Postdoctoral fellowship (2 years) within Generative Modelling in Machine Learning
The Department of Computing science is offering a postdoctoral scholarship to develop generative machine learning models within the project “AI in Medical Imaging”. The scholarship is full-time for two years with access November 1, 2024, or by agreement.
The Department of Computing science is characterized by world-leading research in several scientific fields and a multitude of educations ranked highly in international comparison. The department has been growing rapidly in recent years where focus on an inclusive and bottom-up driven environment are key elements in our sustainable growth. A fellowship is offered in the project “AI in Medical Image Analysis”, focusing on the development and application of generative machine learning models for the analysis of medical images. Our staff represents a diversity of nationalities, backgrounds, and research areas. As a postdoc with us, you will be supported with career planning, networking, administrative and technical functions. See more information at: https://www.umu.se/en/department-of-computing-science/
Our Machine Learning research group, led by Associate Professor Tommy Löfstedt, has many years of experience in machine learning and is engaged in method and theory development, often with applications in e.g. automated radiotherapy and life science. Currently, this includes, for example, structured regularization in machine learning and resource efficiency in deep learning. The group is growing rapidly and has many ongoing collaborations with both national and international research groups and has active contacts with both the private and public sectors. For more information see: https://www.umu.se/en/research/groups/machine-learning/
Is this interesting for you? Welcome with your application latest October 14, 2024.
Project description
Automation tools for imaging applications that use the latest large-scale machine learning methods have gone through a remarkable development in the last few years and perform impressively in an increasing number of tasks. However, these methods lack robust generalization properties, mainly because of the small amounts of images available to develop them. For instance, the amounts of images typically available are not sufficient for clinical implementation, and therefore limits the medical applications that can be considered and also severely hampers the trust in clinical use of such automated decision support systems. The goal of this project is therefore to solve the ever-prevalent problem in imaging applications of not having enough images to develop powerful large-scale machine learning methods. We will develop novel generative machine learning models to synthesise custom high-resolution and high-quality medical images, that can be used to train downstream applied machine learning models.
The main project aim is to develop novel generative machine learning models, which may include multi-modal image generation, conditional synthesis, and design, exploration, and modelling of latent spaces. The project is a collaboration between researchers at the Department of Computing Science and the Department of Radiation Sciences at Umeå University, and the intended application is in image synthesis for automated radiotherapy.
The scholarship project is financed by the Kempe foundations and is for two years with a starting date as agreed. The scholarship amounts to about 350 000 SEK (around 31 350 euro) per year. The scholarship is not subject to taxes.
Qualifications
To qualify as a post-doctoral scholarship holder, the postdoctoral fellow is required to have completed a doctoral degree or a foreign degree deemed equivalent to a doctoral degree. This qualification requirements must be fulfilled no later than at the time of the decision about scholarship recipient.
Priority should be given to candidates who completed their doctoral degree, according to what is stipulated in the paragraph above, no later than three years prior. If there are special reasons, candidates who completed their doctoral degree prior to that may also be eligible. Special reasons include absence due to illness, parental leave, appointments of trust in trade union organizations, military service, or similar circumstances, as well as clinical practice or other forms of appointment/assignment relevant to the subject area.
Candidates must have very good knowledge in project related areas (machine learning, computer vision, etc.), and in particular to have a very strong documented experience in:
Machine learning
Deep learning.
A very good ability to both write and speak English fluently is also required. Good programming skills are also required to be awarded the scholarship.
Meritorious:
Generative modelling
Medical imaging
Computer vision
Image processing and analysis
Numerical methods
Optimization theory.
As a person, you are a good communicator, able to act both independently and in interaction with others, learn new things quickly and can put them in the right context. You are creative and motivated to actively drive your own development towards becoming a competent researcher.
Application
A full application should include:
A cover letter including a description of your research interests, your reasons to apply for the position. The letter should include a description of how you fulfill the specified requirements, and should also contain your contact information.
Curriculum vitae (CV) with publication list.
A verified copy of your doctoral degree certificate, or documentation that clarifies when the degree of doctor is expected to be obtained.
Verified copies of other diplomas (for both BSc and MSc, when applicable), and a list of completed academic courses and grades.
Copy of doctoral thesis and up to three of your relevant publications.
Other documents that the applicant wishes to claim.
Contact information to two persons willing to act as references.
The application must be written in English or Swedish. Your complete application, marked with reference number FS 2.1.6-1545-24, should be sent electronically (in PDF format) to medel@diarie.umu.se (with the reference number on the subject line). The closing date is October 14, 2024.
Further details are provided by associate professor Tommy Löfstedt (tommy.lofstedt@umu.se).