Abstract

Accurate diagnosis of cancer in tissue relies on labor intensive sample preparation after which specialists are required to set a diagnosis through microscopy imaging. Despite the effort involved, the rates of conclusive diagnosis by histological analysis for a range of the most dangerous cancers is only 65-75% [1]. The META-HILIGHT project financed under the ATTRACT Phase 2 project (European Union's Horizon 2020 Research and Innovation Programme) aims to further develop a novel polarization sensitive histopathology modality developed at the University of Oulu [2]. The technique relies on polarimetric mapping of ex vivo tissue samples. The alterations of the polarization associated with tissue cancerous malformations will be explored for the use as diagnostic markers by utilizing supervised Machine Learning (ML) and Deep Learning (DL). META-HILIGHT aims to expand the applicability of this modality by demonstrating a portable stand-alone digital histopathology system based on metamaterial components suitable for scalable production.

© 2023 IEEE