Abstract
Region of interest (ROI) plays an important role in medical image analysis. In this paper, a new approach to ROI extraction based on the curve evolution is proposed. Different from the existent method, the proposed approach is efficient both in segmentation results and computational cost. The deforming curve is modeled as a monotonically marching front under a positive speed field, where a region speed function is derived by minimizing the new defined ROI energy, and integrated with the edge-based speed function. The curve evolution model integrating the ROI information has a large propagation range and could even drive the front in low-contrast and narrow thin areas. Moreover, a multi-initial fast marching algorithm, which permits the user to plant several seed curves as the initial front and evolves them simultaneously, is developed to fast implement the numerical solution. Selective planting seed curves could help the local growth and thus may further improve the segmentation results and reduce the computational cost. Experiments by our approach are presented and compared with that of the other methods, which show that the proposed approach could fast extract low-contrast and narrow thin ROI precisely.
Highlights
Region of interest (ROI) plays an important role in medical image analysis
Considering both the segmentation quality and the computational cost, in this paper, we propose an efficient approach to ROI extraction
The remainder of the paper is as follows: in Section 2, fast marching method is briefly outlined; in Section 3, the curve evolution model is proposed, where a new speed function is introduced by ROI energy minimizing; in Section 4, multi-initial fast marching algorithm is described in detail; in Section 5, experimental results are presented and compared with those of the other methods; in Section 6, conclusions are reported
Summary
Region of interest (ROI) plays an important role in medical image analysis. Quantitative analysis of the shape and the properties of ROI could provide reliable data for diagnosing disease and the follow-up treatment planning [1]. Because the image statistics are variable with the deforming curve, these statistics need to be estimated during the curve evolution, which may bring much computational cost It needs n − 1 curves to segment n regions with each curve corresponding to different curve evolution equation and level set function, which present complex computation. Due to its complex form in speed function, the corresponding level set evolution equation is implemented by Hermes algorithm [13], which is more computationally expensive than the Fast Marching method. Considering both the segmentation quality and the computational cost, in this paper, we propose an efficient approach to ROI extraction. The remainder of the paper is as follows: in Section 2, fast marching method is briefly outlined; in Section 3, the curve evolution model is proposed, where a new speed function is introduced by ROI energy minimizing; in Section 4, multi-initial fast marching algorithm is described in detail; in Section 5, experimental results are presented and compared with those of the other methods; in Section 6, conclusions are reported
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
