Automatic tracking of multiple cell regions in 4D live-cell imaging data

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K16021
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: Kanazawa University
• Principal Investigator: Hirose Osamu 金沢大学, 電子情報学系, 助教 (30549671)
• Research Collaborator: Kawaguchi Shotaro 金沢大学 ; Tokunaga Terumasa 九州工業大学 ; Yoshida Ryo 統計数理研究所 ; Toyoshima Yu 東京大学 ; Teramoto Takayuki 九州大学 ; Kuge Sayuri 九州大学 ; Ishihara Takeshi 九州大学 ; Iino Yuichi 東京大学
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 物体追跡 / 粒子フィルタ / マルコフ確率場 / ベイズ推定 / 人工知能 / 機械学習 / 4Dライブセルイメージングデータ

Outline of Final Research Achievements

In this research, we aimed at developing the methods for automating the detection and tracking of cell regions in 4D live-cell imaging data. Cells in 4D live-cell imaging data are often imaged as ellipsoidal shapes and are densely distributed. For this data, standard methods usually fail to automatic tracking because of cell-switching and coalescence of tracked positions. To address this issue, we utilized typical characteristics in 4D live-cell imaging data; movements of nearly-located cells are strongly correlated. By using the characteristics used as the information for predicting cells' positions, we succeeded to improve tracking performance drastically. The software developed in this research is being distributed on the project website.

Free Research Field

人工知能