Maturity classification of stained reticulocytes using optical microscopy
AssigneeSiemens Healthineers AG
InventorsAnn-Kathrin Reichenwallner, Lukas Richter
Life of this patent
Estimating the term…
Abstract
The present invention relates to a method of maturity classifying reticulocytes from a whole blood sample, comprising: staining the sample with a supravital agglutinating dyeing reagent or a fluorescent agglutinating dye; illuminating the stained sample with a light beam to detect reticulocytes; determining for each reticulocyte the parameters of (i) a fraction (Λ) of the reticulum area (Ar) to the whole cell area (Ac); and (ii) a fraction (Γ) of the perimeter of the reticulum (Ur) to the reticulum area (Ar); and maturity classifying a reticulocyte into 1 of 4 major maturity classes according to the values determined for Λ and Γ.
Claims16
A method of maturity classifying reticulocytes from a whole blood sample, comprising:
- (a) staining the sample with a supravital agglutinating dyeing reagent or a fluorescent agglutinating dye;
- (b) illuminating the stained sample with a light beam in a light detection device to detect reticulocytes;
- (c) determining within an image, via one or more modules executing software on a processor to at least perform morphological segmentation operations, for each reticulocyte parameters of (i) a fraction (Λ) of a reticulum area (Ar) to a whole cell area (Ac) and (ii) a fraction (Γ) of a perimeter of the reticulum (Ur) to the reticulum area (Ar), wherein the image for each reticulocyte is an RGB (red-green-blue)-grayscale image and the reticulum area (Ar) and perimeter (Ur) are determined by the one or more modules using the green channel;
- (d) maturity classifying, via the one or more modules, each reticulocyte into 1 of 4 major maturity classes according to values determined for A and T;
- and
- (e) determining a subject's health status or reaction to a treatment based on one or more numbers of reticulocytes classified in step (d) compared with one or more reference numbers of classified reticulocytes or the subject's previous one or more numbers of classified reticulocytes.
A computer implemented method for maturity classifying reticulocytes from one or more images taken from a whole blood sample, comprising:
- determining within an image, via one or more modules executing software on a processor to at least perform morphological segmentation operations, for each reticulocyte parameters of (i) a fraction (Λ) of a reticulum area (Ar) to a whole cell area (Ac) and (ii) a fraction (Γ) of a perimeter of the reticulum (Ur) to the reticulum area (Ar), wherein the image for each reticulocyte is an RGB (red-green-blue)-grayscale image and the reticulum area (Ar) and perimeter (Ur) are determined by the one or more modules using the green channel;
- maturity classifying, via the one or more modules, the reticulocytes into 4 major classes according to values for Λ and Γ;
- and
- determining a subject's health status or reaction to a treatment based on one or more numbers of classified reticulocytes compared with one or more reference numbers of classified reticulocytes or the subject's previous one or more numbers of classified reticulocytes.
A computer program comprising instructions stored in a non-transitory computer readable medium which, when executed by a computer, cause the computer to:
- stain the sample with a supravital agglutinating dyeing reagent or a fluorescent agglutinating dye;
- illuminate the stained sample with a light beam in a light detection device to detect reticulocytes;
- determine within an image for each reticulocyte parameters of (i) a fraction (Λ) of a reticulum area (Ar) to a whole cell area (Ac), and (ii) a fraction (Γ) of a perimeter of the reticulum (Ur) to the reticulum area (Ar);
- maturity classify a reticulocyte into 1 of 4 major maturity classes according to the values determined for Λ and Γ;
- and
- morphologically compare each stained reticulocyte with an image repository of reticulocytes to classify each stained reticulocyte into the 1 of 4 major maturity classes and tag for further analysis those stained reticulocytes having a different classification than a classification according to the values determined for Λ and θ.
Dependencies are inferred from the claim text.