Capture and storage of magnified images

A method performed by one or more data processing apparatus which comprises, for each magnification level in a sequence of increasing magnification levels:

• imaging one or more regions of interest of a tissue sample at the current magnification level, comprising, for each region of interest:
• using a microscope to generate a magnified image of the region of interest of the tissue sample at the current magnification level;
• using a camera to capture and store the magnified image of the region of interest of the tissue sample at the current magnification level;
• and
• generating data defining one or more refined regions of interest based on the magnified image of the region of interest at the current magnification level, comprising:
• processing the magnified image of the region of interest of the tissue sample at the current magnification level using a detection machine learning model associated with the current magnification level to generate the data defining the one or more refined regions of interest,
• wherein the detection machine learning model associated with the current magnification level is specialized to process magnified images at the current magnification level as a result of being trained on a set of training examples that includes magnified images at the associated magnification level;
• wherein each refined region of interest corresponds to a proper subset of the region of interest of the tissue sample, and
• wherein the refined regions of interest of the tissue sample provide the regions of interest to be imaged at a next magnification level from the sequence of increasing magnification levels.

One or more non-transitory computer-readable storage media encoded with instructions that when executed by one or more computers cause the one or more computers to perform operations comprising, for each magnification level in a sequence of increasing magnification levels:

• imaging one or more regions of interest of a tissue sample at the current magnification level, comprising, for each region of interest:
• using a microscope to generate a magnified image of the region of interest of the tissue sample at the current magnification level;
• using a camera to capture and store the magnified image of the region of interest of the tissue sample at the current magnification level;
• and
• generating data defining one or more refined regions of interest based on the magnified image of the region of interest at the current magnification level, comprising:
• processing the magnified image of the region of interest of the tissue sample at the current magnification level using a detection machine learning model associated with the current magnification level to generate the data defining the one or more refined regions of interest,
• wherein the detection machine learning model associated with the current magnification level is specialized to process magnified images at the current magnification level as a result of being trained on a set of training examples that includes magnified images at the associated magnification level;
• wherein each refined region of interest corresponds to a proper subset of the region of interest of the tissue sample, and
• wherein the refined regions of interest of the tissue sample provide the regions of interest to be imaged at a next magnification level from the sequence of increasing magnification levels.