Methods and apparatus for detecting an entity in a bodily sample

• US 10,663,712 B2
• Filed: 10/18/2019
• Issued: 05/26/2020
• Est. Priority Date: 09/17/2015
• Status: Active Grant

• Current Assignee: S.D. Sight Diagnostics Ltd
• Original Assignee: S.D. Sight Diagnostics Ltd
• Inventors: Eshel, Yochay Shlomo, Lezmy, Natalie, Gluck, Dan, Houri Yafin, Arnon, Pollak, Joseph Joel
• Primary Examiner(s): Johns, Andrew W
• Application Number:
  US16/657,473
• Publication Number:
  US 20200049970A1
• Time in Patent Office: 221 Days
• CPC Class Codes: G01N 2021/8887 based on image processing techniques
• CPC Class Codes: G01N 21/8851 Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges G01N21/8806 and G01N21/93 - G01N21/95692 take precedence; optical measurement of dimensions G01B11/00; optical scanning G02B26/10; image transformation G06T3/00; computerised image enhancement G06T5/00; image processing per se for flaw detection G06T7/0002
• CPC Class Codes: G02B 21/365 Control or image processing arrangements for digital or video microscopes G02B21/361, G02B21/362 take precedence
• CPC Class Codes: G02B 21/367 providing an output produced by processing a plurality of individual source images, e.g. image tiling, montage, composite images, depth sectioning, image comparison
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 20/693 Acquisition
• CPC Class Codes: G06V 20/698 Matching; Classification
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

Apparatus and methods for analyzing a bodily sample are described. A microscope system acquires one or more microscope images of the bodily sample. A computer processor identifies, in the one or more images, at least one element as being a candidate of a given entity. The computer processor extracts, from the one or more images, at least one candidate-informative feature associated with the candidate of the given entity, and at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole. The computer processor processes the candidate-informative feature in combination with the sample-informative feature, and generates an output upon the output device, in response thereto. Other applications are also described.

Claims

1. Apparatus for analyzing a bodily sample, the apparatus comprising:

a microscope system configured to acquire one or more microscope images of the bodily sample;

an output device; and

at least one computer processor configured to;

identify, in the one or more images, at least one element as being a candidate of a given entity,

extract, from the one or more images, at least one candidate-informative feature associated with the candidate of the given entity,

extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole,

process the candidate-informative feature in combination with the sample-informative feature, and

generate an output upon the output device, in response thereto.

3. The apparatus according to claim 1, wherein the at least one computer processor is configured to extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, by extracting, from the one or more images, at least one sample-informative feature selected from the group consisting of:

sizes of non-candidate constituents in the bodily sample, shapes of non-candidate constituents within the bodily sample, intensities of non-candidate constituents within the bodily sample, a quantity of cells of a given cell type within the bodily sample, a distribution of cells of a given cell type within the bodily sample, and a distribution of candidates within the bodily sample.

4. The apparatus according to claim 1, wherein the at least one computer processor is configured to:

identify, in the one or more images, a plurality of elements as being candidates of the given entity, and

extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, by extracting, from the one or more images, an indication of a number of the candidates that have a probability of being the given entity that exceeds a threshold.

5. The apparatus according to claim 1, wherein the at least one computer processor is configured to:

identify, in the one or more images, a plurality of elements as being candidates of the given entity, and

extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, by extracting, from the one or more images, an indication of probabilities of the candidates being the given entity.

6. The apparatus according to claim 1, wherein:

the microscope system is configured to acquire the one or more microscope images of the bodily sample by acquiring one or more microscope images of a bodily sample that is stained with a stain; and

the at least one computer processor is configured to extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, by extracting, from the one or more images, at least one sample-informative feature that is indicative of a quality of staining of the bodily sample by the stain.

7. The apparatus according to claim 1, wherein the at least one computer processor is configured to extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole by extracting, from the one or more images, at least one sample-informative feature that is indicative of a foreign object that is present in the bodily sample.

