Deep learning model for auto-focusing microscope systems
A computing system receives, from an image sensor, at least two images of a specimen positioned on a specimen stage of a microscope system. The computing system provides the at least two images to an autofocus model for detecting at least one distances to a focal plane of the specimen. The computing system identifies, via the autofocus model, the at least one distance to the focal plane of the specimen. Based on the identifying, the computing system automatically adjusts a position of the specimen stage with respect to an objective lens of the microscope system.
1. A microscope system comprising:
an optical system for imaging a specimen, the optical system comprising an objective lens and a specimen stage configured to support the specimen during imaging; and
a focus detection system for automatically focusing the optical system, the focus detection system comprising a computing system that includes a processor and a memory, the memory has programming code stored thereon, which, when executed by the processor, causes the computing system to perform operations, comprising:
receiving, by the computing system from the optical system, only two images of the specimen positioned on the specimen stage, wherein the two images comprise a first image taken above or below a focal plane of the specimen and a second image taken above or below the focal plane;
providing, by the computing system, the two images, as input, to a convolutional neural network to detect at least one distance to the focal plane of the specimen;
identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen by analyzing both the first image and the second image; and
based on the identifying, automatically adjusting, by the computing system, a position of the specimen stage with respect to the objective lens of the microscope system.
2. The microscope system of claim 1 , wherein receiving, by the computing system from the optical system, the two images of the specimen positioned on the specimen stage of the microscope system comprises:
receiving the first image at a first z-position; and
receiving the second image of the two images at a second z-position.
3. The microscope system of claim 2 , wherein an x-position and a y-position of the specimen stage remains constant while the first z-position differs from the second z-position.
4. The microscope system of claim 1 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
generating a range of distances to the focal plane, wherein the range of distances comprises an actual distance to the focal plane.
5. The microscope system of claim 1 , wherein the focal plane is one of multiple focal planes.
6. The microscope system of claim 5 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
identifying at least one first distance to a first focal plane of the multiple focal planes of the specimen; and
identifying at least one second distance to a second focal plane of the multiple focal planes of the specimen.
7. The microscope system of claim 1 , wherein automatically adjusting, by the computing system, the position of the specimen stage with respect to the objective lens of the microscope system comprises:
causing a drive mechanism of the microscope system to adjust the distance between the specimen stage and the objective lens.
8. A method comprising:
receiving, by a computing system from an image sensor, only two images of a specimen positioned on a specimen stage of a microscope system, wherein the two images comprise a first image taken above or below a focal plane of the specimen and a second image taken above or below the focal plane;
providing, by the computing system, the two images, as input, to a convolutional neural network to detect at least one distance to the focal plane of the specimen;
identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen by analyzing both the first image and the second image; and
based on the identifying, automatically adjusting, by the computing system, a position of the specimen stage with respect to an objective lens of the microscope system.
9. The method of claim 8 , wherein receiving, by the computing system from the image sensor, the two images of the specimen positioned on the specimen stage of the microscope system comprises:
receiving the first image of the two images at a first z-position; and
receiving the second image of the two images at a second z-position.
10. The method of claim 9 , wherein an x-position and a y-position of the specimen stage remains constant while the first z-position differs from the second z-position.
11. The method of claim 8 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
generating a range of distances to the focal plane, wherein the range of distances comprises an actual distance to the focal plane.
12. The method of claim 8 , wherein the focal plane is one of multiple focal planes.
13. The method of claim 12 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
identifying at least one first distance to a first focal plane of the multiple focal planes of the specimen; and
identifying at least one second distance to a second focal plane of the multiple focal planes of the specimen.
14. The method of claim 8 , wherein automatically adjusting, by the computing system, the position of the specimen stage with respect to the objective lens of the microscope system comprises:
causing a drive mechanism of the microscope system to adjust the distance between the specimen stage and the objective lens.
15. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by a processor, causes a computing system to perform operations comprising:
receiving, by the computing system from an image sensor, only two images of a specimen positioned on a specimen stage of a microscope system, wherein the two images comprise a first image taken above or below a focal plane of the specimen and a second image taken above or below the focal plane;
providing, by the computing system, the two images, as input, to a convolutional neural network for detecting at least one distance to the focal plane of the specimen;
identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen by analyzing both the first image and the second image; and
based on the identifying, automatically adjusting, by the computing system, a position of the specimen stage with respect to an objective lens of the microscope system.
16. The non-transitory computer readable medium of claim 15 , wherein receiving, by the computing system from the image sensor, the two images of the specimen positioned on the specimen stage of the microscope system comprises:
receiving the first image of the two images at a first z-position; and
receiving the second image of the two images at a second z-position.
17. The non-transitory computer readable medium of claim 15 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
generating a range of distances to the focal plane, wherein the range of distances comprises an actual distance to the focal plane.
18. The non-transitory computer readable medium of claim 15 , wherein the focal plane is one of multiple focal planes.
19. The non-transitory computer readable medium of claim 18 , wherein identifying, via the convolutional neural network, the at least one distance to the focal plane of the specimen comprises:
identifying at least one first distance to a first focal plane of the multiple focal planes of the specimen; and
identifying at least one second distance to second focal plane of the multiple focal planes of the specimen.
20. The non-transitory computer readable medium of claim 15 , wherein automatically adjusting, by the computing system, the position of the specimen stage with respect to the objective lens of the microscope system comprises:
causing a drive mechanism of the microscope system to adjust the distance between the specimen stage and the objective lens.