Research Engineer to Develop Cutting-Edge Vision AI for Breast Radiology

Do you want to contribute to improving human health?

Do you want to contribute to reducing the number of late breast cancer diagnoses and improve survival for our patients? Do you want to develop production-grade visual deep learning models to predict the future risk of breast cancer, and the risk of having genetic risk for breast cancer, based on mammography screening images?

Our research is internationally recognized for the development and evaluation of AI for radiological breast cancer diagnostics with recent publications in Nature, Lancet Digital Health, Radiology, MICCAI, among others. Our group has expertise in radiology, biostatistics and machine learning. We have a strong collaboration with research groups at KTH, and internationally with UC Berkeley.

A transformational project

One in nine women will get diagnosed with breast cancer. For around 20-25% of those women, breast cancer will be their cause of death. By detecting cancer early, many deaths can be avoided. Genetic risk may explain up to around 25% of all breast cancer diagnosis, whilst 75% is not predictable based on genetics. We want to combine AI analysis of screening mammograms, which are performed with 2-year intervals for each woman between 40 and 74 years of age, with the time-constant information from genetics, to deliver precision breast cancer screening.

We were recently awarded funding for Early Detection of breast cancer from Cancerfonden, and our now hiring several people of which the research engineer is one. Our aim is to use our existing AI risk model for breast cancer, improve the robustness across images from various equipment, and to fine-tune to predict the risk of having high-risk genes such as BRCA1. We also want to improve the radiological surveillance strategy for each person who are confirmed of having one of the risk genes, by using both genetic and image information.

We have developed and clinically tested a groundbreaking AI solution at the Karolinska University Hospital to identify women at high risk of having breast cancer despite a negative screening mammogram (i.e., without apparent signs of cancer to the radiologists). Using our AI solution to select a small proportion of women for magnetic resonance imaging diagnosed three to four times more cancer compared to selecting women using traditional methods based on mammographic density.

We now need to extend the model to work with images from multiple mammography equipment types, and further train and fine-tune for the purposes listed above.

Your mission

You will work developing cutting-edge image-based deep learning models for precision screening and surveillance of breast cancer. Your tasks will involve preparing data, running experiments, and testing the results. Your mission will also involve ensuring a continuously updated, structured and well-documented code base. You will be located with our existing group at Karolinska in Solna, Sweden. You will also collaborate and co-locate with our collaborators at KTH, at the nearby SciLifeLab.

Your profile

 Qualifications

You should have:

• Master’s degree in computer science or related field
• A portfolio to demonstrate expertise in visual deep learning
• Experience from software engineering and code base maintenance
• Excellent communication skills in both writing and speaking in English.
• Excellent skills in programming and mathematics.
• Experience with deep learning frameworks, particularly PyTorch.

It is a merit if you have:

• Applied your expertise to medical images, especially radiology and mammography
• Applied your expertise in risk modelling
• Familiarity with medical imaging libraries (e.g., MONAI, TorchIO, SimpleITK)
• Experience from production-grade software engineering
• Academic awards, presentations, publications demonstrating high impact
• Excellent communication skills in both writing and speaking, in Swedish

What do we offer?

A creative and inspiring environment with wide-ranging expertise and interests. Karolinska Institutet is one of the world's leading medical universities. Our vision is to pursue the development of knowledge about life and to promote a better health for all. At Karolinska Institutet, we conduct successful medical research and hold the largest range of medical education in Sweden. Karolinska Institutet is also a state university, which entitles you to several good benefits through our collective agreement. And you get to practice freely in our modern wellness facilities, where trained staff are on site.

Location: Solna

Choose to work at KI – Ten reasons why

Application

Welcome to apply at the latest 22^nd December.

The application is to be submitted through the Varbi recruitment system.In this recruitment, you will apply with your CV without a personal letter. Instead, you will answer some questions about why you are applying for the job in the application form. 

The position is for 1 year, with a possible extension to a permanent position.

Want to make a difference? Join us and contribute to better health for all