Multi-modal computer-aided diagnosis systems and methods for prostate cancer

A non-transitory machine-readable storage medium comprising instructions that, in response to execution by a processor, cause the processor to:

• detect, using a neural network, a lesion from an image of a prostate gland based at least in part on a prostate mask comprising one or more zones of the prostate gland;
• generate a mapping of the lesion from the image to a sector map without user input, wherein the generating the mapping of the lesion comprises identifying a depth region of the lesion, wherein the depth region indicates a location of the lesion along a depth axis, wherein the processor is to generate the mapping of the lesion by using a digital twin;
• provide the sector map comprising a representation of the lesion within the prostate gland mapped from the image to the sector map;
• generate a score using the neural network and the provided sector map comprising the representation of the lesion mapped from the image to the sector map;
• and
• display the sector map with the representation of the of the lesion within the prostate gland,
• wherein the depth region is identified using an apex region, a mid region, and a base region of the prostate gland,
• wherein the generating the mapping comprises calculating one or more polar coordinates relative to a center of the lesion in the image,
• wherein the generating the mapping comprises computing a normalized radius based on the center of the lesion in the image, and
• wherein the generating the mapping comprises computing a de-normalized radius based on the normalized radius and one or more dimensions of the sector map.

A method for computer-aided prostate condition diagnosis, the method comprising:

• detecting, using a neural network, a lesion from an image of a prostate gland based at least in part on a prostate mask comprising one or more zones of the prostate gland;
• generating a mapping of the lesion from the image to a sector map without user input, the generating the mapping of the lesion comprising identifying a depth region of the lesion, wherein the depth region indicates a location of the lesion along a depth axis, wherein the generating of the mapping of the lesion is performed using a digital twin;
• providing the sector map comprising a representation of the lesion within the prostate gland mapped from the image to the sector map;
• generating a score using the neural network and the provided sector map comprising the representation of the lesion mapped from the image to the sector map;
• and
• displaying the sector map with the representation of the of the lesion within the prostate gland,
• wherein the depth region is identified using an apex region, a mid region, and a base region of the prostate gland,
• wherein the generating the mapping comprises calculating one or more polar coordinates relative to a center of the lesion in the image,
• wherein the generating the mapping comprises computing a normalized radius based on the center of the lesion in the image, and
• wherein the generating the mapping comprises computing a de-normalized radius based on the normalized radius and one or more dimensions of the sector map.