US 11,893,811 B2
Method for object detection using hierarchical deep learning
Daniel Clymer, Pittsburgh, PA (US); Jonathan Cagan, Pittsburgh, PA (US); Philip LeDuc, Pittsburgh, PA (US); Liron Pantanowitz, Pittsburgh, PA (US); and Janet Catov, Pittsburgh, PA (US)
Assigned to Carnegie Mellon University, Pittsburgh, PA (US)
Filed by Carnegie Mellon University, Pittsburgh, PA (US); and UNIVERSITY OF PITTSBURGH—OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION, Pittsburgh, PA (US)
Filed on May 21, 2021, as Appl. No. 17/326,541.
Application 17/326,541 is a continuation of application No. 17/073,041, filed on Oct. 16, 2020, granted, now 11,367,189.
Claims priority of provisional application 62/973,697, filed on Oct. 18, 2019.
Prior Publication US 2021/0327061 A1, Oct. 21, 2021
Int. Cl. G06V 20/69 (2022.01); G06T 7/00 (2017.01); G06F 18/10 (2023.01); G06F 18/241 (2023.01); G06F 18/2135 (2023.01); G06F 18/214 (2023.01); G06F 18/2415 (2023.01); G06V 40/14 (2022.01)
CPC G06V 20/69 (2022.01) [G06F 18/10 (2023.01); G06F 18/2135 (2023.01); G06F 18/2148 (2023.01); G06F 18/241 (2023.01); G06F 18/2415 (2023.01); G06T 7/0012 (2013.01); G06V 20/695 (2022.01); G06V 20/698 (2022.01); G06T 2207/10056 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30101 (2013.01); G06V 40/14 (2022.01); G06V 2201/031 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining a high-resolution whole slide image of a post-birth placenta;
analyzing the whole slide image, using a trained machine learning model, to identify one or more blood vessels in the placenta, the analysis occurring at a lower resolution than the native resolution of the whole slide image;
classifying the identified blood vessels at a higher resolution using a trained machine learning classifier that outputs a latent vector for each classified blood vessel:
aggregating the latent vectors for a predetermined number of classified blood vessels;
pooling the aggregated latent vectors by calculating a maximum or minimum of the data for each node of a feature map of the machine learning classifier;
reducing the dimension on the pooled aggregated latent vectors to produce a reduced dimension latent vector; and
performing a binary classification for each whole slide image based on the latent vector.