Exemplar Project 4: Automated workflows for breast cancer diagnosis and treatment

Breast cancer represents a significant public health issue with approximately 2.3 million women worldwide being diagnosed with breast cancer each year. Early detection and treatment is key to improving outcomes. However, this is challenging due to the significant shape changes the breast undergoes during clinical interventions, which make localising tumours difficult. Our aim is to integrate state-of-the-art image processing techniques, personalised 3D biomechanical modelling, and population-based statistical analyses to help address clinical challenges in the interpretation of medical images to improve diagnosis and treatment of breast cancer.

We will describe our latest efforts towards developing automated workflows to predict breast shape under different loading conditions to pinpoint the locations of suspected tumours during breast cancer diagnosis and treatment procedures. This includes: application of machine learning algorithms to automatically segment the breast and its internal tissues from diagnostic medical images; generation of 3D personalised and anatomically accurate models of the torso and upper limbs and the skeleton from segmented image data using a combination of nonlinear geometric fitting and statistical shape analysis techniques; simulation of large deformation mechanics to predict shape changes that occur when the breast is repositioned; and the development of augmented reality platform to visualise simulated tumour positions directly on patients during clinical procedures. Studies using MR images from breast cancer patients are underway in collaboration with breast radiologists at Auckland City Hospital to assess the efficacy of these technologies in the clinical setting.

We will describe how the 12 Labours project is providing infrastructure, resources, and guidelines to supports these developments. This research has the potential to lead to technological advancements in the breast cancer imaging field, which could translate into better health outcomes for women, and improve breast care practices world-wide.