Region Growing for Medical Image Segmentation Using a Modified Multiple-seed Approach on a Multi-core CPU Computer

2014 ◽  
pp. 112-115 ◽  
Author(s):  
Agus Pratondo ◽  
Sim Heng Ong ◽  
Chee Kong Chui
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Region Growing ◽  
Medical Image Segmentation

Research and Improvement of Live-Wire Interactive Algorithm for Medical Image Segmentation

2012 ◽  
Vol 182-183 ◽  
pp. 1065-1068 ◽  
Author(s):  
Ji Ying Li ◽  
Jian Wu Dang
Keyword(s):  
Image Segmentation ◽  
Anisotropic Diffusion ◽  
Medical Image ◽  
Optimal Path ◽  
Region Growing ◽  
Medical Image Segmentation ◽  
Interactive Segmentation ◽  
Diffusion Filtering ◽  
Interactive Algorithm ◽  
Region Growing Segmentation

Traditional Live Wire algorithm distinguished the strength edge of objectives uneasily and executive speed of algorithm is slow. For these problems, an improved Live-Wire algorithm is proposed. First it implements anisotropic diffusion filtering to images and constructs a new expense function, then combined with region growing segmentation algorithm, it implements interactive segmentation to medical images. Improved algorithm avoids the shortcomings of the traditional Live-wire algorithm which is sensitive to noise and can not effectively distinguish the edge of the strength, also reduces the time and blindness of dynamic programming to find the optimal path and improves the accuracy and implementation efficiency of the image segmentation.

Researcher of Region Growing Medical Image Segmentation Based on Adaboost Algorithm

2010 ◽  
Vol 142 ◽  
pp. 21-25
Author(s):  
Peng Wang ◽  
X.F. Ye ◽  
Shi Wei Yin ◽  
Shao Chen Kang ◽  
Jing Lei Xin
Keyword(s):  
Image Segmentation ◽  
Background Subtraction ◽  
Medical Image ◽  
Region Growing ◽  
Medical Image Segmentation ◽  
Segmentation Algorithm ◽  
Adaboost Algorithm ◽  
Actual Application ◽  
Region Extraction ◽  
Seed Pixel

To obtain better region extraction results of medical image, a new segmentation algorithm is proposed based on improved Adaboost algorithm. The seed pixel is selected with background subtraction. The neighborhood point is judged. The primary selected seed is calibrated with label, and then the range of seed is reduced through growing label and the maximal saliency. The optimized Adaboost algorithm is taken as growing criterion to optimally combine the scrappy region when the region growing is over. The experiment result shows that the accuracy and robustness of the algorithm both meet the actual application required.

scholarly journals On Unsupervised Methods for Medical Image Segmentation: Investigating Classic Approaches in Breast Cancer DCE-MRI

2021 ◽  
Vol 12 (1) ◽  
pp. 162
Author(s):  
Carmelo Militello ◽  
Andrea Ranieri ◽  
Leonardo Rundo ◽  
Ildebrando D’Angelo ◽  
Franco Marinozzi ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Image Segmentation ◽  
Medical Image ◽  
Large Scale ◽  
False Negative ◽  
Region Growing ◽  
Medical Image Segmentation ◽  
Machine Learning Techniques ◽  
Unsupervised Segmentation ◽  
Dce Mri

Unsupervised segmentation techniques, which do not require labeled data for training and can be more easily integrated into the clinical routine, represent a valid solution especially from a clinical feasibility perspective. Indeed, large-scale annotated datasets are not always available, undermining their immediate implementation and use in the clinic. Breast cancer is the most common cause of cancer death in women worldwide. In this study, breast lesion delineation in Dynamic Contrast Enhanced MRI (DCE-MRI) series was addressed by means of four popular unsupervised segmentation approaches: Split-and-Merge combined with Region Growing (SMRG), k-means, Fuzzy C-Means (FCM), and spatial FCM (sFCM). They represent well-established pattern recognition techniques that are still widely used in clinical research. Starting from the basic versions of these segmentation approaches, during our analysis, we identified the shortcomings of each of them, proposing improved versions, as well as developing ad hoc pre- and post-processing steps. The obtained experimental results, in terms of area-based—namely, Dice Index (DI), Jaccard Index (JI), Sensitivity, Specificity, False Positive Ratio (FPR), False Negative Ratio (FNR)—and distance-based metrics—Mean Absolute Distance (MAD), Maximum Distance (MaxD), Hausdorff Distance (HD)—encourage the use of unsupervised machine learning techniques in medical image segmentation. In particular, fuzzy clustering approaches (namely, FCM and sFCM) achieved the best performance. In fact, for area-based metrics, they obtained DI = 78.23% ± 6.50 (sFCM), JI = 65.90% ± 8.14 (sFCM), sensitivity = 77.84% ± 8.72 (FCM), specificity = 87.10% ± 8.24 (sFCM), FPR = 0.14 ± 0.12 (sFCM), and FNR = 0.22 ± 0.09 (sFCM). Concerning distance-based metrics, they obtained MAD = 1.37 ± 0.90 (sFCM), MaxD = 4.04 ± 2.87 (sFCM), and HD = 2.21 ± 0.43 (FCM). These experimental findings suggest that further research would be useful for advanced fuzzy logic techniques specifically tailored to medical image segmentation.

Medical image segmentation using 3D seeded region growing

1997 ◽  
Author(s):  
R. K. Justice ◽  
Ernest M. Stokely ◽  
John S. Strobel ◽  
Raymond E. Ideker ◽  
William M. Smith
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Region Growing ◽  
Medical Image Segmentation ◽  
Seeded Region Growing

A medical image segmentation algorithm based on bi-directional region growing

Optik ◽  
2015 ◽  
Vol 126 (20) ◽  
pp. 2398-2404 ◽  
Author(s):  
Xiaoli Zhang ◽  
Xiongfei Li ◽  
Yuncong Feng
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Region Growing ◽  
Medical Image Segmentation ◽  
Segmentation Algorithm ◽  
Image Segmentation Algorithm
Sign in / Sign up

Export Citation Format

Share Document