Optimization of a deep learning model for performing a medical imaging analysis task

An apparatus comprising:

• means for receiving a plurality of input medical images each depicting a lesion;
• means for localizing the lesion based on the plurality of input medical images using a trained localization network to generate one or more localization maps, the trained localization network receiving as input the plurality of input medical images and generating as output the one or more localization maps;
• means for classifying the lesion based on the one or more localization maps to generate a classification result for the lesion;
• and
• means for outputting the classification result.

A non-transitory computer readable medium storing computer program instructions, the computer program instructions when executed by a processor cause the processor to perform operations comprising:

• receiving a plurality of input medical images each depicting a lesion;
• localizing the lesion based on the plurality of input medical images using a trained localization network to generate one or more localization maps, the trained localization network receiving as input the plurality of input medical images and generating as output the one or more localization maps;
• classifying the lesion based on the one or more localization maps to generate a classification result for the lesion;
• and
• outputting the classification result.