US 12,229,670 B2
Temporalizing or spatializing networks
Chandan Aladahalli, Bengaluru (IN); Krishna Seetharam Shriram, Bangalore (IN); and Vikram Melapudi, Bangalore (IN)
Assigned to GE PRECISION HEALTHCARE LLC, Waukesha, WI (US)
Filed by GE Precision Healthcare LLC, Milwaukee, WI (US)
Filed on Jun. 25, 2021, as Appl. No. 17/358,694.
Prior Publication US 2022/0414449 A1, Dec. 29, 2022
Int. Cl. G06N 3/08 (2023.01); G06F 18/25 (2023.01)
CPC G06N 3/08 (2013.01) [G06F 18/25 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a processor that executes computer-executable components stored in a computer-readable memory, the computer-executable components comprising:
a network augmentation component that:
takes a pretrained artificial neural network that is configured to perform an inferencing task on a two-dimensional image;
duplicates, without retraining, an encoder unit of the pretrained artificial neural network one or more times, thereby yielding a plurality of duplicates of the encoder unit of the pretrained artificial neural network, wherein the plurality of duplicates of the encoder unit are arranged in parallel with each other;
combines output from the plurality of duplicates of the encoder unit; and
provides combined output from the plurality of duplicates of the encoder unit to a decoder unit of the pretrained artificial neural network; and
a network application component that employs the pretrained artificial neural network to perform the inferencing task on a stack of two-dimensional images, each of which is input to a respective one of the plurality of duplicates of the encoder unit.