IP Library Granted Patent US 12,482,239
Granted Patent B2
US 12,482,239 · App. 18/298,524 · Granted Nov 25, 2025

Self-supervised learning for medical image quality control

Inventors: Ben Andrew Duffy (Palo Alto, CA); Gajanana Keshava Datta (Los Altos, CA); Enhao Gong (Sunnyvale, CA)
Assignee: Subtle Medical, Inc.
G06V10/7753G06T7/0012G06V10/82G06T2207/10072G06V2201/03
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,482,239
App. No.
18/298,524
Granted
Nov 25, 2025
Kind
B2
Abstract

Provided herein are methods for automated image quality control (QC). The method comprises: generating training data based at least in part on metadata obtained from a data augmentation process; training a model for a QC task based at least in part on the training data. The model is trained using a self-supervised learning algorithm.

Claims (20)

1 . A computer-implemented method for automated image quality control (QC), the method comprising:

generating training data based at least in part on metadata obtained from a data augmentation process; and

training a model for a QC task based at least in part on the training data, wherein the QC task comprises an out-of-distribution (OOD) detection,

wherein the model is trained using a self-supervised learning algorithm and is trained to predict metadata from an input image and

i) wherein an OOD event is determined when the predicted metadata does not match the metadata or

ii) wherein an OOD event is determined when an uncertainty score of the metadata prediction is beyond a predetermined threshold.

2 . The computer-implemented method of claim 1 , wherein supervised learning algorithm is contrastive learning.

3 . The computer-implemented method of claim 1 , wherein the metadata is extracted from a header of an image.

4 . The computer-implemented method of claim 3 , wherein the metadata is used to generate a label for the image.

5 . The computer-implemented method of claim 1 , wherein the QC task is image registration quality control.

6 . The computer-implemented method of claim 5 , wherein the model is trained to predict an alignment based on a similarity map between embeddings of two input images.

7 . The computer-implemented method of claim 6 , wherein the embeddings are local embeddings or global embeddings produced by an encoder-decoder network.

8 . A computer-implemented method for automated image quality control (QC), the method comprising:

generating training data based at least in part on metadata obtained from a data augmentation process, wherein the data augmentation process comprises generating cropped patch with simulated artifact; and

training a model for a QC task based at least in part on the training data, wherein the model is trained using a self-supervised learning algorithm.

9 . The computer-implemented method of claim 8 , further comprising replacing a corresponding patch in an input image with the cropped patch with the simulated artifact.

10 . A computer-implemented method for automated image quality control (QC), the method comprising:

generating training data based at least in part on metadata obtained from a data augmentation process, wherein the training data comprises 3D image including a stack of slices and wherein generating the training data comprises generating a label for the 3D image using a multiple-instance-learning method; and

training a model for a QC task based at least in part on the training data, wherein the model is trained using a self-supervised learning algorithm.

11 . The computer-implemented method of claim 10 , further comprising determining a pooling strategy for combining embeddings of one or more slices based on a selected multiple-instance-learning assumption.

Assignments (2)
GRANT OF SECURITY INTEREST IN PATENTS Recorded May 29, 2026
From: SUBTLE MEDICAL, INC.
To: MS PRIVATE CREDIT ADMINISTRATIVE SERVICES LLC
Reel/Frame 075648/0001 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 12, 2023
From: DUFFY, BEN ANDREW; DATTA, GAJANANA KESHAVA; GONG, ENHAO
To: SUBTLE MEDICAL, INC.
Reel/Frame 063920/0908 →
Continuity (2)
Provisional Application 63330401 · Apr 13, 2022
Related Publication 20230386184A1 · Nov 30, 2023
References Cited (16)
US 20210056412A1 · Jung · 2021 [cited by applicant]
US 20210279868A1 · Ma et al. · 2021 [cited by applicant]
US 20220114733A1 · Feng · 2022 [cited by examiner]
US 20230074706A1 · Xiao · 2023 [cited by examiner]
WO WO2021041125A1 · 2021 [cited by applicant]
WO WO2023200772A1 · 2023 [cited by applicant]
MedAug, Vu et al 2012 (Year: 2012). [cited by examiner]
DICOM metadata, Hu et al 2020 (Year: 2020). [cited by examiner]
Duffy et al. Retrospective correction of motion artifact affected structural MRI images using deep learning of simulated motion. 1st Conference on Medical Imaging with Deep Learning (MIDL 2018), Amsterdam, The Netherlan… [cited by applicant]
Hu et al. Self-Supervised Pretraining with DICOM Metadata in Ultrasound Imaging. Proceedings of Machine Learning Research 1-17, 2020. [cited by applicant]
IXI Dataset. Retrieved from the internet on Aug. 25, 2023 at: https://brain-development.org/ixi-dataset/. 2 pages. [cited by applicant]
PCT/US2023/018141 International Search Report and Written Opinion dated Jun. 29, 2023. [cited by applicant]
Selvaraju et al. Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization. In Proceedings of the IEEE International Conference on Computer Vision 2017 (pp. 618-626). [cited by applicant]
Vu et al. MedAug: Contrastive learning leveraging patient metadata improves representations for chest X-ray interpretation. Proceedings of Machine Learning Research 126:1-14. 2021. [cited by applicant]
Webpage: Openneuro.org. Retrieved from the internet on Aug. 25, 2023. 9 pages. [cited by applicant]
Yun et al. CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features. In Proceedings of the IEEE/CVF International Conference on Computer Vision 2019 (pp. 6023-6032). [cited by applicant]
Cited By (1)
US 12,572,552