RESUMEN
Biological membrane images contain a variety of objects and patterns, which convey information about the underlying biological structures and mechanisms. The field of image analysis includes methods of computation which convert features and objects identified in images into quantitative information about biological structures represented in these images. Microscopy images are complex, noisy, and full of artifacts and consequently require multiple image processing steps for the extraction of meaningful quantitative information. This review is focused on methods of analysis of images of cells and biological membranes such as detection, segmentation, classification and machine learning, registration, tracking, and visualization. These methods could make possible, for example, to automatically identify defects in the cell membrane which affect physiological processes. Detailed analysis of membrane images could facilitate understanding of the underlying physiological structures or help in the interpretation of biological experiments.
Asunto(s)
Membrana Celular/ultraestructura , Procesamiento de Imagen Asistido por Computador , MicroscopíaRESUMEN
Identifying the separate parts in ultrasound images such as bone and skin plays a crucial role in the synovitis detection task. This paper presents a detector of bone and skin regions in the form of a classifier which is trained on a set of annotated images. Selected regions have labels: skin or bone or none. Feature vectors used by the classifier are assigned to image pixels as a result of passing the image through the bank of linear and nonlinear filters. The filters include Gaussian blurring filter, its first and second order derivatives, Laplacian as well as positive and negative threshold operations applied to the filtered images. We compared multiple supervised learning classifiers including Naive Bayes, k-Nearest Neighbour, Decision Trees, Random Forest, AdaBoost and Support Vector Machines (SVM) with various kernels, using four classification performance scores and computation time. The Random Forest classifier was selected for the final use, as it gives the best overall evaluation results.