PhD scholarship in 3D Image Segmentation

Monday 11 Jan 21

Apply for this job

Apply no later than 15 February 2021
Apply for the job at DTU Compute by completing the following form.

Apply online

The section for Image Analysis and Computer Graphics at DTU’s Department for Applied Mathematics and Computer Science (DTU Compute) invites applications for a 3-year PhD position starting in early 2021. This PhD project aims on developing methods for segmentation of large 3D images. The position is part of SOLID (solid.dtu.dk) which is a cross-disciplinary center that focuses on method development and scientific studies within materials science, geoscience, bone research, forensics, and archeology. Research in the center is based on 3D imaging using the world most powerful neutron and X-ray sources, ESS in Lund and ESRF in Grenoble, the two national X-ray infrastructures, DanMAX at MAX IV and DANFIX at DTU, as well as the regional data analysis center QIM (qim.dk). The PhD student will be a part of a larger team of young researchers, and during the next year, we plan to hire eight PhDs and two postdocs in SOLID. Furthermore, the PhD student will be part of the QIM team working on a broad set of image analysis tasks.

In the project you will investigate the state of the art methods for interactive segmentation that allow segmenting large volumetric images. Segmenting photographs and image segmentation for medical applications has for a while been attracting a lot of attention. But interactive and versatile methods for segmenting 3D volumes are still lacking. The volumetric images are characterized by being large in number of voxels and the depicted structures are typically special for the imaged material. This sets special requirements to the tools needed, and carefully considering which new findings from e.g. computer vision can be made applicable for 3D images.

Responsibilities and qualifications
One of the main goals when studying materials based on 3D X-ray or neutron CT imaging is measuring the size and shape of the depicted structures. The typical analysis pipeline starts with segmenting out relevant structures which allows a statistical quantification of the imaged structures. In many cases, the segmentation is difficult, because structures are complex and not easy to separate. Due to the large size of the images, automated methods are required, and typically, advanced analysis methods are used.

There are many segmentation techniques, that require some supervision to guide segmentation. An important example is segmentation using machine learning based on convolutional neural networks, which requires labelled training images. Since X-ray images of materials microstructure are non-standard, there is no existing training data, and the labelling of such data is time consuming. It might, however, not be necessary to have fully annotated training data, and a sparsely labelled data set can be equally efficient. Another example is graph-based segmentation, that rellies on cost functions and initializations. These cost functions are typically defined by a user but could instead be learned from data.

In this project, you will investigate and develop methods for interactive segmentation. A part of this will be to investigate strategies for providing input to 3D segmentation algorithms with the aim of speeding up the segmentation process. The segmentation methods will be used by other researchers in SOLID and elsewhere, and the development will be based on specific use-cases. In particular, we plan to use images of bone as one example.

You will be enrolled in the DTU Compute’s PhD school. The duration of the PhD project is three years, where the primary task is research in interactive segmentation methods, but the project also involves teaching, course work, and other tasks. We expect you to:

  • Be part of a team for investigating and developing new methods for interactive 3D image segmentation.
  • Engage in collaboration with SOLID partners on specific use cases.
  • Contribute with new solutions to enable fast and accurate segmentation.

You should have a two-year Master's degree (120 ECTS points) or a similar degree with an academic level equivalent to a two-year Master's degree. We expect that you have studied topics in computer science and mathematics providing experience in image analysis, computer vision, machine learning, etc. Furthermore, the ability to program in e.g. Python, Matlab, or C++ is important. Also, the ability to work in a multidisciplinary environment is essential, as is a good command of the English language. 

Approval and Enrolment
The scholarship for the PhD degree is subject to academic approval, and the candidate will be enrolled in one of the general degree programmes at DTU. For information about our enrolment requirements and the general planning of the PhD study programme, please see the DTU PhD Guide.

Assessment
The assessment of the applicants will be made by Professor MSO Anders Bjorholm Dahl, Associate professor Vedrana Andersen Dahl and Associate professor Chiara Villa, Copenhagen University.

We offer
DTU is a leading technical university globally recognized for the excellence of its research, education, innovation and scientific advice. We offer a rewarding and challenging job in an international environment. We strive for academic excellence in an environment characterized by collegial respect and academic freedom tempered by responsibility.


Salary and appointment terms
The appointment will be based on the collective agreement with the Danish Confederation of Professional Associations. The allowance will be agreed upon with the relevant union. The period of employment is 3 years starting early 2021.

You can read more about career paths at DTU here

Further information
Further information may be obtained from Anders Bjorholm Dahl, tel.: +45 5189 6913.

You can read more about DTU Compute at www.compute.dtu.dk/english

If you are applying from abroad, you may find useful information on working in Denmark and at DTU at DTU – Moving to Denmark

Application
Your complete online application must be submitted no later than 15 February 2021 (Danish time)Applications must be submitted as one PDF file containing all materials to be given consideration. To apply, please open the link "Apply online", fill out the online application form, and attach all your materials in English in one PDF file. The file must include:

  • A letter motivating the application (cover letter)
  • Curriculum vitae
  • Grade transcripts and BSc/MSc diploma
  • Excel sheet with translation of grades to the Danish grading system (see guidelines and Excel spreadsheet here)

You may apply prior to ob­tai­ning their master's degree but cannot begin before having received it.

All interested candidates irrespective of age, gender, race, disability, religion or ethnic background are encouraged to apply.

DTU Compute
DTU Compute is a unique and internationally recognized academic environment spanning the science disciplines mathematics, statistics, computer science, and engineering. We conduct research, teaching and innovation of high international standard - producing new knowledge and technology-based solutions to societal challenges. We have a long-term involvement in applied and interdisciplinary research, big data and data science, artificial intelligence (AI), internet of things (IoT), smart and secure societies, smart manufacturing, and life science.

Technology for people
DTU develops technology for people. With our international elite research and study programmes, we are helping to create a better world and to solve the global challenges formulated in the UN’s 17 Sustainable Development Goals. Hans Christian Ørsted founded DTU in 1829 with a clear vision to develop and create value using science and engineering to benefit society. That vision lives on today. DTU has 12,000 students and 6,000 employees. We work in an international atmosphere and have an inclusive, evolving, and informal working environment. Our main campus is in Kgs. Lyngby north of Copenhagen and we have campuses in Roskilde and Ballerup and in Sisimiut in Greenland.