System and method for monitoring particles in a fluid using ratiometric cytometry
A particle detection system with a detection mechanism that includes detectors positioned to detect two different ranges of fluorescence produced by particles in the fluid in a flow chamber. Each of the detectors is arranged to generating a trigger signal whenever fluorescence is detected. The system and related method enhance the accuracy and sensitivity of blue-green algae monitoring by utilizing imaging flow cytometry combined with particle analysis and the measurement of the ratio of each particle's phycocyanin to chlorophyll b detected by using the two detectors configured for detection of two different fluorescence ranges, one associated with the phycocyanin and the other associated with the chlorophyll b. Captured images are be used in comparison to known images of a library of images using a support vector machine classifier.
1. A system for analyzing particles in a fluid, the system comprising:
a flow chamber for receiving a sample of fluid to be analyzed;
one or more detectors positioned to detect two different ranges of fluorescence produced by particles in the fluid in said flow chamber, said detectors generating a trigger signal whenever fluorescence is detected;
a classifier training set containing organism type and feature values, wherein the classifier training set comprises a matrix having at least one row of each image of a library of images with particle type and at least another row of feature values for particles; and
a support vector machine classifier training set including related particle images,
wherein feature values of captured images are calculated and the support vector machine classifier training set is applied to captured images to classify the particle associated with each image.
2. The system of claim 1 further comprising a light source arranged to direct excitation light onto said flow chamber.
3. The system of claim 2 further comprising a high intensity light source positioned to backlight said flow chamber in response to said trigger signal.
4. The system of claim 1 further comprising a video system positioned to image particles in the fluid passing through said flow chamber; and
imaging optics for focusing light from said flow chamber onto said video system, wherein said video system images particles in the fluid passing through said flow chamber in response to said trigger signal,
wherein the video system includes means for storing and analyzing particle images captured by said video system.
5. The system of claim 4 wherein the fluid is flowing through said chamber and said video system automatically images particles in the fluid passing through said flow chamber at a predetermined time interval.
6. The system of claim 4 wherein said video system includes a digital camera or a video camera and a framegrabber.
7. The system of claim 1 wherein the flow chamber includes a flow channel configured such that its dimensions match the field of view associated with the detectors.
8. The system of claim 1 wherein the feature values of the matrix comprise measures of size, shape, color, surface characteristics and ratio of light intensity.
9. A method for analyzing particles in a fluid, the method comprising the steps of:
directing a sample of fluid to be analyzed into a flow chamber having a chamber depth;
detecting two different ranges of fluorescence produced by particles in the fluid in said flow chamber;
storing fluorescence data obtained in said detecting step;
creating a classifier training set using organism type and feature values, wherein the classifier training set comprises a matrix having at least one row of each image of a library of images with particle type and at least another row of feature values for particles;
building a support vector machine classifier training set related to particle images;
calculating feature values of captured images; and
applying the support vector machine classifier training set to captured images to classify the organism associated with each image.
10. The method of claim 9 further comprising the steps of:
using imaging optics to focus light from said flow chamber onto a video system;
imaging particles in the fluid in said flow chamber with said video system; and
analyzing images captured by said video system.
11. The method of claim 10 wherein the step of imaging particles in the fluid includes:
directing excitation light onto said flow chamber;
generating a trigger signal whenever fluorescence is detected; and
imaging the particles in the fluid in response to said trigger signal.
12. The method of claim 10 wherein the step of analyzing images captured by said video system includes determining the size of each imaged particle.
13. The method of claim 10 wherein the step of analyzing images captured by said video system includes determining the peak fluorescence of each imaged particle.
14. The method of claim 10 wherein the fluid is flowing in said flow chamber and the step of analyzing images captured by said video system includes determining the time of particle passage for each imaged particle.
15. The method of claim 14 wherein the step of analyzing images captured by said video system includes determining the image location for each imaged particle.
16. The method of claim 10 wherein the step of analyzing images includes:
displaying an interactive scattergram comprising a graph plotting particle size against particle fluorescence on a display;
selecting a particular region of said scattergram; and
displaying the images in said selected region on said display.
17. The method of claim 9 wherein the feature values of the matrix comprise measures of size, shape, color, surface characteristics and ratio of light intensity.