In CHIME we design new studies, develop and validate models for AI-based computational pathology.  We are focused on improving quality in routine cancer diagnoses and on extracting comprehensive information from histopathology image data that enable precision medicine. Most of our studies are large-scale population-based studies with digitial whole slide images together with clinical data.


CHIME currently includes studies in the areas of breast, prostate and colorectal cancer.


Mattias Rantalainen, PhD (Associate professor)


Group webpage: https://ki.se/en/people/matran


Department of Medical Epidemiology and Biostatistics (MEB)

Karolinska Institute

Stockholm, Sweden

Supported by

The Swedish Cancer Foundation

The Swedish Research Council

Karolinska Institutet


Data in short:

> 110,000 WSIs

> 200 TB of image data (and growing!)


Wang Y, Kartasalo K, Weitz P, Acs B, 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. Cancer Research. 2021 Oct 1;81(19):5115-26.

Wang Y, Acs B, Robertson S, Liu B, Solorzano L, Wählby C, Hartman J, Rantalainen M. Improved breast cancer histological grading using deep learning. Annals of Oncology. 2021 Sep 29.

Weitz P, Wang Y, Hartman J, Rantalainen M. An Investigation of Attention Mechanisms in Histopathology Whole-Slide-Image Analysis for Regression Objectives. Proceedings of the IEEE/CVF International Conference on Computer Vision 2021 (pp. 611-619).


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.

Weitz P, Wang Y, Kartasalo K, Egevad L, Lindberg J, Grönberg H, Eklund M, Rantalainen M. Transcriptome-wide prediction of prostate cancer gene expression from histopathology images using co-expression based convolutional neural networks. arXiv preprint arXiv:2104.09310. 2021 Apr 19.

Weitz P, Acs B, Hartman J, Rantalainen M. Prediction of Ki67 scores from H&E stained breast cancer sections using convolutional neural networks.


Liu B., Wang Y.,,Weitz P.,,Lindberg J, Egevad L, Grönberg H, Eklund M, Rantalainen M. Using deep learning to detect patients at risk for prostate cancer despite benign biopsies. arXiv preprint