8. The apparatus according to claim 1, wherein the bodily sample includes a sample that contains blood, and wherein the computer processor is configured to identify at least one element as being a candidate of a given entity by identifying at least one element as being a candidate of a given entity within the blood.

9. The apparatus according to claim 8, wherein the computer processor is configured to identify at least one element as being a candidate of a given entity by identifying at least one element as being a candidate of an entity selected from the group consisting of:

a platelet, a white blood cell, an anomalous white blood cell, a circulating tumor cell, a red blood cell, a reticulocyte, a Howell Jolly body, and a pathogen.

10. The apparatus according to claim 8, wherein the computer processor is configured to perform a blood count on the sample that contains blood, and the computer processor is configured to generate the output by outputting the blood count.

11. The apparatus according to claim 8, wherein the computer processor is configured to perform a complete blood count on the sample that contains blood, and the computer processor is configured to generate the output by outputting the complete blood count.

12. The apparatus according to claim 8, wherein the at least one computer processor is configured to extract, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, by extracting, from the one or more images, a sample informative-feature selected from the group consisting of:

sizes of red blood cells that are present within the bodily sample, an indication of a presence of Howell Jolly bodies within the bodily sample, a concentration of platelets within the bodily sample, a relationship between a number of reticulocytes and a number of mature red blood cells within the bodily sample, and a concentration of reticulocytes within the bodily sample.

13. The apparatus according to claim 1, wherein the at least one computer processor is configured to:

identify within one or more images a plurality of elements as being candidates of one or more given entities,

extract, from the one or more images, candidate-informative features associated with the candidates,

extract, from the candidate-informative feature, two or more sample-informative features that are indicative of characteristics of the bodily sample as a whole, and

determine a characteristic of the bodily sample, by processing the two or more sample-informative features.

14. The apparatus according to claim 13, wherein the computer processor is configured to extract the two or more sample-informative features that are indicative of characteristics of the bodily sample as a whole, by extracting, from the candidate-informative features, two or more sample-informative features selected from the group consisting of:

number of candidates in the sample, type of candidates in the sample, brightness of the candidates relative to background brightness, a probability of candidates being the given entity, number of candidates that have a probability of being a given one of the entities that exceeds a threshold, number of candidates that have a probability of being a given type of the one of the entities that exceeds a threshold, a number of platelets in the sample, brightness of platelets, a number of reticulocytes in the sample, a number of reticulocytes infected by pathogens in the sample, a proximity of the candidates to red blood cells, and a number of red blood cells in the sample.

2. The apparatus according to claim I, wherein the at least one computer processor is configured to extract, from the one or more images, at least one candidate-informative feature associated with the candidate by extracting, from the one or more images, at least one candidate-informative feature associated with the candidate, the candidate-informative feature being a feature selected from the group consisting of:

a size of the candidate, a shape of the candidate, a motion of the candidate, an intensity of the candidate, a location of the candidate within the bodily sample, a property of a cell overlapping the candidate, a color of the candidate, a texture of the candidate, and a sharpness of a boundary of the candidate.

15. A method for analyzing a bodily sample, the method comprising:

acquiring one or more microscope images of the bodily sample, using a microscope;

using at least one computer processor;

in the one or more images, identifying at least one element as being candidate of a given entity;

extracting, from the one or more images, at least one candidate-informative feature associated with the candidate of the given entity;

extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole;

processing the candidate-informative feature in combination with the sample-informative feature; and

generating an output, in response thereto.

16. The method according to claim 15, wherein extracting, from the one or more images, at least one candidate-informative feature associated with the candidate comprises extracting, from the one or more images, at least one candidate-informative feature associated with the candidate, the candidate-informative feature being a feature selected from the group consisting of:

a size of the candidate, a shape of the candidate, a motion of the candidate, an intensity of the candidate, a location of the candidate within the bodily sample, a property of a cell overlapping the candidate, a color of the candidate, a texture of the candidate, and a sharpness of a boundary of the candidate.

