Automatic calibration using machine learning
There is provided a cell analysis apparatus that comprises image capture circuitry for capturing a brightfield image of a cell using brightfield imaging. The cell has been dyed by a functional dye that indicates, during fluorescence imaging and during brightfield imaging, whether the cell has a given characteristic. A model derived by machine learning is stored and used in combination with the brightfield image to determine whether the cell has the given characteristic. There is also provided a method for creating a cell categorisation model, comprising applying a functional dye to one or more samples comprising a plurality of cells. The functional dye indicates during fluorescence imaging and during brightfield imaging whether each of the cells has a given characteristic. A brightfield image and a corresponding fluorescence image for each of the plurality of cells to which the dye has been applied are captured and a machine learning process is used to generate a model that predicts whether a cell has the given characteristic from a brightfield image. The model is generated by using the brightfield image and the corresponding fluorescence image of each of the plurality of cells as training data.
1. A cell analysis apparatus, comprising:
image capture circuitry configured to capture a brightfield image of a cell using brightfield imaging, wherein the cell has been dyed by a functional dye that indicates, during fluorescence imaging and during brightfield imaging, whether the cell has a given characteristic;
storage circuitry configured to store a model derived by machine learning; and
processing circuitry configured to use the model in combination with the brightfield image to determine whether the cell has the given characteristic,
wherein the model has been trained with a plurality of fluorescence images and a corresponding plurality of brightfield images, and
wherein the model is further trained using the plurality of fluorescent images to determine whether the given characteristic is present or not in the corresponding plurality of brightfield images.
2. The cell analysis apparatus of claim 1 , wherein the brightfield image is a colour image.
3. The cell analysis apparatus of claim 1 , wherein the brightfield image is a greyscale image.
4. The cell analysis apparatus of claim 1 , wherein
the given characteristic of the cell is that the cell is dead.
5. The cell analysis apparatus of claim 1 , wherein
the functional dye is an azo dye.
6. The cell analysis apparatus of claim 1 , wherein
the functional dye is Trypan blue.
7. The cell analysis apparatus of claim 1 , wherein
the model comprises a set of weights or parameters derived by using a neural network.
8. The cell analysis apparatus of claim 7 , wherein
the neural network is a convolutional neural network.
9. A method for using a cell analysis model, comprising:
applying a functional dye to a cell to produce a dyed cell, wherein the functional dye is configured to indicate, during fluorescence imaging and during brightfield imaging, whether the cell has a given characteristic;
capturing a brightfield image of the dyed cell using brightfield imaging; and
using a model derived by machine learning to determine whether the cell has the given characteristic from the brightfield image,
wherein the model has been trained with a plurality of fluorescence images and a corresponding plurality of brightfield images, and
wherein the model is further trained using the plurality of fluorescent images to determine whether the given characteristic is present or not in the corresponding plurality of brightfield images.
10. A method for creating a cell categorisation model, comprising:
applying a functional dye to one or more samples comprising a plurality of cells, wherein the functional dye is configured to indicate, during fluorescence imaging and during brightfield imaging, whether each of the cells has a given characteristic;
capturing a brightfield image and a corresponding fluorescence image for each of the plurality of cells to which the functional dye has been applied; and
using a machine learning process to generate a model that predicts whether a cell has the given characteristic from a brightfield image,
wherein the model is generated by using the brightfield image and the corresponding fluorescence image of each of the plurality of cells as training data, and
wherein the model is trained using the corresponding fluorescent image to determine whether the given characteristic is present or not in the brightfield image.
11. The method of claim 10 , wherein
the machine learning process comprises the use of a neural network to generate the model.
12. The method of claim 11 , wherein
the neural network is a convolutional neural network.
13. A non-transitory storage medium comprising the cell categorisation model produced according to the method of claim 10 .