| Presentation | 2018-09-28 Estimation of postmortem time for CT images by using texture analysis Yasushi Hirano, Shoji Kido, Kazuyuki Kinoshita, Kunihiro Inai, Sakon Noriki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The dissection rate for dead human bodies with suspicious death is about 11% in Japan. This rate is markedly lower than other developed countries. This issue is caused by lack of medical doctors who have permissions for internal examinations of dead bodies, resistance of the bereaved for internal examinations and so on. The causes of death and the postmortem time are often estimated by observing surface of their bodies and their medical records, and the reliability of this method is not high enough. Therefore, opportunities to be performed Autopsy Imaging (Ai) have increased. Ai is a method to estimate causes of death and the postmortem time by reading CT images and/or MR images of dead bodies. However, there is a difficulty to widely perform Ai, because there is little number of radiologists who have enough knowledge about postmortem change for reasonable diagnosis. In this study, we propose a method to estimate postmortem time by texture analysis of postmortem change in CT image. First, we calculated texture features in lung regions extracted from postmortem CT images taken at hourly intervals of miniature pigs. Next, we created a model for estimating postmortem time using the Regression Forest. Finally, we estimated the postmortem time for human by using this model. As a result, it is shown the postmortem time was estimated with a mean absolute error of 1.80 hours. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Estimation of postmortem time / Texture analysis / Chest CT images / Autopsy imaging (Ai) |
| Paper # | MI2018-35 |
| Date of Issue | 2018-09-21 (MI) |
| Conference Information | |
| Committee | MI |
|---|---|
| Conference Date | 2018/9/28(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Tokyo Women's Medical University |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Medical Imaging, Utilization of Medical Image Processing for Surgery, etc. |
| Chair | Kensaku Mori(Nagoya Univ.) |
| Vice Chair | Yoshiki Kawata(Tokushima Univ.) / Yuichi Kimura(Kinki Univ.) |
| Secretary | Yoshiki Kawata(Aichi Inst. of Tech.) / Yuichi Kimura(Nagoya Inst. of Tech.) |
| Assistant | Ryo Haraguchi(Univ. of Hyogo) / Yasushi Hirano(Yamaguchi Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Medical Imaging |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Estimation of postmortem time for CT images by using texture analysis |
| Sub Title (in English) | |
| Keyword(1) | Estimation of postmortem time |
| Keyword(2) | Texture analysis |
| Keyword(3) | Chest CT images |
| Keyword(4) | Autopsy imaging (Ai) |
| 1st Author's Name | Yasushi Hirano |
| 1st Author's Affiliation | Yamaguchi University(Yamaguchi Univ.) |
| 2nd Author's Name | Shoji Kido |
| 2nd Author's Affiliation | Yamaguchi University(Yamaguchi Univ.) |
| 3rd Author's Name | Kazuyuki Kinoshita |
| 3rd Author's Affiliation | University of Fukui(Univ. of Fukui) |
| 4th Author's Name | Kunihiro Inai |
| 4th Author's Affiliation | University of Fukui(Univ. of Fukui) |
| 5th Author's Name | Sakon Noriki |
| 5th Author's Affiliation | University of Fukui(Univ. of Fukui) |
| Date | 2018-09-28 |
| Paper # | MI2018-35 |
| Volume (vol) | vol.118 |
| Number (no) | MI-225 |
| Page | pp.pp.11-15(MI), |
| #Pages | 5 |
| Date of Issue | 2018-09-21 (MI) |