Vision-based force measurement using neural networks for biological cell microinjection.

Area of Science:

• Biophysics
• Biomaterials Science
• Robotics

Background:

• Accurate force measurement is crucial in micromanipulation of biological cells.
• Traditional force sensors are often challenging or infeasible to implement in microscale biological applications.
• Existing methods may lack precision or adaptability for diverse cell types.

Purpose of the Study:

• To develop and validate a vision-based force measurement method using artificial neural networks (ANNs) for bio-micromanipulation.
• To enable precise force quantification on spherical biological cells without direct physical contact.
• To provide a versatile tool applicable to various micromanipulation tasks and spherical elastic objects.

Main Methods:

• Utilizing an artificial neural network model trained with image processing techniques to estimate applied loads.
• Developing a bio-micromanipulation system to collect image and force data for ANN training.
• Extracting geometric features from cell images to quantify deformation during indentation.
• Training the ANN with visual data as input and measured force as output.

Main Results:

• The proposed ANN model accurately estimates applied forces to biological cells based on visual input.
• Geometric features effectively describe cell shape and deformation during micromanipulation.
• The system demonstrates capability for precise force sensing in bio-micromanipulation setups.
• The method is validated for zebrafish embryo membranes and applicable to other spherical elastic objects.

Conclusions:

• The vision-based ANN method provides a precise and adaptable solution for force measurement in bio-micromanipulation.
• This approach overcomes limitations of traditional force sensors in microscale biological applications.
• The findings support applications in cell injection, oocyte manipulation, and other delicate biological procedures.