Funded by ERA PerMed
Karolinska Institutet (Sweden): Mattias Rantalainen (Phd), Department of Medical Epidemiology and Biostatistics (coordinator)
Tampere University (Finland): Pekka Ruusuvuori (PhD), Faculty of Medicine and Health Technology
University of Eastern Finland (Finland): Leena Latonen (PhD), Institute of Biomedicine,
Zealand University Hospital, Region Zealand (Denmark): Anne-Vibeke Laenkholm (MD), Department of Surgical Pathology
Manual histopathological assessment of biopsies or resected tumours is the main mode to detect breast cancer and to establish diagnosis. However, there is a shortage of pathology expertise and a high inter-assessor variability between pathologists. This leads to prolonged response times and unequal access to top-quality assessments. Misclassifications cause both over- and under-treatment, and can have severe consequences for individual patients.
In the ABCAP programme we will develop and validate novel state-of-the-art deep learning-based computer models for improved routine histopathology classification and for refined patient stratification. ABCAP is based on large population samples to ensure representative models. Through comprehensive validation of the developed models, evidence for clinical translation will be established. Improved quality of breast cancer histopathology assessments will contribute towards reducing both over- and under-treatment of breast cancer patients.