Wyniki wyszukiwania - BazTech

1

A sparse pair-preserving centroid-based supervised learning method for high-dimensional biomedical data or images

Kalina Jan, Matonoha Ctirad

Biocybernetics and Biomedical Engineering

|

2020

|

Vol. 40, no. 2

774--786

EN

In various biomedical applications designed to compare two groups (e.g. patients and controls in matched case-control studies), it is often desirable to perform a dimensionality reduction in order to learn a classification rule over high-dimensional data. This paper considers a centroid-based classification method for paired data, which at the same time performs a supervised variable selection respecting the matched pairs design. We propose an algorithm for optimizing the centroid (prototype, template). A subsequent optimization of weights for the centroid ensures sparsity, robustness to outliers, and clear interpretation of the contribution of individual variables to the classification task. We apply the method to a simulated matched case-control study dataset, to a gene expression study of acute myocardial infarction, and to mouth localization in 2D facial images. The novel approach yields a comparable performance with standard classifiers and outperforms them if the data are contaminated by outliers; this robustness makes the method relevant for genomic, metabolomic or proteomic high-dimensional data (in matched case-control studies) or medical diagnostics based on images, as (excessive) noise and contamination are ubiquitous in biomedical measurements.

2

Integrating Ant Colony Optimization with Level Set Method for Biomedical Image Boundary Detection

Rahebi J., Hardalaç F.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 5

218-221

EN

In this paper, ACO-based level set method is introduced to tackle the biomedical image boundary detection problem. The proposed ACO based level set method boundary detection approach is able to construct a pheromone matrix that represents the boundary information presented at each pixel position of the image, according to the movements of a number of ants which are dispatched to move on the image, then this result is initial contour for zero level set function in boundary of image that is segmented. Furthermore, the movements of these ants steers by the local variation of the image’s intensity values that it cause the contour move toward the object and exactly found boundaries. ACO-based method determines the initial contour to reduce the iteration steps. Such improvements simplify level set manipulation and lead to more robust segmentation. Experimental results show that the proposed method is can preserve the detail of the object and can be used to reduce the capacity of more computational tasks in research.

PL

W artykule opisano zastosowanie optymalizacji ACO (ang. Ant Colony Optimization) i metody poziomic w wychwytywaniu naruszenia brzegów obrazów biomedycznych. Proponowana metoda wykrywania granic tworzy matrycę feromonów, reprezentujących informacje brzegowe dla każdego z pikseli obrazu, w oparciu o ruch mrówek poruszających się po nim. Dane te stanowią wartość początkową dla funkcji ustalającej poziomicę zerową granicy obrazu. Pozwala to na redukcję ilości iteracji algorytmu. Wyniki badań eksperymentalnych potwierdzają skuteczność działania metody.

3

Segmentation of color biomedical images

Ablameyko S., Belotserkovsky A., Oczeretko E., Półjanowicz W.

Image Processing & Communications

|

2003

|

Vol. 9, no 1-2

17-29

EN

The task of color cell image segmentation is considered. Overview of cell image segmentation algorithms is given. A new homogeneity and color characteristics histogram based algorithm of color cell image segmentation is proposed. It includes a modified edge detection algorithm based on Sobel operator and the procedure of RGB-HSV transformation that was applied to decrease number of operations and increase the segmentation quality.