The Predictive Medicine group focus on medical research that is driven by statistical machine learning, artificial intelligence and big data. Currently our main focus is on computational pathology.

Mattias Rantalainen

Lecturer senior

The Predictive Medicine group focus on medical research that is driven by statistical machine learning, artificial intelligence and big data. Currently our main focus is on computational pathology.

About me

I am an associate professor (docent) at the Department of Medical Epidemiology and Biostatistics where I lead the Predictive Medicine group. 

I was previously a research fellow (2009-2013) with a joint affiliation at the Department of Statistics (University of Oxford) and at the Wellcome Trust Centre for Human Genetics (University of Oxford) where I worked with Professor Chris Holmes and Dr. Cecilia Lindgren. I was awarded a MRC Centenary Early Career Award (2012-2013) and a Medical Research Council (MRC) Special Training Fellowship in Biomedical Informatics (2009-2012). I had a postdoctoral position working on the data analysis work package for the MolPAGE consortium in Prof Chris Holmes’ group at the Department of Statistics in Oxford (2008).  I completed my PhD at Imperial College London where I developed novel multivariate pattern recognition methods with applications in metabonomics, working together with Professor Elaine Holmes and Professor Jeremy Nicholson.  I have an undergraduate degree in Engineering Biology (combined BSc/MSc) from Umeå University in Sweden.

Research description

The Predictive Medicine group focus on medical research that is driven by statistical machine learning, artificial intelligence (AI) and large population representative data sets. Currently our research is centered on projects in the computational pathology domain.

We develop and apply statistical and machine learning methodologies for predictive modelling in biomedical applications with a particular interest in precision medicine and cancer research.

Our mission is to drive quantitative approaches to cancer precision medicine and to develop and validate novel patient stratification models, including prognostic and treatment predictive applications. To achieve this we develop methods and models that allow us to transform large biomedical data into clinically relevant predictions at the individual level. 

Our research is based on large datasets (big-data) across multiple modalities including comprehensive molecular profiling (e.g. DNA- and RNA-sequencing), clinical information and medical imaging data (histopathology).

Main research areas

  • Computational pathology: AI (deep learning) driven research focusing on developing and validating models in the domain of cancer histopathology. The research is based on large population representative studies (we have >70,000 WSIs collected in-house). The main focus is on breast, prostate and colorectal cancer ( &  
  • Cancer precision medicine: Development and validation of predictive models for improved patient stratification, based on comprehensive molecular phenotyping data (e.g. DNA- and RNA-sequencing)

Current research grants (as PI)

Our work is supported by Swedish Research Council (VR), Swedish Cancer Society, ERA PerMed and Karolinska Institutet.

Doctoral students

  • Yinxi Wang (main supervisor)
  • Philippe Weitz (main supervisor)
  • Abhinav Sharma (main supervisor)
  • Quang Thinh Trac (co-supervisor)
  • Pablo Gonzalez Ginestet (co-supervisor)

Other group members

  • Bojing Liu (Postdoc)
  • Balasz Acs (Associated postdoc)
  • Duong Tran (Technician)

Former group members / supervised students

  • Boxi Zhang (MSc student)
  • Youcheng Zhang (MSc student)
  • Peter Ström (PhD student, co-supervisor)
  • Charlotte Von Heijne Widlund (visiting scientist)
  • Mei Wang (Postdoc; currently Research Assistant Professor, School of life science, Peking University, China)
  • Arvind Mer (Postdoc; currently Research fellow at Princess Margaret Bioinformatics and Computational Genomics Laboratory, Toronto, Canada)
  • Nghia Vu (Postdoc; currently Assistant Professor, Karolinska Institutet, MEB)

Examples of publications

Wang Y, Kartasalo K, Valkonen M, Larsson C, Ruusuvuori P, Hartman J, Rantalainen M. Predicting molecular phenotypes from histopathology images: a transcriptome-wide expression-morphology analysis in breast cancer. arXiv preprint arXiv:2009.08917. 2020 Sep 18 (pre-print)

Acs B, Rantalainen M, Hartman J., Artificial intelligence as the next step towards precision pathology. Journal of Internal Medicine. 2020 Mar 3.

Ström, P., Kartasalo, K., Olsson, H., Solorzano, L., Delahunt, B., Berney, D.M., Bostwick, D.G., Evans, A.J., Grignon, D.J., Humphrey, P.A., Iczkowski, K.A.. Kench, J.G., Kristiansen, G., van der Kwast, T.H., Leite, K.R.M., McKenneym, J.K., Oxley, J., Pan, C.C., Samaratunga, H., Srigley, J.R., Takahashi, H., Tsuzuki, T., Varma, M.,Zhou, M., Lindberg, J.,  Lindeskog,  C., Ruusuvuori, P, Wählby, C., Grönberg, H.,  Rantalainen,  M., Egevad, L., Eklund, M.,   Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology. 2020 Jan 8.

Wang, M., Lindberg, J., Klevebring, D., Nilsson, C., Lehmann, S., Grönberg, H., Rantalainen, M., Development and validation of a novel RNA sequencing-based prognostic score for acute myeloid leukemia, J Natl Cancer Inst, 2018 Mar 18

Vu, T.N., Wills, Q.F., Kalari, K.R., Niu, N., Wang, L., Pawitan, Y., Rantalainen M., Isoform-level gene expression patterns in single-cell RNA-sequencing data, Bioinformatics, 2018 Feb 1

Wang, M., Klevebring, D., Lindberg, K., Czene, K., Grönberg, H., Rantalainen M.; Determining breast cancer histological grade from RNA sequencing data. Breast Cancer Research, 2016, 18(1).

Vu, N.T., Wills, Q.F., Kalari, K.R., Niu, N., Wang, L., Rantalainen M.§, Pawitan Y.§. Beta-Poisson model for single-cell RNA-seq data analyses. Bioinformatics, 2016, btw202.

Stålhammar, G., Martinez, N.F., Lippert M., Tobin, N.P., Mølholm, I., Kis, L., Rosin, G., Rantalainen, M., Pedersen, L, Bergh, J, Grunkin, M.; Digital image analysis outperforms manual biomarker assessment in breast cancer. Modern Pathology, 2016, Feb 26.

Teaching portfolio

Loading bibliometrics...