User-Agent Cooperation in Multiagent IVUS Image Segmentation

Abstract

Automated interpretation of complex images requires elaborate knowledge and model-based image analysis, but often needs interaction with an expert as well. This research describes expert interaction with a multiagent image interpretation system using only a restricted vocabulary of high-level user interactions. The aim is to minimize inter- and intra-observer variability by keeping the total number of interactions as low and simple as possible. The multiagent image interpretation system has elaborate high-level knowledge-based control over low-level image segmentation algorithms. Agents use contextual knowledge to keep the number of interactions low but, when in doubt, present the user with the most likely interpretation of the situation. The user, in turn, can correct, supplement, and/or confirm the results of image-processing agents. This is done at a very high level of abstraction such that no knowledge of the underlying segmentation methods, parameters or agent functioning is needed. High-level interaction thereby replaces more traditional contour correction methods like inserting points and/or (re)drawing contours. This makes it easier for the user to obtain good results, while inter- and intra-observer variability are kept minimal, since the image segmentation itself remains under control of image-processing agents. The system has been applied to intravascular ultrasound (IVUS) images. Experiments show that with an average of 2-3 high-level user interactions per correction, segmentation results substantially improve while the variation is greatly reduced. The achieved level of accuracy and repeatability is equivalent to that of manual drawing by an expert.

Automatic Coronary Wall Segmentation in Intravascular Ultrasound Images Using Binary Morphological Reconstruction

“…Prior studies have documented the importance of IVUS, and IVUS segmentation researches for many coronary disease investigations (dos S. Filho et al 2005; Coronary wall segmentation using binary morphological reconstruction d M. C. MORAES and S. S. FURUIE 1497 Bovenkamp et al 2009;Brusseau et al 2004;Gil et al 2006;Katouzian et al 2008b;Papadogiorgaki et al 2008;Sonka et al 1995;Unal et al 2008). In spite of the good results presented by these studies, the pursuit of achieving an optimum method, which could be reliable, complete and practical, is still in progress.…”

“…Much significant IVUS segmentation work has been published lately (Bovenkamp et al 2009;Brusseau et al 2004;Gil et al 2006;Katouzian et al 2008b;Papadogiorgaki et al 2008;Sonka et al 1995;Unal et al 2008). In a study by Bovenkamp et al (2009), a multi-agent image interpretation system is described.…”

Non-Coronary Vessel Exploration Under Intravascular Ultrasound: Principles and Applicability

“…Different approaches have been described to detect the luminal and external elastic membrane borders from 2-D IVUS images, including texture-based methods (Papadogiorgaki et al, 2007), knowledge-based graph searching (Bovenkamp et al, 2009), region growing (Sanz-Requena et al, 2007), radial gradient searching (Luo et al, 2003), and active contours (Sanz-Requena et al, 2007;Takagi et al, 2000). These 2-D border-detection techniques are now used online in newer IVUS imaging consoles, allowing the quantitative evaluation of lumen and plaque measurements.…”

Mixed Integer Evolution Strategies for Parameter Optimization

“…In Bovenkamp et al (2004Bovenkamp et al ( , 2009) a state of the art multiagent system is developed to detect lumen, vessel, shadows, sidebranches, and calcified plaques in IVUS images. IVUS image processing agents interact with other agents through communication, act on the world by controlling and adapting image processing operations and perceive that same world by accessing image processing results.…”