We advertise two three-year doctoral positions and one two-year postdoctoral fellow position in machine learning and data mining for personalized cancer screening at SimulaMet and Cancer Registry of Norway, Oslo, Norway. The positions are available as a part of the Norwegian-funded research project DeCipher in collaboration between SimulaMet, Cancer Registry of Norway, Lawrence Livermore National Laboratory (LLNL), and Karolinska University Hospital. The project combines the unique Norwegian and Swedish registry data including extensive screening histories and epidemiological expertise, with LLNL’s exceptional HPC computing infrastructure, data analysis and modelling experience, and with Simula’s expertise in machine learning methodology.

Project Description

Cancer is a major cause of morbidity and mortality worldwide. A large proportion of these incidents are preventable. For example, a mass-screening program against cervical cancer established in the Nordic countries has demonstrated a reduction in morbidity and mortality almost by 80%. Despite this success, it remains a major challenge to improve the screening program, such as minimize overscreening and undertreatment, and hence reduce expenditure in a broad public health perspective.

Current knowledge about the disease together with a wealth of available data and modern technologies can offer far better personalized prevention, by deriving an individual-based time till the next screening.

By intelligent use of existing registries and health data, the DeCipher project aims to develop a data-driven framework to provide a personalized time-varying risk assessment for cancer initiation and identify subgroups of individuals and factors leading to similar disease progression. Specifically, the following three sub-projects will be investigated:

1. Development of matrix factorization approaches for reconstruction of individualized disease trajectories from longitudinal screening data (PostDoc at SimulaMet)
2. Development of geometric deep learning framework for reconstruction of individualized disease trajectories from longitudinal screening data (PhD fellow at Cancer Registry)
3. Development of data fusion approaches based on matrix and tensor factorizations for joint analysis of longitudinal screening data and lifestyle information as well as other relevant health data (PhD fellow at SimulaMet)

The developed methods will be validated and tested on Norwegian and Swedish datasets. The DeCipher results will allow for improvement of individual’s preventive cancer healthcare while reducing the cost of screening programs.

The successful candidates are expected to develop a set of machine learning and data mining approaches to unravel patterns in the cancer screening and health data and learn discriminating features. The successful candidates will enjoy an inspiring and resourceful environment, with the possibility of travelling to conferences, short- and long-term research visits to LLNL, and establishing new collaborations.

Candidate Profile

PostDoc Fellow at SimulaMet for Development of Matrix Factorization Approaches. We consider interested candidates who obtained their PhD in applied mathematics or a related discipline (computer science, statistics, signal processing), preferentially in the medical/biology/public health domain, within the last five years. Required qualifications include

• strong background in optimization, machine learning, signal processing and numerical linear algebra proved by strong conference and journal publications,
• experience in developing novel theoretically grounded machine learning methods and applying them to real-life problems,
• previous experience with matrix factorization is a plus,
• ability to carry interdisciplinary research,
• fluency in MATLAB and/or Python.

Contact person: Valeriya Naumova (valeriya@simula.no)

PhD Fellow at Cancer Registry for Development of Geometric Deep Learning Approaches. We consider interested candidates who have a BS and MS degree in computer science, applied mathematics, statistics, electrical engineering or other related discipline, with top grades. We prefer a candidate with interest or documented experience in the medical/biology/public health domain. Required qualifications include

• strong background in optimization, signal processing, applied harmonic analysis, and/or machine learning,
• an average GPA of at least B in Bachelor and A in Masters,
• willingness to carry out interdisciplinary research,
• fluency in MATLAB and Python,
• familiarity with Tensorflow/Keras libraries is a plus.

Contact person: Jan Nygård (Jan.Nygard@kreftregisteret.no)

PhD Fellow at SimulaMet for Development of Data Fusion Approaches. We consider interested candidates with BS and MS degrees in computer science or a related discipline (applied mathematics, statistics, signal processing), with top grades. We prefer a candidate with interest or documented experience in the medical/biology/public health domain. The successful candidate will join a group working on data fusion methods and will have a specific focus on cancer screening as an application. Required qualifications include

• strong background in numerical linear algebra, optimization, and machine learning/data mining proven by the grades and/or conference/journal publications,
• an average GPA of at least B in Bachelor and A in Masters,
• willingness to carry out interdisciplinary research,
• fluency in MATLAB and/or Python,
• previous experience with matrix/tensor factorizations is a plus,

Contact person: Evrim Acar (evrim@simula.no)

The candidates will also have to demonstrate an excellent level of spoken and written English, possess good interpersonal and communication skills and show willingness to work as part of an international team. Both hiring institutions strive to achieve a good balance between male and female employees, and women are particularly encouraged to apply.

Application requirements

Candidates are requested to upload a motivation letter, CV including a list of publications, and contact information of three references (including the main supervisor). Postdoc applications should include a research statement (max. two pages). PhD student applicants should include their transcripts. After initial screening, relevant candidates will be asked to provide further documentation of education and work experience.Applications can only be submitted via our online system, applications submitted by email will not be considered.

The expected starting date is October 01, 2019. Applications deadline is August 15, 2019. However, applications will be screened continuously and a decision will be made as soon as we find the right candidate.

We offer

• An engaging and interactive working environment with highly competent and motivated researchers.
• An informal and inclusive international working environment.
• High-standard facilities and offices located in Oslo.
• Professional courses and workshops led by international experts such as Communication of Scientific Research, Innovation and Entrepreneurship, and Writing Effective Research Proposals.
• Access to cutting-edge HPC infrastructure.
• A competitive salary.

The advertisement is subject to approval by the Research Council Norway.

Simula Metropolitan Center for Digital Engineering uses Semac´s background check in our recruitment process.According to the Norwegian Freedom and Information Act (Offentleglova) information about the applicant may be included in the public applicant list, also in cases where the applicant has requested non-disclosure.

Apply for position