Advanced Computer Vision Approaches in Biomedical Image Analysis
چکیده
Copyright © 2014 Anke Meyer-Baese et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Medical imaging is today becoming one of the most important visualization and interpretation methods in biology and medicine. The past decade has witnessed a tremendous development of new, powerful instruments for detecting, storing , transmitting, analyzing, and displaying images. These instruments are greatly amplifying the ability of biochemists, biologists, medical scientists, and physicians to see their objects of study and to obtain quantitative measurements to support scientific hypotheses and medical diagnoses. Noise, artifacts, and weak contrast are the cause of a decrease in image quality and make the interpretation of medical images very difficult. These sources of interference, which are of a different nature for mammograms than for ultrasound images, are responsible for the fact that conventional or traditional analysis and detection algorithms are not always successful. The biomedical imaging scene is one of the most challenging since we have to deal not only with non-Gaussian, nonstationary, and nonlinear processes (transients, bursts, and ruptures) but also with mixtures of components interacting in a quite complicated form. Therefore, much of the research done today is geared towards improvement of the reduced quality of the available biomedical imaging material. The aim of this special issue is to present the current state-of-the-art in the theory of advanced computer vision approaches and applications to biomedical image analysis and modeling. In this special issue, we have invited articles that explore novel problems in biomedical imaging that require advanced computer vision approaches. Three papers address the important aspect of segmentation in medical images, two papers deal with automated detection, and three others propose theoretical and computational or hardware solutions for encountered artifacts. One of the papers evaluates the feasibility of computer aided malignant tumor detection using the traditional texture analysis applied on two-compartment-based parameter pseu-doimages of dynamic contrast enhanced magnetic resonance (DCE-MR) breast image data. The innovation lies in the through-plane assessment capability. The paper shows that neural network classification using the textural parameters for automated screening of two-compartment-based parameter pseudoimages of DCE-MRI as input data represents a useful tool for the radiologists in the preassessment stage by revealing possible cancerous regions and in the postassess-ment stage by reviewing the segmentation especially in analyzing complex DCE-MRI cases. A paper …