17. The method according to claim 15, wherein extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, comprises extracting, from the one or more images, at least one sample-informative feature selected from the group consisting of:

sizes of non-candidate constituents in the bodily sample, shapes of non-candidate constituents within the bodily sample, intensities of non-candidate constituents within the bodily sample, a quantity of cells of a given cell type within the bodily sample, a distribution of cells of a given cell type within the bodily sample, and a distribution of candidates within the bodily sample.

18. The method according to claim 15, wherein:

identifying at least one element as being a candidate of a given entity comprises identifying, in the one or more images, a plurality of elements as being candidates of the given entity, and

extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, comprises extracting, from the one or more images, an indication of a number of the candidates that have a probability of being the given entity that exceeds a threshold.

19. The method according to claim 15, wherein:

identifying at least one element as being a candidate of a given entity comprises identifying, in the one or more images, a plurality of elements as being candidates of the given entity, and

extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, comprises extracting, from the one or more images, an indication of probabilities of the candidates being the given entity.

20. The method according to claim 15, wherein:

acquiring one or more microscope images of the bodily sample comprises acquiring one or more microscope images of the bodily sample when the sample is stained with a stain; and

extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, comprises extracting, from the one or more images, at least one sample-informative feature that is indicative of a quality of staining of the bodily sample by the stain.

21. The method according to claim 15, wherein extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole comprises extracting, from the one or more images, at least one sample-informative feature that is indicative of a foreign object that is present in the bodily sample.

22. The method according to claim 15, wherein the bodily sample includes a sample that contains blood, and wherein identifying at least one element as being a candidate of a given entity comprises identifying at least one element as being a candidate of a given entity within the blood.

23. The method according to claim 22, wherein identifying at least one element as being a candidate of a given entity comprises identifying at least one element as being a candidate of an entity selected from the group consisting of:

a platelet, a white blood cell, an anomalous white blood cell, a circulating tumor cell, a red blood cell, a reticulocyte, a Howell Jolly body, and a pathogen.

24. The method according to claim 22, wherein generating the output comprises outputting a blood count.

26. The method according to claim 22, wherein extracting, from the one or more images, at least one sample-informative feature that is indicative of a characteristic of the bodily sample as a whole, comprises extracting, from the one or more images, a sample informative-feature selected from the group consisting of:

sizes of red blood cells that are present within the bodily sample, an indication of a presence of Howell Jolly bodies within the bodily sample, a concentration of platelets within the bodily sample, a relationship between a number of reticulocytes and a number of mature red blood cells within the bodily sample, and a concentration of reticulocytes within the bodily sample.

27. The method according to claim 15, wherein:

identifying at least one element as being a candidate of a given entity comprises identifying, within the one or more images, a plurality of elements as being candidates of one or more given entities,

extracting, from the one or more images, at least one candidate-informative feature associated with the candidate of the given entity comprises extracting, from the one or more images, candidate-informative features associated with the candidates,

extracting, from the one or more images, at least one sample-informative feature that is indicative of contextual information related to the bodily sample comprises extracting, from the candidate-informative feature, two or more sample-informative features that are indicative of characteristics of the bodily sample as a whole, and

the method comprises determining a characteristic of the bodily sample, by processing the two or more sample-informative features.

28. The method according to claim 27, wherein extracting the two or more sample-informative features that are indicative of characteristics of the bodily sample as a whole, comprises extracting, from the candidate-informative features, two or more sample-informative features selected from the group consisting of:

number of candidates in the sample, type of candidates in the sample, brightness of the candidates relative to background brightness, a probability of candidates being the given entity, number of candidates that have a probability of being a given one of the entities that exceeds a threshold, number of candidates that have a probability of being a given type of the one of the entities that exceeds a threshold, a number of platelets in the sample, brightness of platelets, a number of reticulocytes in the sample, a number of reticulocytes infected by pathogens in the sample, a proximity of the candidates to red blood cells, and a number of red blood cells in the sample.

25. The method according to claim wherein generating the output comprises outputting a complete blood count.