US 12,340,907 B2
Systems, methods, and apparatuses for implementing advancements towards annotation efficient deep learning in computer-aided diagnosis
Zongwei Zhou, Tempe, AZ (US); and Jianming Liang, Scottsdale, AZ (US)
Assigned to Arizona Board of Regents on behalf of Arizona State University, Scottsdale, AZ (US)
Filed by Arizona Board of Regents on behalf of Arizona State University, Scottsdale, AZ (US)
Filed on Apr. 8, 2022, as Appl. No. 17/716,929.
Claims priority of provisional application 63/173,250, filed on Apr. 9, 2021.
Claims priority of provisional application 63/188,981, filed on May 14, 2021.
Prior Publication US 2022/0328189 A1, Oct. 13, 2022
Int. Cl. G06F 18/21 (2023.01); G06V 10/26 (2022.01); G06V 10/44 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G16H 30/40 (2018.01); G16H 40/67 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01)
CPC G16H 50/20 (2018.01) [G06V 10/7753 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a memory to store instructions;
a processor to execute the instructions stored in the memory;
wherein the system is specially configured to learn annotation-efficient deep learning from non-labeled medical images to generate a trained deep-learning model by applying a multi-phase model training process, comprising:
pre-training a model by executing a one-time learning procedure using an initial annotated image dataset;
iteratively re-training the model by executing a fine-tuning learning procedure using newly available annotated images without re-using any images from the initial annotated image dataset;
selecting a plurality of most representative samples related to images of the initial annotated image dataset and the newly available annotated images by executing an active selection procedure based on the which of a collection of un-annotated images exhibit either a greatest uncertainty or a greatest entropy;
extracting generic image features from the initial annotated image dataset, the newly available annotated images, and the plurality of most representative samples selected;
updating the model using the generic image features extracted; and
outputting the model as the trained deep-learning model for use in analyzing a patient medical image which is not included in any training image for the trained deep-learning model.