Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Deep learning for damaged tissue detection and segmentation in Ki-67 brain tumor specimens based on the U-net model

Swiderska-Chadaj Z., Markiewicz T., Gallego J., Bueno G., Grala B., Lorent M.

Bulletin of the Polish Academy of Sciences. Technical Sciences

2018

Vol. 66, nr 6

849--856

The pathologists follow a systematic and partially manual process to obtain histological tissue sections from the biological tissue extracted from patients. This process is far from being perfect and can introduce some errors in the quality of the tissue sections (distortions, deformations, folds and tissue breaks). In this paper, we propose a deep learning (DL) method for the detection and segmentation of these damaged regions in whole slide images (WSIs). The proposed technique is based on convolutional neural networks (CNNs) and uses the U-net model to achieve the pixel-wise segmentation of these unwanted regions. The results obtained show that this technique yields satisfactory results and can be applied as a pre-processing step for automatic WSI analysis in order to prevent the use of the damaged areas in the evaluation processes.

MIAP – Web-based platform for the computer analysis of microscopic images to support the pathological diagnosis

Markiewicz T., Korzynska A., Kowalski A., Swiderska-Chadaj Z., Murawski P., Grala B., Lorent M., Wdowiak M., Zak J., Roszkowiak L., Kozlowski W., Pijanowska D.

Biocybernetics and Biomedical Engineering

2016

Vol. 36, no. 4

597--609

The aim of the project is to design and to implement a web-based platform for the computer analysis of microscopic images which support the pathological diagnosis. The use of the platform will be free of charge. It offers: quantitative analysis of staining tissue sample' images, archiving microscopic images, peer consultation, and join work independently from distance between scientific collaborating centers to registered doctors, scientists and students. The use of proposed platform allows: (i) to save pathologists' time spend on quantitative analysis, (ii) to reduce consulting costs by replacing sending of the physical preparations by placing their images (mostly virtual slide) on the platform server, (iii) to increase reproducibility, comparability and objectivity of quantitative evaluations. These effects have a direct impact on improving the effectiveness and decreasing the costs of patients' treatment. This paper presents the main ideas of the project which deliver web-based system working as multi-functional, integrated, modular and scalable computer system. The details of hardware solutions, concept of the workflow in the platform, the programming language and interpreters, the specific tools and algorithms, and the user interfaces are described below. The practical solutions for web-based services in the area of medical image analysis, storage and retrieval are also presented and discussed.