Uncertainty-refined image segmentation under domain shift
Abstract
A method for digital image segmentation is provided. The method comprises training a neural network for image segmentation with a labeled training dataset from a first domain, wherein a subset of nodes in the neural net are dropped out during training. The neural network receives image data from a second, different domain. A vector of N values that sum to 1 is calculated for each image element, wherein each value represents an image segmentation class. A label is assigned to each image element according to the class with the highest value in the vector. Multiple inferences are performed with active dropout layers for each image element, and an uncertainty value is generated for each image element. The label of any image element with an uncertainty value above a predefined threshold is replaced with a new label corresponding to the class with the next highest value.
- Inventors:
- Issue Date:
- Research Org.:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1924873
- Patent Number(s):
- 11379991
- Application Number:
- 16/887,311
- Assignee:
- National Technology & Engineering Solutions of Sandia, LLC (Albuquerque, NM)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
G - PHYSICS G06 - COMPUTING G06N - COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
- DOE Contract Number:
- NA0003525
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 05/29/2020
- Country of Publication:
- United States
- Language:
- English
Citation Formats
Martinez, Carianne, Potter, Kevin Matthew, Donahue, Emily, Smith, Matthew David, Snider, Charles J., Korbin, John P., Roberts, Scott Alan, and Collins, Lincoln. Uncertainty-refined image segmentation under domain shift. United States: N. p., 2022.
Web.
Martinez, Carianne, Potter, Kevin Matthew, Donahue, Emily, Smith, Matthew David, Snider, Charles J., Korbin, John P., Roberts, Scott Alan, & Collins, Lincoln. Uncertainty-refined image segmentation under domain shift. United States.
Martinez, Carianne, Potter, Kevin Matthew, Donahue, Emily, Smith, Matthew David, Snider, Charles J., Korbin, John P., Roberts, Scott Alan, and Collins, Lincoln. Tue .
"Uncertainty-refined image segmentation under domain shift". United States. https://www.osti.gov/servlets/purl/1924873.
@article{osti_1924873,
title = {Uncertainty-refined image segmentation under domain shift},
author = {Martinez, Carianne and Potter, Kevin Matthew and Donahue, Emily and Smith, Matthew David and Snider, Charles J. and Korbin, John P. and Roberts, Scott Alan and Collins, Lincoln},
abstractNote = {A method for digital image segmentation is provided. The method comprises training a neural network for image segmentation with a labeled training dataset from a first domain, wherein a subset of nodes in the neural net are dropped out during training. The neural network receives image data from a second, different domain. A vector of N values that sum to 1 is calculated for each image element, wherein each value represents an image segmentation class. A label is assigned to each image element according to the class with the highest value in the vector. Multiple inferences are performed with active dropout layers for each image element, and an uncertainty value is generated for each image element. The label of any image element with an uncertainty value above a predefined threshold is replaced with a new label corresponding to the class with the next highest value.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2022},
month = {7}
}
Works referenced in this record:
Transaction periodicity forecast using machine learning-trained classifier
patent-application, December 2021
- Chaturvedi, Anubhav
- US Patent Application 16/915790; 20210406896
A method to determine a degree of abnormality, a respective computer readable medium and a distributed cancer analysis system
patent-application, May 2021
- Lahrmann, Bernd
- US Patent Application 17/254970; 20210142908
Uncertainty guided semi-supervised neural network training for image classification
patent-application, July 2021
- Sedai, Suman; Antony, Bhavna Josephine; Garnavi, Rahil
- US Patent Application 16/738378; 20210216825
Learning apparatus and method for learning a model corresponding to a function changing in time series
patent-application, November 2018
- Kajino, Hiroshi
- US Patent Application 15/581814; 20180314978
Learning of Detection Model Using Loss Function
patent-application, October 2020
- Goto, Takuya; Nakano, Hiroki; Sakamoto, Masaharu
- US Patent Application 16/507625; 20200311483
Adapting a generative adversarial network to new data sources for image classification
patent, January 2020
- Madani, Ali; Moradi, Mehdi; Syeda-Mahmood, Tanveer F.
