Granted Patent B2

US 12,230,023 · App. 18/511,920 · Granted Feb 18, 2025

Inventors: Sadao Ota (Tokyo, JP); Issei Sato (Tokyo, JP); Katsuhito Fujiu (Tokyo, JP); Satoko Yamaguchi (Tokyo, JP); Kayo Waki (Tokyo, JP); Yoko Itahashi (Tokyo, JP); Ryoichi Horisaki (Osaka, JP)

Assignees: The University of Tokyo; Osaka University

G06V10/82C12M1/34G01N15/14G01N15/1404G01N15/1429G01N15/1434G01N15/1459G01N15/147G01N21/01G01N21/27G01N21/64G01N21/65G06F18/2178G06F18/28G06V10/772G06V10/7784G06V20/698G01N2015/1006G01N2015/1413G01N2015/1415G01N2015/145G06F2218/12

Abstract

Claims (24)

1. A method, comprising:

(a) providing one or more flow cytometers comprising a flow path and a light receiving unit, wherein the light receiving unit comprises one or more sensors;

(b) flowing one or more objects through the flow path of the one or more flow cytometers;

(c) irradiating the one or more objects with a structured illumination pattern as the one or more objects move relative to the structured illumination pattern in a light irradiation region, wherein the light irradiation region is located on at least a portion of the flow path in which the one or more objects flow through;

(d) detecting at least one electromagnetic wave from the one or more objects using the one or more sensors, wherein the one or more objects comprises at least one target object;

(e) converting the least one electromagnetic wave into one or more time-series electrical signals using the light receiving unit; and

(f) analyzing the one or more time-series electrical signals using a machine learning classification model to obtain one or more classification results.

2. The method of claim 1 , wherein the structured illumination pattern is generated by providing light from a light source through one or more optical elements, wherein the one or more flow cytometers further comprises the light source and the one or more optical elements.

3. The method of claim 2 , wherein the one or more optical elements comprises a plurality of regions with different optical characteristics.

4. The method of claim 1 , wherein one or more light properties in the light irradiation region is controlled or adjusted at least in part based on the one or more classification results.

5. The method of claim 1 , wherein the one or more flow cytometers is coupled to one or more logic circuits, wherein the one or more logic circuits comprise the machine learning classification model.

6. The method of claim 1 , further comprising sorting the one or more objects based on the one or more classification results.

7. The method of claim 6 , wherein the at least one target object is separated from the one or more objects.

8. The method of claim 1 , wherein the machine learning classification model is trained using one or more time-series electrical training signals from a plurality of training objects, wherein the one or more time-series electrical training signals is stored in a storage unit.

9. The method of claim 1 , wherein the machine learning classification model is trained using one or more time-series electrical training signals from a plurality of training objects using one or more processors.

10. The method of claim 1 , further comprising updating the machine learning classification model based on the one or more classification results.

11. The method of claim 1 , wherein the machine learning classification model is trained by supervised machine learning in combination with unsupervised machine learning.

12. The method of claim 1 , wherein the machine learning classification model is trained through machine learning using one or more time-series electrical training signals obtained from one or more labeled training objects.

13. The method of claim 12 , wherein the one or more labeled training objects are labeled with a fluorescent label.

14. The method of claim 12 , wherein the one or more labeled training objects comprises a classification label indicating whether the one or more labeled training objects is the at least one target object or is not the at least one target object.

15. The method of claim 1 , wherein the one or more time-series electrical signals comprise compressed spatial information corresponding to the one or more objects.

16. The method of claim 1 , further comprising, in (d), passing the at least one electromagnetic wave from the one or more objects through one or more optical elements prior to detecting the at least one electromagnetic wave using the one or more sensors.

17. The method of claim 1 , further comprising capturing an image of the one or more objects using an imaging sensor, wherein the light-receiving unit further comprises the imaging sensor.

18. The method of claim 1 , further comprising, subsequent to (e), storing the one or more time-series electrical signals using a storage unit; and wherein in (f), the analyzing the one or more time-series electrical signals comprises using the one or more time-series electrical signals stored in the storage unit to obtain the one or more classification results.

Assignments (1)

Priority Claims (1)

Continuity (5)

References Cited (400)