IP Library › Granted Patent US 12,555,284
Granted Patent B2
US 12,555,284 · App. 18/276,665 · Granted Feb 17, 2026

Training data synthesizer for contrast enhancing machine learning systems

Inventor: Liran Goshen (Pardes-Hanna, IL)
Assignee: KONINKLIJKE PHILIPS N.V.
G06T11/60G06T5/70G06T2207/30168
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 12,555,284
App. No.
18/276,665
Granted
Feb 17, 2026
Kind
B2
Abstract

A system (DSS) and related method for synthesizing training data or machine learning, based on a set (TD) including two types of training imagery, high image quality, IQ, imagery and low IQ imagery. The system comprises a data synthesizer (DSY), configured to register the at least two types of imagery and to transfer i) image information from high IQ imagery to the registered low IQ imagery to obtain synthesized high IQ imagery, or ii) image information from low IQ imagery to the registered high IQ imagery to obtain synthesized low IQ imagery. The synthesized data may be used for improved training of machine learning models for IQ enhancement.

Claims (23)

1 . A system for synthesizing medical training data for training a machine learning model, comprising:

a memory that stores a plurality of instructions; and

a processor coupled to the memory and configured to execute the plurality of instructions to:

register at least two types of imagery comprising high image quality (IQ) imagery and low image quality (IQ) imagery, wherein the IQ comprises at least one of an image contrast level and an image artifact level;

denoise the registered imagery to remove a noise component;

transfer i) image information from the high IQ imagery to the registered low IQ imagery to obtain synthesized high IQ imagery, or ii) image information from the low IQ imagery to the registered high IQ imagery to obtain synthesized low IQ imagery, wherein the image information is data associated with the IQ;

add the noise component to the synthesized low or high IQ imagery; and

provide the synthesized high and/or low IQ imagery for training the machine learning model.

2 . The system of claim 1 , wherein the high IQ imagery is recorded by an imaging apparatus while a contrast agent is present in a field of view of the imaging apparatus, and the low IQ imagery is recorded by the imaging apparatus while a smaller amount of the contrast agent is present in the field of view of the imaging apparatus than during the high IQ imagery.

3 . The system of claim 1 , wherein the synthesized high or low IQ imagery is added to the medical training data to obtain an enlarged medical training data.

4 . The system of claim 1 , wherein a proportion of imagery to be synthesized is adjustable via a user interface.

5 . A computer-implemented method of synthesizing medical training data for training a machine learning model, comprising:

registering at least two types of imagery comprising high image quality (IQ) imagery and low image quality (IQ) imagery, wherein the IQ comprises at least one of an image contrast level and an image artifact level;

denoise the registered imagery to remove a noise component;

transferring i) image information from the high IQ imagery to the registered low IQ imagery to obtain synthesized high IQ imagery, or ii) image information from the low IQ imagery the registered high IQ imagery to obtain synthesized low IQ imagery, wherein the image information is data associated with the IQ;

add the noise component to the synthesized low or high IQ imagery; and

providing the synthesized high and/or low IQ imagery for training the machine learning model.

6 . A computer-implemented method for enhancing an image, comprising:

receiving an input image; and

processing the input image into an enhanced image based on a machine learning model trained according to the method of claim 5 .

7 . A non-transitory computer-readable medium for storing executable instructions, which cause a method to be performed to enhance an image, the method comprising:

receiving an input image; and

processing the input image into an enhanced image based on a machine learning model trained according to the method of claim 5 .

