US 12,468,928 B2
Real time medical image processing using deep learning accelerator with integrated random access memory
Poorna Kale, Folsom, CA (US)
Assigned to Micron Technology, Inc., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on Aug. 6, 2020, as Appl. No. 16/987,112.
Prior Publication US 2022/0044108 A1, Feb. 10, 2022
Int. Cl. G06N 3/08 (2023.01); A61B 5/055 (2006.01); A61B 6/00 (2006.01); A61B 6/03 (2006.01); A61B 8/00 (2006.01); G06N 3/04 (2023.01); G06T 7/00 (2017.01); G11C 11/4096 (2006.01)
CPC G06N 3/08 (2013.01) [A61B 5/055 (2013.01); A61B 6/032 (2013.01); A61B 6/54 (2013.01); A61B 8/54 (2013.01); G06N 3/04 (2013.01); G06T 7/0012 (2013.01); A61B 2560/0475 (2013.01); G06T 2207/10081 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/10116 (2013.01); G06T 2207/10132 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30196 (2013.01); G11C 11/4096 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A device, comprising:
an integrated circuit, comprising:
random access memory configured to:
store first data representative of parameters of an artificial neural network, second data representative of instructions executable to implement matrix computations of the artificial neural network using at least the first data stored in the random access memory, and third data representative of an image generated in an imaging apparatus, the imaging apparatus configured to image a portion of a person; and
store, in a predefined location of the random access memory, an indication for triggering autonomous execution of the instructions;
a signal emitter configured to propagate at least one signal into the portion of the person;
a response sensor configured to detect at least one response signal generated based on the at least one signal interacting with the portion of the person and write the third data representative of the image based on the at least one response signal generated based on the at least one signal interacting with the portion of the person;
at least one interface configured to receive the third data representative of the image and store the third data into the random access memory; and
at least one processing unit configured to execute the instructions having matrix operands to implement the matrix computations of the artificial neural network, the artificial neural network configured to:
generate, using the image as an input and in response to the autonomous execution of the instructions triggered in response to writing of the input into the random access memory, an output in the random access memory, the output including at least one digital annotation configured to guide or assist an imaging device in performing imaging of the portion of the person;
overlay the at least one digital annotation onto the image to generate an overlaid image; and
generate, based on the overlaid image, a prompt to guide or assist the imaging device in performing the imaging.