Development of Medical Diagnosis Supporting System for Endoscope Image and Cell Image
| Project/Area Number |
17K00252
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Chubu University |
|---|
Principal Investigator |
IWAHORI Yuji 中部大学, 工学部, 教授 (60203402)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2018: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2017: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Keywords | コンピュータビジョン / Shape from Shading / 内視鏡画像 / CNN / 血管検出 / ポリープ検出 / パターン分類 / 良性・悪性 / 細胞画像 / 良性・悪性分類 / シーン分類 / 医療画像 / 知覚情報処理 / 機械学習 |
|---|
| Outline of Final Research Achievements |
This research developed an improved approach to recover the shape and size of polyp from endoscope images using the blood vessel information to estimate the moving distance of endoscope into the depth direction and that of the surface reflectance parameter as a reliable approach. Other approach to recover the shape and size of polyp using medical suture was also developed using the geometrical relation from only one image. Further extension using blood vessel instead of the medical suture was also developed. The research also challenged the image classification problem of benign or malignant for the observed polyp using three kinds of endoscope images with the higher accuracy and cell detection and cell classification of benign or malignant in a microscope cell image.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では一般的な白色内視鏡画像でポリープの3次元形状のみならずその絶対的な大きさを復元するための新たな手法を研究し，開発した．従来はステレオ構造やレーザーレンジファインダーを応用した特殊なハードウェアベースの内視鏡を開発した例があるが，通常内視鏡で形状と大きさを推定する手がかりが必要であった．研究ではこのため，手術用縫合糸やさらに血管の情報を用いて，高精度に検出をするCNNベースの方法のほか，対応点抽出問題や画像分類問題においてもCNNと転移学習を用いて精度向上を図る手法として，３種類の内視鏡画像を用いて精度向上を図ったほか，細胞画像の検出・分類問題においても新規で有効な手法を提案した．
|
Report
(4 results)
Research Products
(145 results)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Journal Article] Classification of Cell Nuclei using CNN Features2019
Author(s)
Yuji Iwahori, Yuya Tsukada, Takashi Iwamoto, Kenji Funahashi, Jun Ueda, M. K. Bhuyan
-
Journal Title
International Conference on Intelligence Science, Springer Studies in Computational Intelligence (SCI) (Selected Paper among IEEE/ACIS ICIS 2019)
Volume: 849
Pages: 195-208
DOI
ISBN
9783030252120, 9783030252137
Related Report
Peer Reviewed / Int'l Joint Research
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Presentation] Estimation of User Location for Hearing-Dog Robot2017
Author(s)
Hoshito Kudo, Satoshi Tanaka, Yukihiro Yoshida, Tsuyoshi Nakamura, Masayoshi Kanoh, Koji Yamada, Daimu Oiwa, Yuji Iwahori, Shinji Fukui
Organizer
IFSA-SCIS 2017 (Joint 17th World Congress of International Fuzzy Systems Association and 9th International Conference on Soft Computing and Intelligent Systems)
Related Report
Int'l Joint Research
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
