In this talk, we present MitosisAnalyser, a framework for automated mitosis detection and cell tracking in time-lapse phase contrast microscopy images. In cancer research, live-cell imaging experiments constitute a key part in chemotherapy drug development, where abnormal mitotic behaviour is a key indicator of successful treatment. Derived from mathematical concepts like the circular Hough transform and variational segmentation methods, we present our holistic and automated imaging pipeline and show results for real data while validating them.

This presentation is part of Minisymposium “MS44 - 3D Image Depth/Texture/Reflectivity Tracking, Modelling and Reconstruction”
organized by: Catherine Higham (University of Glasgow) , Roderick Murray-Smith (University of Glasgow) .

Authors:

Joana Grah (Graz University of Technology)

Jenny Harrington (Cancer Research UK Cambridge Institute)

Siang B. Koh (Massachusetts General Hospital Cancer Center, Boston)

Jeremy A. Pike (University of Birmingham)

Alexander Schreiner (PerkinElmer, Hamburg)

Martin Burger (University of Muenster)

Carola-Bibiane Schönlieb (University of Cambridge)

Stefanie Reichelt (Cancer Research UK Cambridge Institute)

Keywords:

computer vision, image segmentation, partial differential equation models