masterhead masterhead  masterhead

High-speed 3D Tracking of Chlamydomonas with Dark-Field Microscope


Based on a high-speed 3D tracking of chlamydomonas with phase-contrast microscope, we are trying 3D tracking with dark-field microscope in cooperation with Associate Professor Wakabayashi, Chemical Resources Laboratory, Tokyo Institute of Technology, and Professor Kikkawa, Department of Cell Biology and Anatomy, Graduate School of Medicine, the University of Tokyo.

A fact is known that some organisms have phototaxis or chemotaxis, which is considered to be closely related to biological activity. As one of them, when chlamydomonas, which is a kind of swimming cells, indecates the phototaxis, we are investigating a relation between a direction of an eyespot or a motion of a flagellum and a light position. A dark-field microscopy is suitable to observe the eyespot and the flagellum of the chlamydomonas, and 3D tracking technique under dark-field microscopy is desirable for precise observation. Therefore, we aim to three-dimensionally track the freely swimming chlamydomonas with a microorganism tracking system, and to observe the eyespot and the motion of the flagellum in detail.

Because we observe scattered light by a sample in the dark-field microscopy, an image is different from those by other observation methods which observe transmitted light (see figure below). Therefore, it is impossible to simply apply a DFDi method which we have developed. Thus, we have developed a method to estimate a depth position of a measuring object under dark-field microscopy. This depth position estimation is based on a correlation between a variance of a pixel value near a cell center and the depth position at a vicinity of a focal plane.

We have experimentally confirmed that this method has sufficient precision of the depth position estimation for the tracking control. We have succeeded the 3D tracking of chlamydomonas under dark-field microscopy by controlling the focus position by the proposed method, extracting the cell position by image processing in image plane direction, and conducting feedback control of an automatic stage with a container position including the target.

the comparison of three images observed by each different method

Different image characteristics in each observation method


A movie captured at 1000 fps, and played at 30 fps (1/33 speed)
Ishikawa Senoo Laboratory, Department of Information Physics and Computing, Department of Creative Informatics,
Graduate School of Information Science and Technology, University of Tokyo
Ishikawa Senoo Laboratory WWW admin:
Copyright © 2008 Ishikawa Senoo Laboratory. All rights reserved.