- US Patent Document 10,540,578
Training artificial neural networks with constraints
patent-application, September 2021
- Kennel, Matthew Bochner; Zoldi, Scott Michael
- US Patent Application 16/823193; 20210295175
Methods to estimate effectiveness of a medical treatment
patent-application, September 2020
- Tang, Qi; Ling, Wodan; Qi, Youran
- US Patent Application 16/790216; 20200275885
Unification of models having respective target classes with distillation
patent-application, February 2021
- Vongkulbhisal, Jayakorn; Vinayavekhin, Phongtharin
- US Patent Application 17/046014; 20210034985
Training Artificial Neural Networks Using Context-Dependent Gating With Weight Stabilization
patent-application, August 2020
- Masse, Nicolas Y.; Grant, Gregory D.; Freedman, David J.
- US Patent Application 16/774343; 20200250483
Robust anti-adversarial machine learning
patent-application, May 2020
- Baker, James K.
- US Patent Application 16/619278; 20200143240
User classification from data via deep segmentation for semi-supervised learning
patent-application, May 2021
- Burkhart, Michael; Shan, Kyle
- US Patent Application 16/681239; 20210142152
Annealed dropout training of neural networks
patent-application, October 2016
- Goel, Vaibhava; Rennie, Steven John; Thomas, Samuel
- US Patent Application 14/920304; 20160307098
Information processing apparatus
patent-application, January 2019
- Gotou, Takefumi
- US Patent Application 16/038486; 20190034803
Face detection
patent-application, March 2017
- Saberian, Mohammad; Farfade, Sachin Sudhakar; Li, Jia
- US Patent Application 14/859040; 20170083752
Encoded data along tape based on colorspace schemes
patent, November 2020
- Goodsitt, Jeremy; Walters, Austin
- US Patent Document 10,833,852
Learning apparatus and method for learning a model corresponding to real number time-series input data
patent-application, May 2018
- Osogami, Takayuki
- US Patent Application 15/358947; 20180144266
Learning coach for machine learning system
patent-application, June 2020
- Baker, James K.
- US Patent Application 16/334204; 20200184337
Medical image classification based on a generative adversarial network trained discriminator
patent, March 2021
- Madani, Ali; Moradi, Mehdi; Syeda-Mahmood, Tanveer F.
- US Patent Document 10,937,540
Progressively-trained scale-invariant and boundary-aware deep neural network for the automatic 3D segmentation of lung lesions
patent, January 2022
- Chen, Antong; Goldmacher, Gregory; Zhou, Bo
- US Patent Document 11,232,572
Image processing apparatus, image processing method, and non-transitory computer-readable storage medium
patent-application, January 2019
- Aoba, Masato; Komori, Yasuhiro
- US Patent Application 16/028633; 20190012582
Training and application method of neural network model, apparatus, system and storage medium
patent-application, May 2020
- Liu, Junjie; Chen, Tse-Wei; Wen, Dongchao
- US Patent Application 16/670940; 20200151514
Computer Program For Performing Drawing-Based Security Authentication
patent-application, October 2020
- Kim, Young Soo; Jo, Young Bak; Hwang, Yeongtae
- US Patent Application 16/848550; 20200338454
Training a neural network model
patent-application, November 2020
- Mavroeidis, Dimitrios; Gebre, Binyam Gebrekidan; Trajanovski, Stojan
- US Patent Application 16/768783; 20200372344
Method, device, and computer program for improving the reconstruction of dense super-resolution images from diffraction-limited images acquired by single molecule localization microscopy
patent-application, August 2020
- Zimmer, Christophe; Ouyang, Wei
- US Patent Application 16/635053; 20200250794
Method and apparatus for multi-level stepwise quantization for neural network
patent-application, November 2021
- Kim, Jin Kyu; Kim, Byung Jo; Kim, Seong Min
- US Patent Application 17/317607; 20210357753
Information processing apparatus and method
patent-application, August 2020
- Tachibana, Fumihiko
- US Patent Application 16/564286; 20200257983
Generative adversarial network medical image generation for training of a classifier
patent, March 2020
- Madani, Ali; Moradi, Mehdi; Syeda-Mahmood, Tanveer F.
- US Patent Document 10,592,779
V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation
conference, October 2016
- Milletari, Fausto; Navab, Nassir; Ahmadi, Seyed-Ahmad
- 2016 Fourth International Conference on 3D Vision (3DV)
Device and method for multiclass object detection
patent-application, April 2012
- Mei, Shuqi; Wu, Weiguo
- US Patent Application 13/257617; 20120089545