Particle Trajectory Imaging With AI for Fast Motility Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for characterizing motile cellular particles, such as spermatozoa, are computationally expensive and time-consuming, especially when dealing with large numbers of particles, requiring significant computational resources and time to track and characterize each particle.
Innovation Solution
A method utilizing defocused or lensless imaging modalities in conjunction with a supervised learning artificial intelligence algorithm, such as a convolutional neural network, to analyze particle trajectories in a sample, reducing computational requirements by combining multiple images to form a single trajectory image and calculating average displacement parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional tracking methods are used to characterize each particle individually, then measurement precision is improved, but productivity deteriorates due to computational expense and time consumption
Solution Approach 1:
The patent combines multiple individual particle trajectory images into a single composite trajectory image that contains information from all particles. This merging allows a single AI algorithm execution to analyze all particles simultaneously, transforming the computational complexity from O(N) separate tracking operations to a single integrated analysis, thereby dramatically improving productivity while maintaining measurement precision through the AI algorithm's ability to extract individual trajectory parameters from the composite image
Solution Approach 2:
The patent creates a composite trajectory image that serves as a simplified representation containing essential motion information from all particles without requiring separate processing of each particle's full trajectory data. This copying approach preserves the necessary measurement information while reducing computational burden, enabling fast analysis without sacrificing characterization accuracy
2Quantity of substance
If the number of particles to be tracked increases, then quantity of substance improves, but device complexity worsens due to increased computational requirements
Solution Approach 1:
The patent merges all particle trajectories into a single composite image, allowing the AI algorithm to process all particles in one operation regardless of particle count. This approach decouples computational complexity from the number of particles, enabling the system to handle large quantities of particles without proportionally increasing computational requirements or device complexity
Solution Approach 2:
The patent transitions from processing particle data in temporal sequence (multiple time points for each particle) to spatial representation (all particles simultaneously in a composite image). This dimensional transformation allows parallel processing of all particles, reducing computational complexity from linear scaling with particle count to a constant-time operation relative to particle number
3Measurement precision
If focused imaging is used to achieve high spatial resolution, then measurement precision improves, but the field of view and depth of field are limited
Solution Approach 1:
The patent combines multiple defocused or lensless images captured over time into a composite trajectory image that preserves sufficient spatial information for trajectory analysis. This merging approach allows the use of defocused imaging (which provides larger field of view and depth of field) while still achieving accurate particle characterization through the AI algorithm's ability to extract motion parameters from the composite image
Solution Approach 2:
The patent changes the imaging parameter from focused to defocused or lensless mode, trading off some spatial resolution for increased field of view and depth of field. The AI algorithm compensates for the reduced spatial resolution by learning to extract accurate trajectory information from the defocused images, thereby maintaining measurement precision while expanding the observable area
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method significantly reduces analysis time and computational expense while effectively characterizing large numbers of motile particles, providing accurate estimates of velocity and motility characteristics with high correlation to ground truth values.
Implementation Method 1
acquisition of at least one image of the sample during an acquisition period, using an image sensor
Implementation Method 2
The microscope has an objective lens defining an object plane, extending into the sample, and an image plane, coinciding with the detection plane of an image sensor
Implementation Method 3
each image of the sample is acquired according to a defocused imaging modality or according to a lensless imaging modality, so that each particle forms, on each image, a diffraction pattern
Data Source
Figure 1~2
Figure 3~4
Figure 5A~5B
AI summary
A method for characterizing at least one mobile particle (10i) in a sample (10), the method comprising: a) acquiring at least one image (I, In) of the sample during an acquisition period, using an image sensor (20), defining an observation field, the acquisition period comprising different acquisition times (tn); b) using the image or each image resulting from a), forming a trajectory image (I) representing the particles of the sample, in the observation field, at the different acquisition times; c) using the trajectory image resulting from b) as an input image for a detection algorithm, programmed to detect particles, and for a supervised learning artificial intelligence algorithm, programmed to calculate at least one average displacement parameter for different detected particles.