Image analyzing device
An image analyzing device ( 1 ) includes an image holder ( 8 ) that holds an image, a trained model registration part ( 10 ) configured to register trained models created by machine learning, a trained model holder ( 12 ) that holds the trained models registered by the trained model registration part ( 10 ), an algorithm holder ( 14 ) that holds a plurality of analysis algorithms for executing analysis processing of an image, a recipe creation part ( 18 ) configured, for an image to be analyzed optionally selected from among images held in the image holder ( 8 ), to create an analysis recipe for analyzing the image to be analyzed by combining a trained model selected from the trained models held in the trained model holder ( 10 ) and an analysis algorithm optionally selected from the plurality of analysis algorithms held in the algorithm holder ( 14 ), and an analysis execution part ( 20 ) configured to execute analysis of the image to be analyzed based on the analysis recipe created by the recipe creation part ( 18 ).
1. An image analyzing device comprising:
an image holder that holds an image;
a trained model registration part configured to register trained models created by machine learning;
a trained model holder that holds the trained models registered by the trained model registration part;
an algorithm holder that holds a plurality of analysis algorithms for executing analysis processing of an image;
a recipe creation part configured, for an image to be analyzed optionally selected from among images held in the image holder, to create an analysis recipe for analyzing the image to be analyzed by combining a trained model selected from the trained models held in the trained model holder and an analysis algorithm optionally selected from the plurality of analysis algorithms held in the algorithm holder; and
an analysis execution part configured to execute analysis of the image to be analyzed based on the analysis recipe created by the recipe creation part.
2. The image analyzing device according to claim 1 , wherein
the algorithm holder holds an analysis algorithm for creating a trained model by performing machine learning,
the recipe creation part is configured to create an analysis recipe including machine learning for creating a trained model of the image to be analyzed in a case where the recipe creation part determines, based on information input by a user, that the trained model needs to be created, and is configured, when creating the analysis recipe including machine learning, to select an image for training to be used for the machine learning from among images held in the image holder and to create a data set for training including the image for training and the image to be analyzed, and
the trained model registration part is configured to register a trained model created by the machine learning using the data set for training.
3. The image analyzing device according to claim 2 , wherein
the image is an image obtained by imaging each of a plurality of cell culture wells provided in a cell culture plate, and each image is held in the image holder in a state of being associated with a culture condition of a cell in a cell culture well being imaged, and
in a case where there are a plurality of the images to be analyzed and a plurality of the images for training when creating the analysis recipe including the machine learning, the recipe creation part is configured to create a plurality of the data sets for training by combining the image to be analyzed and the image for training corresponding to each other based on the culture condition associated with each image.
4. The image analyzing device according to claim 3 , wherein the recipe creation part is configured, when creating a plurality of the data sets for training, to present a list of the image to be analyzed and the image for training constituting each data set for training to a user together with a culture condition associated with each data set for training.
5. The image analyzing device according to claim 3 , wherein
the recipe creation part is configured, based on information input by a user, to set a part of a plurality of the data sets for training as a data set for evaluation for evaluating a created trained model,
the analysis execution part is configured to create a trained model by performing the machine learning using the data set for training and to execute evaluation using the data set for evaluation for the created trained model, and
the trained model registration part is configured to register the trained model having a highest evaluation result.
6. The image analyzing device according to claim 5 , wherein the recipe creation part is configured, when a user desires automatic selection of the data set for training to be the data set for evaluation, to classify a plurality of the data sets for training into a plurality of sections according to the culture condition and to set at least one of the data sets for training belonging to each section as the data set for evaluation.
7. The image analyzing device according to claim 1 , further comprising:
an algorithm registration part configured to register a new analysis algorithm not held in the algorithm holder, wherein
the algorithm holder is configured to hold an analysis algorithm registered by the algorithm registration part.
8. An image analyzing device comprising:
an image holder that holds a plurality of images obtained by imaging each of a plurality of cell culture wells provided in a cell culture plate in a state of being associated with a culture condition of a cell in the imaged cell culture well;
a recipe creation part configured to select a plurality of images for training corresponding to a plurality of images to be analyzed optionally selected from among images held in the image holder, to create a plurality of data sets for training by combining, based on the culture condition associated with each image, the image to be analyzed and the image for training corresponding to each other, and to create an analysis recipe for creating a trained model for image analysis by performing machine learning using the plurality of data sets for training; and
an analysis execution part configured to create the trained model based on an analysis recipe created by the recipe creation part.
9. The image analyzing device according to claim 8 , wherein the recipe creation part is configured, when the plurality of data sets for training have been created, to present a list of the image to be analyzed and the image for training constituting each of the plurality of data sets for training to a user together with a culture condition associated with each of the plurality of data sets for training.
10. The image analyzing device according to claim 8 , wherein
the recipe creation part is configured to set a part of the plurality of data sets for training as a data set for evaluation for evaluating a created trained model based on information input by a user, and
the analysis execution part is configured to create a trained model by performing the machine learning using the plurality of data sets for training and to execute evaluation using the data set for evaluation for the created trained model.
11. The image analyzing device according to claim 10 , wherein the recipe creation part is configured, when a user desires automatic selection of the data set for training to be the data set for evaluation, to classify the plurality of data sets for training into a plurality of sections according to the culture condition and to set at least one of the data sets for training belonging to each section as the data set for evaluation.