Automatic classification of skin lesions

Pathology services in the Western Norway Health Region – a center for applied digitization: Automatic classification of skin lesions

Financed by Helse Vest

Samarbeidspartnere

prof. Kjersti Engang, University of Stvanger

prof. Carlos Monteagudo, INICLIVA, Valencia, Spain

Formål

To develop a Computer-aided Diagnostic (CAD) system for whole slide images that will help pathology departments to diagnose malignant melanoma more efficiently, by (1) reducing examination time, (2) reducing diagnostic variations and (3) increase diagnostic accuracy.

Problemstilling

Melanoma of the skin is the cancer type with the largest increase in incidence during the last decade. Currently, after excision of a suspicious skin lesion, the tissue is fixed in buffered formalin and a histological section is examined under a microscope by a trained expert pathologist. The pathologists use their deep domain knowledge to identify several possible complex morphological and cytological features. Unfortunately, this manual diagnosis of skin cancer is a time- and labour intensive task and not always that easy and reproducible. In addition, the sheer amount of skin biopsies causes logistic and personnel issues. The main challenge is to find the few malignant melanomas among all the other benign or less malign skin cancers.