Method and apparatus for automated platelet identification within a whole blood sample from microscopy images

Patent metadata

Field	Value
Publication number	US-10061973-B2
Application number	US-201615270307-A
Country	US
Kind code	B2
Filing date	Sep 20, 2016
Priority date	Dec 30, 2011
Publication date	Aug 28, 2018
Grant date	Aug 28, 2018

Abstract

A method and apparatus for identifying platelets within a whole blood sample. The method includes the steps of: a) adding at least one colorant to the whole blood sample, which colorant is operable to tag platelets; b) disposing the blood sample into a chamber defined by at least one transparent panel; c) imaging at least a portion of the sample quiescently residing within the chamber to create one or more images; and d) identifying one or more platelets within the sample using an analyzer adapted to identify the platelets based on quantitatively determinable features within the image using a analyzer, which quantitatively determinable features include intensity differences.

First claim

What is claimed is:

1. A method for identifying platelets within a substantially undiluted whole blood sample, comprising:
adding at least one fluorescent colorant to the whole blood sample which colorant is operable to tag the platelets;
imaging at least a region of the whole blood sample quiescently residing within a chamber to create at least one region image, which region image has an area, and which imaging includes illuminating at least a portion of the quiescent sample with one or more first wavelengths of light in a manner that causes the fluorescent colorant to emit at least one second wavelength of light, and which image comprises per incremental units, each per incremental unit having an intensity value representative of the at least one second wavelength of light emitted relative to that per incremental unit;
identifying one or more peak intensity values within local sub-regions of the at least one region image using the per incremental unit intensity values, which local sub-regions are each of a predetermined area less than the region area; and
identifying one or more platelets within the whole blood sample, which identifying includes comparing per incremental unit intensity values in a respective local sub-region to distinguish per incremental unit intensity values associated with plasma from per incremental unit intensity values associated with platelets.

2. The method of claim 1 , further comprising identifying anomaly image portions and estimating the number of platelets in the anomaly image portions.

3. The method of claim 1 , wherein the identifying one or more platelets within the whole blood sample further includes using one or more determined directional intensity contrasts.

4. The method of claim 3 , wherein the one or more determined directional intensity contrasts each are based on an incremental decrease in intensity within at least one image in a respective direction extending away from the respective local peak intensity value.

5. The method of claim 1 , wherein the identifying one or more platelets within the whole blood sample includes evaluating the at least one region image to determine a presence of one or more platelet clusters within the sample.

6. The method of claim 1 , wherein each per incremental unit is at least one pixel, and each local sub-region of a predetermined area includes a determined number of pixels.

7. The method of claim 1 , wherein the identifying one or more platelets within the whole blood sample using the identified local peak intensity values within the at least one region image further includes determining a Gaussian distribution of a decrease in the per incremental unit intensity values surrounding a respective one of the local peak intensity values.

8. The method of claim 7 , further comprising:
determining an outer region perimeter of a platelet candidate in a particular local sub-region;
comparing the per incremental unit intensity value of one or more per incremental units located at the outer region perimeter to one or more per incremental unit intensity values associated with plasma within the particular local sub-region; and
determining the presence or absence of a cluster of platelets based on the comparison of the per incremental unit intensity value of one or more per incremental units located at the outer region perimeter to the one or more per incremental unit intensity values associated with plasma within the particular local sub-region.

9. The method of claim 8 , further comprising determining the number of platelets within the cluster of platelets.

10. The method of claim 1 , wherein the identifying one or more platelets within the whole blood sample includes determining the per incremental units of the image associated with each of the one or more platelets.

11. The method of claim 10 , wherein the identifying one or more platelets within the whole blood sample further includes determining an area of a respective one of the one or more platelets using the determined per incremental units of the image associated with the respective platelet.

12. The method of claim 10 , wherein the identifying one or more platelets within the whole blood sample further includes determining one or both of a major axis length and a minor axis length of a respective one of the one or more platelets using the determined per incremental units of the image associated with the respective platelet.

13. An apparatus for identifying platelets within a substantially undiluted whole blood sample, which whole blood sample includes at least one fluorescent colorant operable to tag the platelets, the apparatus comprising:
an analysis chamber configured to quiescently hold the sample; and
an analysis device having a programmable analyzer, the analyzer programmed with stored instructions that when executed cause the analysis device to:
image at least a portion of the whole blood sample quiescently residing within the analysis chamber to create at least one image, which imaging includes illuminating at least a portion of the quiescent sample with one or more first wavelengths of light in a manner that causes the fluorescent colorant to emit at least one second wavelength of light, and which image comprises per incremental units, each per incremental unit having an intensity value representative of the at least one second wavelength of light emitted relative to that per incremental unit;
identify one or more peak intensity values within local areas of the at least one image using the per incremental unit intensity values, which local regions are of a predetermined area; and
identify one or more platelets within the whole blood sample, which identifying includes comparing per incremental unit intensity values in a respective local sub-region to distinguish per incremental unit intensity values associated with plasma from per incremental unit intensity values associated with platelets.

Assignees

• Abbott Point Of Care Inc

Inventors

• Yu Changhua
• Xie Min
• Wu Yiming

Classifications

• G06T2207/30242

  Counting objects in image · CPC title

• G01N33/86 Primary

  involving blood coagulating time {or factors, or their receptors} · CPC title

• G06K9/00147 Primary

  Physics · mapped topic

• G06T7/0012

  Biomedical image inspection · CPC title

• G06K2209/05

  Physics · mapped topic