| Presentation | 2009-07-15 Advanced Pancreas Segmentation from Three Dimensional Contrast Enhanced CT Images by Ensemble Learning Hiroaki OCHIAI, Akinobu SHIMIZU, Hidefumi KOBATAKE, Shigeru NAWANO, Kenji SHINOZAKI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper presents an ensemble segmentation to improve performance of pancreas segmentation from three dimensional contrast enhanced CT images. An ensemble segmentation is a process that consists of several hundred weak segmentation processes, which are combined into one using a machine learning algorithm, such as boosting. So far, various boosting algorithms have been proposed. In this paper, we focus on robustness against outlier, robustness against mislabel, cost for false positive and negative, convergence of learning algorithm and classify boosting algorithms into four groups according to the characteristics. We compare the performance of generated segmentation algorithms in each group and determine the best algorithm at each group. Finally, we combine the best algorithms to construct a pancreas segmentation process. In this study, we used three phase contrast enhanced CT volumes from 33 cases to train the segmentation algorithms and applied them to CT volumes from unknown 20 cases for testing. The paper shows the experimental results and discusses its effectiveness by comparing to our previous one. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | CT / pancreas / segmentation / boosting |
| Paper # | MI2009-52 |
| Date of Issue |
| Conference Information | |
| Committee | MI |
|---|---|
| Conference Date | 2009/7/8(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Medical Imaging (MI) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Advanced Pancreas Segmentation from Three Dimensional Contrast Enhanced CT Images by Ensemble Learning |
| Sub Title (in English) | |
| Keyword(1) | CT |
| Keyword(2) | pancreas |
| Keyword(3) | segmentation |
| Keyword(4) | boosting |
| 1st Author's Name | Hiroaki OCHIAI |
| 1st Author's Affiliation | Tokyo University of Agriculture and Technology() |
| 2nd Author's Name | Akinobu SHIMIZU |
| 2nd Author's Affiliation | Tokyo University of Agriculture and Technology |
| 3rd Author's Name | Hidefumi KOBATAKE |
| 3rd Author's Affiliation | Tokyo University of Agriculture and Technology |
| 4th Author's Name | Shigeru NAWANO |
| 4th Author's Affiliation | Center for Radiological Sciences, International University of Health and Welfare |
| 5th Author's Name | Kenji SHINOZAKI |
| 5th Author's Affiliation | National Kyusyu Cancer Center |
| Date | 2009-07-15 |
| Paper # | MI2009-52 |
| Volume (vol) | vol.109 |
| Number (no) | 127 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |