Our group won the first place in the CAMELYON17 grand challenge

Digital pathology, is a new, rapidly expanding field in medical imaging in which whole-slide scanners are used to digitise glass slides at high resolution (up to 160nm per pixel). The availability of whole-slide images (WSI) has garnered the interest of the medical image analysis community, resulting in increasing numbers of publications on histopathologic image analysis.

Whole-slide images are generally stored in a multi-resolution pyramid structure. Image files contain multiple down-sampled versions of the original image. Each image in the pyramid is stored as a series of tiles, to facilitate rapid retrieval of subregions of the image.

A typical whole-slide image is approximately 200000 x 100000 pixels on the highest resolution level with 3 byte RGB pixel format. This means 55.88GB of uncompressed pixel data from a single level.

The aim of CAMELYON17 is to evaluate new and existing algorithms for automated detection and classification of breast cancer metastases in whole-slide images of histological lymph node sections. Last year’s CAMELYON16 focused on the detection of lymph node metastases, both on a lesion-level and on a slide-level. This year we move up to patient-level analysis, which requires combining the detection and classification of metastases in multiple lymph node slides into one outcome: a pN-stage. This task has high clinical relevance and would normally require extensive microscopic assessment by pathologists. Therefore, an automated solution for assessing the pN-stage in breast cancer patients, would hold great promise to reduce the workload of pathologists, while at the same time, reduce the subjectivity in diagnosis.

Our group won the first place in the CAMELYON17 grand challenge