February 1, 2006

Engineers Use Computer Game Approach to Automate In Vitro Fertilization

  — The process would reduce cost and improve consistency of the cell micro-injection process

Leonardo Mattos uses a computer screen and joystick to control micro-manipulation tools for in vitro fertilization. (Photos: Becky Kirkland)
Dr. Edward Grant demonstrates the current way that cell micro-injection is performed using a microscope and manually controlled micro-manipulation tools. (Photos: Becky Kirkland)

From stem cells to in vitro fertilization to transgenic mice, cell micro-injection is the first step that makes biomedical research in these areas possible. Cell micro-injection is the process of selecting a single cell and injecting it with DNA material. At present the process is a manual, time-consuming and tedious process that can take technicians as much as a year to learn. It is one reason that biomedical research is so expensive, with each transgenic mouse costing as much as $65,000, for example.

At North Carolina State University, engineers are working to create a new system for cell micro-injection that uses computer games and robotics to semi-automate the process, reducing the time and labor factors and increasing the accuracy and efficiency, which will eventually lower the cost of the genetically altered cells and transgenic mice.

	“Basically we would be capturing the soul of the individual and putting it in the robot.”

Currently the micro-injection process requires technicians to sit at a microscope attached to a micro-manipulation system and manually maneuver a tiny pipette to collect a single cell while manually maneuvering a very tiny needle to inject the material into the cell. The risk of error is high even for highly skilled technicians.

Dr. Edward Grant, professor of electrical and computer engineering and biomedical engineering and director of the Center For Robotics and Intelligent Machines, and Leonardo Mattos, a doctoral student in electrical and computer engineering, have created an interactive computer interface linked to the micro-manipulation system that treats the micro-injection process like a computer game, complete with joystick and onscreen images. The researchers are using machine learning techniques and experimental results to develop a controller for the micro-injection process that can be used to train technicians more quickly by creating a virtual environment for practicing techniques and honing skills.

The semi-automation will also increase efficiency and accuracy since the computer can be programmed to correct for human error. Using the joystick, a technician can manipulate a cell with the pipette and inject it with material without ever having to look through the microscope and without having to spend months learning the skill.

“Eventually we plan to fully automate the process by having users ‘teach’ the robot and computer how to perform everything from cell selection to injection of the material,” said Grant. “Basically we would be capturing the soul of the individual and putting it in the robot.”

The engineers are currently refining the system, which uses a knowledge-based controller, and will begin training the robot using different individuals at different skill levels.

Contributions to the project have been made by Dr. John Muth, assistant professor of electrical and computer engineering at NC State. The project is a joint research project with Dr. Randy Thresher and Dr. Al Banes, both from the University of North Carolina – Chapel Hill, through the NC State and UNC–CH joint Department of Biomedical Engineering.

— weston —

Media contact:
Jennifer Weston, (919) 515-3848, weston@ncsu.edu

Technical contact:
Dr. Edward Grant, (919) 515-7354, egrant@eos.ncsu.edu

/ News Index / News Archives Index /