CNN Label Occlusion Compensation in Specimen Characterization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing automated systems for characterizing specimen containers face challenges in accurately determining the presence of hemolysis, icterus, and lipemia in serum or plasma portions due to label occlusion, leading to subjective and labor-intensive visual inspections and reduced confidence in test results.

Innovation Solution

A method and apparatus using a quality check module with a convolutional neural network (CNN) to classify label configurations and adjust spectral responses, enabling effective characterization of serum or plasma portions even when occluded by labels, by identifying label counts and configurations and compensating for label interference.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated systems use visual inspection to characterize serum or plasma portions, then productivity is improved, but measurement precision deteriorates due to label occlusion causing subjective and labor-intensive inspections

Engineering Contradiction:
Improveautomation of specimen characterizationVSAvoidaccuracy of interferent detection
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent introduces a quality check module with CNN-based image analysis as an intermediary between the specimen container and the characterization system. This intermediary captures images of the specimen container, uses the CNN to identify label configurations and determine label counts, then adjusts spectral responses to compensate for label interference, enabling accurate automated characterization despite label occlusion

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically adjusts spectral response parameters based on detected label configurations. By changing the analytical parameters (spectral responses) according to the number and position of labels, the system compensates for label interference and maintains measurement precision across different labeling scenarios

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple labels are placed on specimen containers for tracking and identification, then reliability of specimen identification is improved, but difficulty of detecting and measuring serum or plasma characteristics worsens due to label occlusion

Engineering Contradiction:
Improvespecimen identification accuracyVSAvoidvisibility of serum or plasma portion
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system performs preliminary characterization of the label configuration before conducting the actual serum or plasma analysis. By first identifying the number and position of labels using the quality check module, the system can pre-adjust its detection parameters and spectral responses to compensate for the specific occlusion pattern, enabling accurate measurement despite the labels

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transitions from direct visual inspection of the serum or plasma portion to a multi-dimensional approach: capturing images from multiple angles, analyzing label configurations in the spatial domain, and adjusting spectral responses in the frequency domain. This dimensional transformation allows the system to overcome occlusion in the direct view dimension

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3610269B1Methods and apparatus for determining label count during specimen characterization
Publication Date: 2024.11.20 SIEMENS HEALTHCARE DIAGNOSTICS INC
  • EP3610269B1 patent drawingFigure 1~2A
  • EP3610269B1 patent drawingFigure 2B~3A
  • EP3610269B1 patent drawingFigure 3B~3C

AI summary

A method of characterizing a serum and plasma portion of a specimen in regions occluded by one or more labels. The characterization may be used for Hemolysis, Icterus, and/or Lipemia, or Normal detection. The method captures one or more images of a labeled specimen container including a serum or plasma portion, processes the one or more images to provide segmentation data and identification of a label-containing region, and classifying the label-containing region with a convolutional neural network (CNN) to provide a pixel-by-pixel (or patch-by-patch) characterization of the label thickness count, which may be used to adjust intensities of regions of a serum or plasma portion having label occlusion. Optionally, the CNN can characterize the label-containing region as one of multiple pre-defined label configurations. Quality check modules and specimen testing apparatus adapted to carry out the method are described, as are other aspects.