Abstract :
The recent expand in the utilization of Whole Slide scanners in Digital Pathology gave birth to a production of massive amount of data and the need of integration of Digital Pathology Systems (DPS’s) into modern Laboratory Information Systems (LIS’s). In this context, the problem of automatically retrieving a particular image from a large set of digital images that contains similar medical visual content has gained fruitful ground. This work investigates the fast and consistent properties of the Speeded-Up Robust Features (SURF) algorithm in order to search in the content of a digital pathology image, detect and find similarities for content-based image retrieval. An important aspect of this work is the diversity of Whole Slide Scanners. The proposed methodology that involves the process of the comparison of digital pathology images, mostly WSI, with the use of the SURF algorithm was proved robust to various condition changes.
https://hal.inria.fr/hal-01821063
Contributor : Hal Ifip <>
Submitted on : Friday, June 22, 2018 - 11:45:37 AM Last modification on : Friday, June 22, 2018 - 12:00:46 PM Long-term archiving on: : Wednesday, September 26, 2018 - 6:58:12 PM
A. Kallipolitis, I. Maglogiannis. Content Based Image Retrieval in Digital Pathology Using Speeded Up Robust Features. 14th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), May 2018, Rhodes, Greece. pp.374-384, ⟨10.1007/978-3-319-92007-8_32⟩. ⟨hal-01821063⟩