Methods and apparatus for multi-view characterization

Patent metadata

Field	Value
Publication number	US-10746753-B2
Application number	US-201716072412-A
Country	US
Kind code	B2
Filing date	Jan 24, 2017
Priority date	Jan 28, 2016
Publication date	Aug 18, 2020
Grant date	Aug 18, 2020

Abstract

A model-based method of classifying a specimen in a specimen container. The method includes capturing images of the specimen and container at multiple different exposures times, at multiple different spectra having different nominal wavelengths, and at different viewpoints by using multiple cameras. From the captured images, 2D data sets are generated. The 2D data sets are based upon selection of optimally-exposed pixels from the multiple different exposure images to generate optimally-exposed image data for each spectra. Based upon these 2D data sets, various components are classified using a multi-class classifier, such as serum or plasma portion, settled blood portion, gel separator (if present), tube, air, or label. From the classification data and 2D data sets, a 3D model can be generated. Specimen testing apparatus and quality check modules adapted to carry out the method are described, as are other aspects.

First claim

What is claimed is:

1. A method of characterizing a specimen contained within a specimen container, comprising:
providing classified 2D data sets obtained by processing a plurality of 2D images of the specimen container containing a specimen taken from multiple viewpoints, the classified 2D data sets being classified as:
serum or plasma,
settled blood portion,
gel separator,
air,
tube, and
label;

correlating locations in the classified 2D data sets to a consolidated 3D data set; and
forming a consolidated 3D model based upon the consolidated 3D data set.

2. The method of claim 1 , wherein the plurality of 2D images are taken at multiple different exposure times at each of the multiple viewpoints.

3. The method of claim 2 , wherein the multiple different exposure times comprise between about 0.1 ms and about 256 ms.

4. The method of claim 2 , wherein the plurality of 2D images are taken at multiple different spectra having different nominal wavelengths.

5. The method of claim 4 , wherein the multiple different spectra comprise three or more wavelengths between about 400 nm and 700 nm.

6. The method of claim 4 , wherein the multiple different spectra comprise wavelengths of about 634 nm+/−35 nm, about 537 nm+/−35 nm, and about 455 nm+/−35 nm.

7. The method of claim 4 , wherein the classified 2D data sets are derived from optimally-exposed image data for each wavelength at multiple different exposure times.

8. The method of claim 1 , wherein the plurality of 2D images represent a 360 degree view of the specimen container that is based on multiple lateral images, with each lateral image overlapping adjacent images.

9. The method of claim 1 , wherein a number of the multiple viewpoints comprises 3 or more.

10. The method of claim 1 , comprising computing statistics of optimally-exposed pixels at different wavelengths to generate statistical data.

11. The method of claim 10 , wherein the computing statistics of the optimally-exposed pixels from optimally-exposed image data for the different wavelengths comprises calculating a mean value, a standard deviation, and/or covariance from a collection of corresponding pixels from each wavelength.

12. The method of claim 10 , wherein selection of the optimally-exposed pixels comprises selection of pixels from the images that include intensities of between about 180-254 based upon a range of 0-255.

13. The method of claim 10 , wherein a multi-class classifier is used to generate the classified 2D data sets.

14. The method of claim 13 , wherein a multi-class classifier comprises a support vector machine or a random decision tree.

15. The method of claim 13 , wherein the multi-class classifier is generated from multiple training sets.

16. The method of claim 1 , wherein the consolidated 3D model is displayed or stored.

17. The method of claim 1 , wherein the correlating locations in the classified 2D data sets to the consolidated 3D data set is based upon a virtual voxel grid for each classified 2D data set.

18. The method of claim 1 , wherein the classified 2D data sets are further classified as cap.

19. A quality check module adapted to characterize a specimen and specimen container, comprising:
a plurality of cameras arranged around the specimen container and configured to capture multiple images of the specimen container and specimen from multiple viewpoints, each of the plurality of cameras adapted to generate a plurality of 2D images taken at multiple different exposure times and multiple different wavelengths or one or more wavelength ranges;
a computer coupled to the plurality of cameras and adapted to process image data from the plurality of 2D images, the computer configured and capable of being operated to:
provide classified 2D data sets obtained by processing the plurality of 2D images taken from multiple viewpoints, the classified 2D data sets being classified as:
serum or plasma portion,
settled blood portion,
gel separator (if present),
air,
tube, and
label;

correlate locations in the 2D data sets to a consolidated 3D data set; and
form a consolidated 3D model based upon the consolidated 3D data set.

20. A specimen testing apparatus adapted to image a specimen contained within a specimen container, comprising:
a track;
a carrier on the track configured to contain the specimen container;
a plurality of cameras arranged around the track and configured to capture a plurality of 2D images of the specimen container and specimen from multiple viewpoints, each of the plurality of cameras configured to generate a plurality of images at multiple different exposure times and multiple different wavelengths or one or more wavelength ranges;
a computer coupled to the plurality of cameras and adapted to process image data from the plurality of 2D images, the computer configured and capable of being operated to:
provide classified 2D data sets obtained by processing the plurality of 2D images taken from the multiple viewpoints, the classified 2D data sets being classified as:
serum or plasma portion,
settled blood portion,
gel separator (if present),
air,
tube, and
label;

correlate locations in the 2D data sets to a consolidated 3D data set; and
form a consolidated 3D model based upon the consolidated 3D data set.

Assignees

• Siemens Healthcare Diagnostics Inc

Inventors

• Kluckner Stefan
• Chang Yao-Jen
• Chen Terrence
• Pollack Benjamin S

Classifications

• G01F23/804

  containing circuits handling parameters other than liquid level · CPC title

• G01N2015/055

  for hematocrite determination · CPC title

• G01N2035/1018

  Detecting inhomogeneities, e.g. foam, bubbles, clots · CPC title

• G01N2015/047

  by static multidetectors · CPC title

• G01N15/042

  by centrifuging and investigating centrifugates (centrifuges per se B04B) · CPC title