Computerized Medical Imaging and Graphics 61 (2017) 1

Contents lists available at ScienceDirect

Computerized Medical Imaging and Graphics journal homepage: www.elsevier.com/locate/compmedimag

Preface

MARK

After the 13th European Congress on Digital Pathology, organised in Berlin in 2016, a selection of papers has been carefully reviewed to go for this special issue, covering a series of discriminative approaches highlighted during the conference, in Digital Pathology. With the recent decision of the FDA to permit marketing of a first whole slide imaging (WSI) system that allows for review and interpretation of digital surgical pathology slides prepared from biopsied tissue, digitalization in pathology made a giant leap into daily routine work. As expected, Deep Learning belongs to the selected approaches, in opposition to more traditional random forest approaches. The natural trend is to work on Whole Slide Image (WSI), now when the computer capabilities are definitely allowing it, in order to have a better overview of the anatomopathological semiology. The context targeted here is related to an automatic recognition of gastric carcinoma in Hematoxylin and Eosin (H & E) stained WSI. Another important issue presented in this selection is related to the automatic quantification of immunohistochemical (IHC) stain in breast Tumour Microarray (TMA) using colour analysis. Several methods have been analysed including colour standardisation methods, compared against colour model segmentation. This lead to an interesting discussion which we are pleased to submit to our readers. Finally an adaptive algorithm for automated Ki67 hotspot detection has been presented for breast cancer biopsies, in a context in which we clearly see the interest of computer aided tools for a preliminary filtering of the image, in the interest of the efficiency and the effectiveness of Pathologist's daily work. Algorithms of that type will be used in a broad manner in the near future. “Automation” constitutes the link that connects the series of state of the art publications in area of Medical Image Computing and, in particular in Digital Pathology. Those methods are only supporting Clinicians - and in our case Pathologists - and are at the top of the loop, in which the proposed tools are having the role of supporting quantification and traceability. The diagnostic decision will always belong to the them, but the computer aided support will be more and more a part or their future panel of services, in the interest of an increasing quality of care for the final benefit of the patients. Daniel Racoceanu Pontifical Catholic University of Peru, San Miguel, Lima, Peru Sorbonne Université - Université Pierre et Marie Curie, Paris, France E-mail address: [email protected]

Peter Hufnagl1 Charité – Universitätsmedizin Berlin, Berlin, Germany HTW University of Applied Sciences, Berlin, Germany E-mail address: [email protected]

1

Tel.: +49 30 450 536 140; Fax: +49 30 450 536 910.

http://dx.doi.org/10.1016/j.compmedimag.2017.10.001

0895-6111/ © 2017 Published by Elsevier Ltd.