Interactive Medical Image Segmentation Research Articles

Interactive Medical Image Segmentation Using Snake and Multiscale Curve Editing

Image segmentation is typically applied to locate objects and boundaries, and it is an essential process that supports medical diagnosis, surgical planning, and treatments in medical applications. Generally, this process is done by clinicians manually, which may be accurate but tedious and very time consuming. To facilitate the process, numerous interactive segmentation methods have been proposed that allow the user to intervene in the process of segmentation by incorporating prior knowledge, validating results and correcting errors. The accurate segmentation results can potentially be obtained by such user-interactive process. In this work, we propose a novel framework of interactive medical image segmentation for clinical applications, which combines digital curves and the active contour model to obtain promising results. It allows clinicians to quickly revise or improve contours by simple mouse actions. Meanwhile, the snake model becomes feasible and practical in clinical applications. Experimental results demonstrate the effectiveness of the proposed method for medical images in clinical applications.

REGION MERGING VIA GRAPH-CUTS

In this paper, we discuss the use of graph-cuts to merge the regions of the watershed transform optimally. Watershed is a simple, intuitive and efficient way of segmenting an image. Unfortunately it presents a few limitations such as over-segmentation and poor detection of low boundaries. Our segmentation process merges regions of the watershed over-segmentation by minimizing a specific criterion using graph-cuts optimization. Two methods will be introduced in this paper. The first is based on regions histogram and dissimilarity measures between adjacent regions. The second method deals with efficient approximation of minimal surfaces and geodesics. Experimental results show that these techniques can efficiently be used for large images segmentation when a pre-computed low level segmentation is available. We will present these methods in the context of interactive medical image segmentation.

Interactive 3D medical image segmentation with energy-minimizing implicit functions

We present an interactive segmentation method for 3D medical images that reconstructs the surface of an object using energy-minimizing, smooth, implicit functions. This reconstruction problem is called variational interpolation. For an intuitive segmentation of medical images, variational interpolation can be based on a set of user-drawn, planar contours that can be arbitrarily oriented in 3D space. This also allows an easy integration of the algorithm into the common manual segmentation workflow, where objects are segmented by drawing contours around them on each slice of a 3D image.Because variational interpolation is computationally expensive, we show how to speed up the algorithm to achieve almost real-time calculation times while preserving the overall segmentation quality. Moreover, we show how to improve the robustness of the algorithm by transforming it from an interpolation to an approximation problem and we discuss a local interpolation scheme.A first evaluation of our algorithm by two experienced radiology technicians on 15 liver metastases and 1 liver has shown that the segmentation times can be reduced by a factor of about 2 compared to a slice-wise manual segmentation and only about one fourth of the contours are necessary compared to the number of contours necessary for a manual segmentation.

SU‐GG‐I‐95: Simpler and Faster Watershed Medical Image Segmentation Algorithm

Purpose: A watershed medical image segmentation method has a drawback of producing a region for each local minimum, resulting in oversegmentation. To alleviate this problem, a post merging process such as a hybrid graph merging algorithm is used, which impedes interactive segmentation with a graphical user interface. An effective watershed algorithm without a post merging process is proposed. Method and Materials: Our algorithm, as the preprocessing, includes the edge‐preserving noise reduction and gradient‐based boundary sharpening steps. Because the initial oversegmentation depends on the gradient image intensity variations, the noise reduction algorithm can reduce the problem appreciably. A well‐known statistical edge‐preserving noise reduction algorithm is implemented to preserve the boundaries of the image objects which have different physiological properties. Gradient‐based boundary sharpening stage follows the novel edge‐preserving noise reduction step. Boundary sharpening process combined with watershed segmentation results in an effective watershed algorithm without a post merging process. The combination of the standard gradient values with two cross operators makes the edges sharper and thinner. A threshold value is controlled interactively with a steppedup graphical user interface in the boundary sharpening process. Results: We applied the method to both CT and MR images. Interactive real time medical image segmentation was possible with a graphical user interface. Threshold values were manipulated interactively with a mouse to get an appropriate segmentation of the abdomen CT image. It was also confirmed that this method produced the same segmentations as those of hybrid‐merging algorithm for the MR brain images. Conclusion: In the gradient‐based watershed, computing time is negligible and storage requirement can be reduced appreciably because a post hybrid merging algorithm is not necessary. Moreover, interactive real time segmentation is possible with the steppedup graphical user interface.

SketchSnakes: Sketch-line initialized Snakes for efficient interactive medical image segmentation

We present an intuitive, fast and accurate 2D interactive segmentation method that combines a general subdivision-curve Snake possessing powerful editing capabilities, with a novel sketch-line user initialization process, and a pen input device. Using the pen (or a mouse), the Snake is quickly and precisely initialized with a few quick sketch lines drawn across the width of the target object. The smooth contour constructed using these lines is extremely close to the position and shape of the object boundary. This makes the Snake’s task of snapping to the object boundary much simpler and hence more likely to succeed in noisy images with minimal user editing. We apply our Snake to the segmentation of several 2D medical images to demonstrate it’s efficiency, accuracy and robustness. We also compare SketchSnakes to Adobe Photoshop’s Magnetic Lasso (Adobe Systems Inc., Adobe Photoshop User Guide, 2002) as well as a recent graph-cut based image cutout tool known as Snap (Digital Film Tools LLC, Snap User Guide, 2007) in order to highlight SketchSnakes effectiveness.

An interactive medical image segmentation system based on the optimal management of regions of interest using topological medical knowledge

This paper presents an original interactive system for efficient medical image segmentation in computer aided diagnosis. The main originality concerns the method used to manage, according to an a priori topological-based structural model, regions of interest (ROIs) within which computations can be constrained. The goal is then to avoid the processing of irrelevant image points, therefore improving and accelerating segmentations. In the case of a hierarchical modeling procedure, our ROI management method enables, for delineating a given medical structure, to optimally determine image points of interest by taking previously segmented structures into account. We propose a mathematical formulation of the method as well as a possible implementation within an interactive system. We also detail an experience report focussing on the segmentation of several abdominal structures from a CT image. It illustrates the behavior and the potential of our method.

A Flexible, Generic Model for Anatomic Shape: Application to Interactive Two-Dimensional Medical image Segmentation and Matching

A representation called a radial contour model (RCM) is described for two-dimensional anatomic shapes. The model, which is a type of a geometric constraint network (GCN), is both flexible, in that it can deform to fit a particular instance of an anatomic shape, and generic, in that it captures all examples of a particular anatomic shape class. The model is implemented in a program, called SCANNER (version 0.7), for interactive model-based two-dimensional image segmentation and matching. Use of the model allows the segmenter to direct the search for edges in the image, and to fill in edges where none are present. Evaluations were done using models of 15 cross-sectional shapes appearing on CT images from 16 patients. Results from 480 trials show that the model-based approach reduces segmentation time by nearly a factor of 3 over manual methods, and correctly classifies 72.9% of the contours. The results not only suggest that the RCM will be useful for several current medical image segmentation tasks, but also support the hypothesis that geometric constraint networks are a viable approach to anatomic shape representation.