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 10, 2023
From: GOSHEN, LIRAN
To: KONINKLIJKE PHILIPS N.V.
Reel/Frame 064545/0417 →
Priority Claims (1)
EP 21157094 · Feb 15, 2021 · regional
Continuity (1)
Related Publication 20240312086A1 · Sep 19, 2024
References Cited (56)
US 8938110B2 · Goshen · 2015 [cited by applicant]
US 9159124B2 · Goshen · 2015 [cited by applicant]
US 9547889B2 · Goshen · 2017 [cited by applicant]
US 10282820B2 · Goshengoshen · 2019 [cited by applicant]
US 20190108634A1 · Zaharchuk · 2019 [cited by applicant]
US 20190333199A1 · Ozcan · 2019 [cited by examiner]
US 20210390696A1 · Iwase · 2021 [cited by examiner]
US 20220044074A1 · Li · 2022 [cited by examiner]
US 20220207680A1 · Wang · 2022 [cited by examiner]
US 20240037732A1 · Gong · 2024 [cited by examiner]
EP 3576050A1 · 2019 [cited by applicant]
EP 3739522A1 · 2020 [cited by applicant]
WO WO20200234051A1 · 2020 [cited by applicant]
PCT International Search Report, International application No. PCT/EP2022/052955, Jun. 21, 2022. [cited by applicant]
Zeiler M. D. et al., “Adaptive Deconvolutional Networks for Mid and High Level Feature Learning”, 2011 International Conference on Computer Vision, Barcelona, Spain, pp. 2018-2025, 2011. [cited by applicant]
Goshen L. et al., “An Iodine-Calcium Separation Analysis and Virtually Non-Contrasted Image Generation Obtained with Single Source Dual Energy MDCT”, 2008 IEEE Nuclear Science Symposium Conference Record, pp. 3868-3870,… [cited by applicant]
Ioffe S. et al., “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift”, online TarXiv:1502.03167v3 [cs.LG] Mar. 2, 2015. [cited by applicant]
Li S. et al., “Pixel-Level Image Fusion: A Survey of the State of the Art”, Information Fusion, Elsevier, US, vol. 33, May 19, 2016, pp. 100-112, XP029596965. [cited by applicant]
Maayan F-A et al., “GAN-Based Synthetic Medical Image Augmentation for increased CNN Performance in Liver Lesion Classification”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 148… [cited by applicant]
Yi X. et al., “Generative Adversarial Network in Medical Imaging: A Review”, arXiv:1809.07294v4 [cs.CV] , Sep. 2019. [cited by applicant]
Tang Y. et al., “CT Image Enhancement Using Stacked Generative Adversarial Networks and Transfer Learning for Lesion Segmentation Improvement”, arXiv:1807.07144v1 [cs.CV], Jul. 18, 2018. [cited by applicant]
Dar S.H. et al., “Image Synthesis in Multi-Contrast MRI with Conditional Generative Adversarial Networks”, arXiv:1802.01221, Feb. 2018. [cited by applicant]
Armanious K. et al., “MedGAN: Medical Image Translation Using GANs”, arXiv:1806.06397v2 [cs.CV] , Apr. 4, 2019. [cited by applicant]
Wang C.L et al., “Frequency, Outcome, and Appropriateness of Treatment of Nonionic Iodinated Contrast Media Reactions”, AJR American Journal of Roentgenology, vol. 191, No. 2, pp. 409-415 , Aug. 2008. [cited by applicant]
Mortele K.J. et al., “Universal Use of Nonionic lodinated Contrast Medium for CT: Evaluation of Safety in a Large Urban Teaching Hospital”, AJR American Journal of Roentgenology, vol. 184, Issue 1, pp. 31-34, 2005. [cited by applicant]
Wong P.C.Y. et al., “Pathophysiology of Contrast-Induced Nephropathy”, International Journal of Cardiology, vol. 158, Issue 2, pp. 186-192, Jul. 2, 2012. [cited by applicant]
Wong G.T.C. et al., “Contrast-Induced Nephropathy”, British Journal of Anesthesia (BJA), vol. 99, Issue 4, pp. 474-483, 2007. [cited by applicant]
Solomon R. et al., “Follow-Up of Patients with Contrast Induced Nephropathy”, Kidney International Suppl, vol. 100, pp. S46-S50, 2006. [cited by applicant]
Barrett B.J. et al., “Preventing Nephropathy Induced by Contrast Medium”, New England Journal of Medicine, vol. 354, pp. 379-386, Clinical Practice, 2006. [cited by applicant]
Barrett B.J. et al., “Contrast Induced Nephropathy in Patients with Chronic Kidney Disease Undergoing Computed Tomography: A Double-Blind Comparison of Iodixanol and Iopamidol”, Investigative Radiology, vol. 41, No. 11,… [cited by applicant]
Nash K. et al., “Hospital-Acquired Renal Insufficiency”, American Journal of Kidney Diseases, vol. 39, No. 5, pp. 930-936, 2002. [cited by applicant]
Kobayashi K. et al., “Screening Methods for Early Detection of Hepatocellular Carcinoma”, Hepatology, vol. 5 Issue 6, pp. 1100-1105, 1985. [cited by applicant]
