Methods and apparatus for bio-fluid specimen characterization using neural network having reduced training

US11386291B2 · kind B2 · utility

Assignee

• Siemens Healthcare Diagnostics Inc. · US

Inventors

• Stefan Kluckner · Berlin, DE
• Yao-Jen Chang
• Kai Ma · Plainsboro, US
• Vivek Singh · Monmouth Junction, US
• Terrence Chen · Princeton, US
• Benjamin S. Pollack · Budd Lake, US

Key dates

• Filing date: Jan 8, 2019
• Grant date: Jul 12, 2022
• Expiry date: Jan 8, 2039

Classification

• Technology area (CPC G): Physics
• CPC primary: G06V2201/06
• WIPO field: Measurement
• WIPO sector: Instruments

Abstract

A method of training a neural network (Convolutional Neural Network-CNN) including reduced graphical annotation input is provided. The training method can be used to train a Testing CNN that can be used for determining Hemolysis (H), Icterus (I), and/or Lipemia (L), or Normal (N) of a serum or plasma portion of a test specimen. The training method includes capturing training images of multiple specimen containers including training specimens, generating region proposals of the serum or plasma portions of the training specimens; and selecting the best matches for the location, size and shape of the region proposals for the multiple training specimens. The obtained features (network and weights) from the training CNN can be used in a testing CNN. Quality check modules and testing apparatus adapted to carry out the training method, and characterization methods using abounding box regressor are described, as are other aspects.