AUTHOR=Zhou Zhengsong , Wan Hongli , Zhang Haoyu , Chen Xumiao , Wang Xiaoyu , Lili Shiluo , Zhang Tao 

TITLE=Segmentation of Spontaneous Intracerebral Hemorrhage on CT With a Region Growing Method Based on Watershed Preprocessing

JOURNAL=Frontiers in Neurology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.865023

DOI=10.3389/fneur.2022.865023

ISSN=1664-2295

ABSTRACT=Intracerebral hemorrhage (ICH) poses a great threat to human life due to its high incidence and poor prognosis. Identification of bleeding location and quantification of the volume based on CT images are of great significance for assisting the diagnosis and treatment of ICH. In this study, a region growing algorithm based on watershed preprocessing (RG-WP) was proposed to segment and quantify the hemorrhage. The lowest points yielded by watershed algorithm were used as seed points for region growing, then hemorrhage was segmented based on region growing method. At the same time, in order to integrate the rich experience of clinicians with the algorithm, seed points selecting manually on the basis of watershed segmentation was remained. With application to segmentation on CT images of 55 ICH patients, performance of the RG-WP algorithm was evaluated by comparing it with manual segmentations delineated by professional clinicians as well as the traditional ABC/2 method and the deep learning algorithm U-net. The mean deviation of hemorrhage volume of the RG-WP algorithm from manual segmentation was -0.12mL (range: -1.05, 1.16), while of the ABC/2 from the manual was 1.05mL (range: -0.77, 9.57). Strong agreement of the algorithm and the manual was confirmed with a high intraclass correlation coefficient (ICC) (0.998, 95% CI: 0.997-0.999), which was superior to that of the ABC/2 and the manual (0.972, 95% CI: 0.953-0.984). The Sensitivity (Sen), positive predictive value (PPV), Dice Similarity index (DSI), and Jaccard index (JI) of the RG-WP algorithm compared to manual were 0.92 ± 0.04, 0.95 ± 0.04, 0.93 ± 0.02, and 0.88 ± 0.04, respectively, showing high consistency. Besides, accuracy of the algorithm was also comparable to that of the deep learning method U-net, with Sen, PPV, DSI and JI of 0.91 ± 0.09, 0.91 ± 0.06, 0.91 ± 0.05, and 0.91 ± 0.06, respectively.