De Ledinghen V. et al., “Detection of Nodules in Liver Cirrhosis: Spiral Computed Tomography or Magnetic Resonance Imaging? A Prospective Study of 88 Nodules in 34 Patients”, European Journal of Gastroenterology & Hepat… [cited by applicant]
Napel S. et al., “CT Angiography with Spiral CT and Maximum Intensity Projection”, Radiology, vol. 185, No. 2, pp. 607-610, 1992. [cited by applicant]
Marks M.P. et al., “Diagnosis of Carotid Artery Disease: Preliminary Experience with Maximumintensity-Projection Spiral CT Angiography”, AJR American Journal of Roentgenology, 160(6):1267-1271, 1993. [cited by applicant]
Diederichs C.G. et al., “Blurring of Vessels in Spiral CT Angiography: Effects of Collimation Width, Pitch, Viewing Plane, and Windowing in Maximum Intensity Projection”, Journal of Computer Assisted Tomography, 20(6):9… [cited by applicant]
Jeon Y.W. et al., “Dynamic CT Perfusion Imaging for the Detection of Crossed Cerebellar Diaschisis in Acute Ischemic Stroke”, Korean Journal of Radiology, vol. 13, No. 1, pp. 12-19, 2012. [cited by applicant]
Lin K., et al., “Measuring Elevated Microvascular Permeability and Predicting Hemorrhagic Transformation in Acute Ischemic Stroke Using First-Pass Dynamic Perfusion CT Imaging”, AJNR American Journal of Neuroradiology, … [cited by applicant]
Eastwood J.D. et al., “Correlation of Early Dynamic CT Perfusion Imaging with Whole-Brain MR Diffusion and Perfusion Imaging in Acute Hemispheric Stroke”, AJNR American Journal of Neuroradiol, vol. 24, No. 9, pp. 1869-1… [cited by applicant]
Nakaguchi H. et al., “Efficacy of Dynamic CT Perfusion Imaging in Conjunction with Three Dimensional CT Angiography for the Evaluation of Acute Ischemic Stroke”, Neurological Surgery, 31(1):17-25 , 2003. [cited by applicant]
Mayer T.E. et al., “Dynamic CT Perfusion Imaging of Acute Stroke”, AJNR American Journal of Neuroradiol, vol. 21, No. 8, pp. 1441-1449, Sep. 2000. [cited by applicant]
Noël P. B. et al., “A Method for Improving Iodine Contrast Enhancement in Abdominal Computed Tomography: Experimental Study in a Pig Model”, European Radiology, vol. 23, No. 4, pp. 985-990, 2013. [cited by applicant]
Teague S.D. et al., “Potential Clinical and Economic Benefits of Low-Contrast-Dose CT Angiography”, Applied Radiology, 38(3), 40A., Mar. 2009. [cited by applicant]
Freiman M. et al., “Unsupervised Abnormality Detection Through Mixed Structure Regularization (MSR) in Deep Sparse Auto-Encoders”, Medical Physics, vol. 46, issue 5, pp. 2223-2231, 2019. [cited by applicant]
Long J. et al., “Fully Convolutional Networks for Semantic Segmentation”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3431-3440, 2015. [cited by applicant]
Ronneberger O. et al., “U-Net: Convolutional Networks for Biomedical Image Segmentation.”, Medical Image Computing and Computer-Assisted Intervention (MICCAI), Springer, LNCS, vol. 9351, pp. 234-241, 2015. [cited by applicant]
Robbins H. et al., “A Stochastic Approximation Method”, The Annals of Mathematical Statistics, vol. 22, Issue 3, op. 400-407, Sep. 1951. [cited by applicant]
Kingma D. P. et al., “ADAM: A Method for Stochastic Optimization”, arXiv preprint arXiv:1412.6980, 2014. [cited by applicant]
Zhu J-Y et al., “Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks”, Proceedings of the IEEE International Conference on Computer Vision, pp. 2223-2232, 2017. [cited by applicant]
He K. et al., “Deep Residual Learning for Image Recognition”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016. [cited by applicant]
He K. et al., “Identity Mappings in Deep Residual Networks”, European Conference on Computer Vision, pp. 630-645, Springer, Cham., Oct. 2016. [cited by applicant]
Huang G. et al., “Densely Connected Convolutional Networks”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4700-4708, 2017. [cited by applicant]
Goodfellow I.J. et al., “Generative Adversarial Nets”, NIPS'14: Proceedings of the 27th International Conference on Neural Information Processing Systems, vol. 2, pp. 2672-2680, 2014. [cited by applicant]
Qi G-J et al., “Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities”, Computer Vision and Pattern Recognition (cs.CV), arXiv preprint arXiv:1701.06264 (2017). [cited by applicant]
Arjovsky M. et al., “Wasserstein GAN”, Machine Learning (stat.ML); Machine Learning (cs.LG), arXiv:1701.07875 [stat.ML], 2017. [cited by applicant]
Mitchell T. M. et al., “Machine Learning”, p. 2, section 1.1, McGraw-Hill, 1997. [cited by applicant]