision of improving the current treatments of mental disorders
Application deadline: 15th February 2023
- Cover letter (1 page in length)
- Transcript of MSc or of doctoral studies as well as corresponding degree certificate
- List of publications (if available)
- Contact details of possible referees from 1-2 people. We will contact your referees if required.
All materials should be submitted in a PDF format using the link on this webpage.
Please write in your application that you've seen the job at Jobfinder