Project: Advancing Breast Cancer histopathology towards AI-based Personalised medicine

Acronym	ABCAP (Reference Number: ERAPERMED2019-224)
Duration	01/04/2020 - 31/03/2023
Project Topic	Manual histopathological assessment of biopsies and resected tumours is currently the main mode to detect presence of breast cancer (BC), identify clinically relevant cancer, and to establish diagnosis. However, there is currently a shortage of pathology expertise in many parts of the world, and there is also a high inter-assessor variability between pathologists. This leads to prolonged response times, unnecessary patient anxiety, and unequal access to top-quality histopathology assessments for cancer patients. Misclassifications in histopathology assessments will cause both over- and under-treatment, and can have severe consequences for individual patients. We hypothesise that it is now possible to develop advanced image-based prediction models based on artificial intelligence (AI) and deep-learning (DL) techniques for BC histopathology assessment that match or outperform the performance of top-level human pathology experts. In this research programme we will develop and validate state-of-the-art AI-based models for BC routine histopathology classification and for improved patient stratification in respect to prognosis and treatment response. Through both retrospective and prospective validation of the developed models, we will establish evidence towards future clinical translation. Our studies are based on large-scale population samples (>125,000 whole slide images in total, multiple cohorts), ensuring unbiased data and models. Novel methodologies for stain-free and multi-stain analysis will also be developed, which has the potential to contribute to improved models and advancing the field of digital pathology. The project aims to improve the quality of BC histopathology assessments by reducing errors and inter-assessor variability, enhancing patient stratification and reducing over- and under-treatment of patients, while also contributing toward more efficient and reliable routine pathology.
Network	ERA PerMed
Call	2nd Joint Transnational Call for Proposals (2019)

Project partner

Number	Name	Role	Country
1	Karolinska Institutet	Coordinator	Sweden
2	Tampere University	Partner	Finland
3	University of Eastern Finland	Partner	Finland
4	Zealand University Hospital	Partner	Denmark