2022 Fiscal Year Final Research Report
Development of a cell tracking algorithm using deep learning
Project/Area Number |
20H03244
|
Research Category |
Grant-in-Aid for Scientific Research (B)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Review Section |
Basic Section 43060:System genome science-related
|
Research Institution | Keio University |
Principal Investigator |
Funahashi Akira 慶應義塾大学, 理工学部(矢上), 教授 (70324548)
|
Project Period (FY) |
2020-04-01 – 2023-03-31
|
Keywords | 画像解析 / 機械学習 / 深層学習 / 細胞系譜 / 細胞追跡 / 発生・分化 |
Outline of Final Research Achievements |
Using deep learning, we have developed an image processing algorithm that can accurately track both cell migration and cell division in 3D time-series fluorescence microscopy images of mouse embryonic development, which was previously difficult. Existing tracking algorithms have difficulty detecting both cell migration and cell division simultaneously, and the accuracy of cell division tracking is very low. In this research project, we developed a tracking algorithm that combined deep learning and integer programming and accurately tracked cells up to the 40-cell stage.
|
Free Research Field |
定量生物学
|
Academic Significance and Societal Importance of the Research Achievements |
学術的意義として、当研究課題で開発したOHモデルおよびMHモデルは細胞分裂を含む細胞動態を90%近い精度で細胞追跡を行うことに成功し、初期マウス発生過程における胚の1細胞ごとの動態を定量的に比較することが可能であることが示された点が挙げられる。 社会的意義としては本研究課題で提案された細胞追跡アルゴリズムを用いることで、経験的に定められた指標に代わる、産仔作出能との関連性が高い「胚の質を評価し得る指標」の確立が期待される点が挙げられる。
|