US 12,387,839 B2
Method and system for labeling medical images
Virginia Boyce Hill, Chicago, IL (US); Michael Alexander Iorga, Chicago, IL (US); Aggelos K. Katsaggelos, Chicago, IL (US); and Todd B. Parrish, Chicago, IL (US)
Assigned to Northwestern University, Evanston, IL (US)
Filed by Northwestern University, Evanston, IL (US)
Filed on Sep. 13, 2022, as Appl. No. 17/943,390.
Claims priority of provisional application 63/243,373, filed on Sep. 13, 2021.
Prior Publication US 2023/0078011 A1, Mar. 16, 2023
Int. Cl. G16H 30/40 (2018.01); G06F 40/40 (2020.01); G06T 7/00 (2017.01); G16H 30/20 (2018.01)
CPC G16H 30/40 (2018.01) [G06F 40/40 (2020.01); G06T 7/0012 (2013.01); G16H 30/20 (2018.01); G06T 2207/10081 (2013.01); G06T 2207/20081 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for labeling massive quantities of image report data, the system comprising:
a memory configured to store image report data; and
a processor operatively coupled to the memory and configured to:
apply a natural language processing algorithm to identify one or more characteristics of the image report, wherein the one or more characteristics include n-grams identified in the image report data;
identify one or more phrases in the image report data, wherein the one or more phrases are identified based on a model of encodings of the n-grams identified in the image report data;
label the image report data based at least in part on the one or more identified phrases to form a labeled dataset; and
analyze a received image report using the labeled dataset to determine a priority of the received image report.