Systems and methods for generating encoded representations for multiple magnifications of image data

A computer-implemented method of generating an encoded representation for one or more query images, the method comprising operating at least one processor to, for each query image:

• obtain a plurality of sub-images for the query image, the plurality of sub-images comprising at least one sub-image corresponding to each magnification level of a plurality of magnification levels, each sub-image of the plurality of sub-images comprising same dimensions as other sub-images of the plurality of sub-images, the dimensions being characterized by pixels;
• apply at least one artificial neural network model to pool and convolute sub-images for each magnification level to extract and aggregate feature vectors representative of image characteristics of the query image at that magnification level;
• and
• generate the encoded representation for the query image based on the feature vectors extracted for the plurality of magnification levels.

A system for generating an encoded representation for one or more query images, the system comprising:

• a communication component to provide access to the one or more images via a network;
• and
• at least one processor in communication with the communication component, the processor being operable to, for each query image:
• obtain a plurality of sub-images for the query image, the plurality of sub-images comprising at least one sub-image corresponding to each magnification level of the plurality of magnification levels, each sub-image of the plurality of sub-images comprising same dimensions as other sub-images of the plurality of sub-images, the dimensions being characterized by pixels;
• for each magnification level of the plurality of magnification levels, apply at least one artificial neural network model to pool and convolute a group of sub-images for that magnification level to extract and aggregate feature vectors representative of image characteristics of the query image at that magnification level;
• and
• generate the encoded representation for the query image based on the feature vectors extracted for the plurality of magnification levels.