Apparatus and methods of training models of diagnostic analyzers

Patent metadata

Field	Value
Publication number	US-12475681-B2
Application number	US-202017755477-A
Country	US
Kind code	B2
Filing date	Oct 22, 2020
Priority date	Oct 31, 2019
Publication date	Nov 18, 2025
Grant date	Nov 18, 2025

Abstract

A method of training a model of a diagnostic apparatus includes providing one or more first tube assemblies of a first type and one or more second tube assemblies of a second type in a diagnostic apparatus; capturing one or more first images of at least a portion of each of the one or more first tube assemblies and the second tube assemblies using the imaging device. Training the model includes identifying tube assemblies of the first type and tube assemblies of the second type based on the one or more first images and the one or more second images. Tubes assemblies of the first type are grouped into a first group and tube assemblies of the second type are grouped into a second group. Other methods and apparatus are disclosed.

First claim

What is claimed is:

1 . A method of training a model of a diagnostic apparatus, comprising:
providing one or more first tube assemblies of a first type in a diagnostic apparatus; providing one or more second tube assemblies of a second type in the diagnostic apparatus; capturing one or more first images of at least a portion of each of the one or more first tube assemblies using an imaging device; capturing one or more second images of at least a portion of each of the one or more second tube assemblies using the imaging device; training a model to identify tube assemblies of the first type and tube assemblies of the second type based on one or more physical characteristics of a tube assembly identified in the one or more first images and the one or more second images; grouping tube assemblies of the first type into a first group by the trained model, the first group associated with first tests to be performed by the diagnostic apparatus or a first chemical additive or contents in the first type; and grouping tube assemblies of the second type into a second group by the trained model, the second group associated with second tests to be performed by the diagnostic apparatus or a second chemical additive or contents in the second type.

2 . The method of claim 1 , further comprising:
transporting the one or more first tube assemblies to the imaging device located in the diagnostic apparatus; and transporting the one or more second tube assemblies to the imaging device.

3 . The method of claim 1 , wherein providing the first tube assemblies and providing the second tube assemblies comprises locating the one or more first tubes assemblies of the first type in a first container and locating the one or more second tube assemblies of the second type in a second container.

4 . The method of claim 1 , comprising:
providing one or more third tube assemblies of a third type in the diagnostic apparatus; and capturing one or more third images of at least a portion of each of the one or more third tube assemblies using the imaging device, wherein training the model comprises training the model to identify tube assemblies of the third type based on the one or more third images, and further comprising grouping tube assemblies of the third type into a third group.

5 . The method of claim 4 , comprising grouping tube assemblies of the first type and tube assemblies of the third type in the first group and tube assemblies of the second type in the second group.

6 . The method of claim 1 , wherein training the model comprises training a discriminative model.

7 . The method of claim 1 , wherein training the model comprises training a neural network.

8 . The method of claim 1 , wherein training the model comprises training a convolutional neural network.

9 . The method of claim 1 , wherein training the model comprises training a support vector machine.

10 . The method of claim 1 , wherein training the model comprises analyzing the one or more first images to identify at least one color, geometric, material, or dimensional-gradient characteristic of the one or more first tube assemblies and analyzing the one or more second images to identify at least one color, geometric, material, or dimensional-gradient characteristic of the one or more second tube assemblies.

11 . The method of claim 10 , wherein the one or more first tube assemblies comprise a cap and wherein the at least one characteristic is a color of the cap.

12 . The method of claim 11 , further comprising displaying images of at least one color of a cap of the one or more first tube assemblies.

13 . The method of claim 10 , wherein the one or more first tube assemblies comprise a cap and wherein the at least one characteristic is a geometric feature of the cap.

14 . The method of claim 10 , wherein the one or more first tube assemblies comprise a cap and wherein the at least one characteristic is opacity of the cap.

15 . The method of claim 14 , wherein the at least one characteristic is opacity of at least one wavelength of light.

16 . The method of claim 1 , wherein training the model generates a trained model, and further comprising transporting the trained model to another diagnostic apparatus.

17 . The method of claim 1 , wherein tube assemblies in the first group are used for a first analysis of biological samples and tube assemblies in the second group are used for a second analysis of biological samples.

18 . A method of operating a diagnostic apparatus, comprising:
training a model of the diagnostic apparatus, comprising:
providing one or more first tube assemblies of a first type in a diagnostic apparatus;
providing one or more second tube assemblies of a second type in the diagnostic apparatus;
capturing one or more first images of at least a portion of each of the one or more first tube assemblies using an imaging device;
capturing one or more second images of at least a portion of each of the one or more second tube assemblies using the imaging device; and
training the model to identify tube assemblies of the first type and tube assemblies of the second type based on one or more physical characteristics of a tube assembly identified in the one or more first images and the one or more second images;
grouping tube assemblies of the first type into a first group by the trained model, the first group associated with first tests to be performed by the diagnostic apparatus or a first chemical additive or contents in the first type; grouping tube assemblies of the second type into a second group by the trained model, the second group associated with second tests to be performed by the diagnostic apparatus or a second chemical additive or contents in the second type; loading one or more tube assemblies containing specimens located therein into the diagnostic apparatus; imaging the one or more tube assemblies containing specimens; identifying the one or more tube assemblies containing specimens as being of the first type or the second type using the model; and grouping the one or more tube assemblies containing specimens into the first group or the second group based on the identifying.

19 . The method of claim 18 , wherein training the model comprises analyzing the one or more first images to identify at least one color, geometric, material, or dimensional-gradient characteristic of the one or more first tube assemblies and analyzing the one or more second images to identify at least one color, geometric, material, or dimensional-gradient characteristic of the one or more second tube assemblies.

20 . A diagnostic apparatus, comprising:
a location configured to store one or more first tube assemblies of a first type and one or more second tube assemblies of a second type; an imaging device configured to image at least a portion of the one or more first tube assemblies and at least a portion of the one or more second tube assemblies; a transport device configured to transport the one or more first tube assemblies and the one or more second tube assemblies at least to the imaging device; and a controller including a processor coupled to a memory, the memory having instructions stored therein that, when executed by the processor:
train a model to identify tube assemblies of the first type and tube assemblies of the second type based on one or more physical characteristics of a tube assembly identified in at least one image of the at least a portion of the one or more first tube assemblies and at least one i

Assignees

• Siemens Healthcare Diagnostics Inc

Inventors

• Nalam Venkat Rayal Raj Prasad
• Pollack Benjamin S
• Chang Yao-Jen
• Narasimhamurthy Venkatesh
• Singh Vivek
• Kapoor Ankur

Classifications

• A61B5/150343

  Collection vessels for collecting blood samples from the skin surface, e.g. test tubes, cuvettes (sampling from within body cavities A61B5/150015, A61B5/150045) · CPC title

• G06V10/774

  Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V20/60

  Type of objects · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V10/56

  relating to colour · CPC title