Robust and highly performant ring detection algorithm for 3d particle tracking using 2d microscope imaging.

Area of Science:

• Microscopy and imaging techniques
• Biophysics and cellular dynamics
• Fluid dynamics and microfluidics

Background:

• Three-dimensional (3D) particle tracking is crucial for understanding microscopic dynamics in fields like fluid dynamics, cell biology, and microbiology.
• Determining 3D particle positions from 2D imaging is possible by analyzing diffraction rings from out-of-focus fluorescent particles.

Purpose of the Study:

• To present a novel ring detection algorithm for robust and efficient 3D particle tracking.
• To enable real-time analysis of multiple particle trajectories with high accuracy.

Main Methods:

• The algorithm is based on the circle Hough transform, adapted to detect diffraction rings from fluorescent particles.
• It employs a classification approach to handle varying numbers of particles and overcome challenges in complex parameter spaces.
• The method is optimized for high performance and low memory usage, utilizing open-source software.

Main Results:

• The algorithm demonstrates a high detection rate, exceeding 94%, with a low false-detection rate of only 1%.
• It successfully tracks particles in real-time (70 Hz) within a microfluidic experiment, even when particles are close or signals overlap.
• The algorithm shows robustness against ring occlusion, inclusions, and overlaps.

Conclusions:

• The developed algorithm provides an efficient and accurate solution for real-time 3D multi-particle tracking.
• Its robustness and performance make it suitable for various applications involving microscopic dynamics and noisy data.
• The open-source implementation facilitates distribution and modification for broader scientific use.