TY - GEN
T1 - An improved threshold method based on histogram entropy for the blood vessel segmentation
AU - Guo, Shuxiang
AU - Yang, Qiuxia
AU - Gao, Baofeng
AU - Cai, Xiaojuan
AU - Zhao, Yan
AU - Xiao, Nan
AU - He, Yanlin
AU - Zhang, Chaonan
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/2
Y1 - 2017/7/2
N2 - Interest is growing in the interventional surgery training system used before the treatment of vessel diseases. As one of the elementary component of the simulator, an accurate reconstruction of blood vessel obtaining from cross-sectional images is ugly needed. Digital Subtraction Angiography (DSA) data is used as a criterion for reconstruction result of blood vessel. In this paper, an improved threshold method is proposed to segment blood vessel from medical images. Firstly, with optimization characteristic, the genetic algorithm is used to determine the pre-segmentation greyscale from best histogram entropy. Secondly, we enlarge greyscale around the best greyscale to adjust image intensity value which can greatly singularize the region of interest. And then, classical threshold method (Otsu algorithm) is used to determine the best threshold value which can separate blood vessel from background. To test the improved method, comparative trial was set to testify the improvement. Finally, a series of DSA images were obtained to demonstrate this method and the final experimental results showed the effectiveness of the improved method.
AB - Interest is growing in the interventional surgery training system used before the treatment of vessel diseases. As one of the elementary component of the simulator, an accurate reconstruction of blood vessel obtaining from cross-sectional images is ugly needed. Digital Subtraction Angiography (DSA) data is used as a criterion for reconstruction result of blood vessel. In this paper, an improved threshold method is proposed to segment blood vessel from medical images. Firstly, with optimization characteristic, the genetic algorithm is used to determine the pre-segmentation greyscale from best histogram entropy. Secondly, we enlarge greyscale around the best greyscale to adjust image intensity value which can greatly singularize the region of interest. And then, classical threshold method (Otsu algorithm) is used to determine the best threshold value which can separate blood vessel from background. To test the improved method, comparative trial was set to testify the improvement. Finally, a series of DSA images were obtained to demonstrate this method and the final experimental results showed the effectiveness of the improved method.
KW - Best Histogram Entropy
KW - Blood Vessel Segmentation
KW - Genetic Algorithm
KW - Image Intensity
KW - Threshold Algorithm
UR - http://www.scopus.com/inward/record.url?scp=85050487171&partnerID=8YFLogxK
U2 - 10.1109/CBS.2017.8266103
DO - 10.1109/CBS.2017.8266103
M3 - Conference contribution
AN - SCOPUS:85050487171
T3 - 2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
SP - 221
EP - 225
BT - 2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
Y2 - 17 October 2017 through 19 October 2017
ER -
