Particle Classification through the Analysis of the Forward Scattered Signal in Optical Tweezers.

Area of Science:

• Biotechnology
• Optics
• Machine Learning

Background:

• Optical tweezers are crucial for manipulating micro-sized particles in biotechnology.
• Current particle classification relies on imaging, scattered light, or spectroscopy.
• A need exists for faster, more efficient classification methods.

Purpose of the Study:

• To develop a novel method for classifying trapped particles using optical tweezers.
• To leverage temporal scattering signal dynamics for identification.
• To validate the accuracy and efficiency of the proposed approach.

Main Methods:

• Utilized temporal data signals from laser scattering captured by a quadrant photodetector.
• Applied a pre-processing strategy combining Fourier transform and principal component analysis for feature extraction.
• Tested various machine learning algorithms for classification performance.

Main Results:

• Achieved classification accuracy of approximately 90% across multiple machine learning models.
• Demonstrated the feasibility of using temporal scattering signal dynamics for particle classification.
• Obtained high accuracy with short signal acquisition times (500 milliseconds).

Conclusions:

• The proposed method offers a viable alternative for particle classification in optical trapping.
• This approach enables faster and computationally efficient classification, suitable for real-time applications.
• The findings pave the way for enhanced optical trapping technologies in biotechnology.