How AI Reconstructs Clearer MRI Scans Using Progressive Networks
This patent describes using a special type of artificial intelligence called a progressive generative adversarial network to reconstruct high-resolution magnetic resonance images, making them clearer and more detailed for doctors.
Original patent title: “Progressive generative adversarial network in medical image reconstruction”
This patent describes using a special type of artificial intelligence called a progressive generative adversarial network to reconstruct high-resolution magnetic resonance images, making them clearer and more detailed for doctors. Granted to Siemens Healthcare in 2022 with 21 claims and 1 forward citation, and it is expected to expire in 2040.
Coverage
What does this patent actually cover?
This patent outlines a method for reconstructing magnetic resonance (MR) images from raw scan data. It uses an image processor to apply a 'regularizer' during reconstruction, which is a machine-learned progressive generator from a progressive generative adversarial network (PGAN). This generator is an 'image-to-image U-net type network' integrated into an 'unrolled iterative reconstruction algorithm' (ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 1). The PGAN generator is trained by starting at a low resolution and progressively adding new layers to model finer details as training advances (AbstractabstractA short summary at the front of the patent describing the invention. Not legally binding.Read more →, Claim 7). For example, after an MR system scans a patient and collects k-space measurements, this method reconstructs a high-quality image by iteratively refining it with the PGAN, then displays the clear MR image.
The gap
What does this patent NOT cover?
- Does not cover medical image reconstruction methods that do not specifically use a generative adversarial network (GAN) as a regularizer.
- Does not cover GANs that are not progressively trained, meaning they don't start at low resolution and add layers for higher detail (ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 7, 8).
- Does not cover reconstruction algorithms that are not an 'unrolled iterative reconstruction' (ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 1, 5).
- Does not cover generators that are not an 'image-to-image U-net type network' (ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 1).
- Does not cover the reconstruction of images from medical modalities other than Magnetic Resonance (MR) imaging (ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 1).
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
Key facts
What made this novel
The innovation lies in using a 'progressive' training approach for the generative adversarial network. Instead of training a full-resolution network from scratch, it starts with low-resolution images and gradually adds layers to learn finer details. This makes the network more stable and better at handling the complex, high-resolution information needed for medical imaging.
The Patent Drawing
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
Siemens Healthineers MR imaging systems
Advanced diagnostic MRI scanners
AI-powered medical image processing software
Clinical radiology workflows
Why it matters
The bigger picture
Clearer and more detailed MR images are crucial for accurate medical diagnoses. This technology aims to improve image quality, potentially allowing doctors to detect subtle abnormalities more easily. It could also lead to faster scan times, making the MR experience more comfortable for patients and increasing patient throughput in clinics.
Filed
June 10, 2020
Granted
August 23, 2022
Market context
Who's building on this
Companies in this space
Siemens Healthineers, the assigneeassigneeThe entity that owns the patent — usually the inventor's employer or a company.Read more →, is a major player in medical imaging and continues to develop advanced AI solutions for their MR systems. Other large medical technology companies like GE Healthcare and Philips Healthcare are also heavily invested in using AI and machine learning to enhance their imaging products and improve diagnostic capabilities.
Market impact
This patent contributes to the ongoing shift towards AI-enhanced medical diagnostics. By enabling higher quality and potentially faster MR image reconstruction, it can lead to more efficient clinical workflows and improved diagnostic accuracy. This technology helps differentiate advanced MR systems in a competitive market, pushing the boundaries of what AI can achieve in healthcare.
Claim 1 — Plain English
What this patent covers
This patent outlines a method for reconstructing magnetic resonance (MR) images from raw scan data. It uses an image processor to apply a 'regularizer' during reconstruction, which is a machine-learned progressive generator from a progressive generative adversarial network (PGAN). This generator is an 'image-to-image U-net type network' integrated into an 'unrolled iterative reconstruction algorithm' (Claim 1). The PGAN generator is trained by starting at a low resolution and progressively adding new layers to model finer details as training advances (Abstract, Claim 7). For example, after an MR system scans a patient and collects k-space measurements, this method reconstructs a high-quality image by iteratively refining it with the PGAN, then displays the clear MR image.
The clever bit
The innovation lies in using a 'progressive' training approach for the generative adversarial network. Instead of training a full-resolution network from scratch, it starts with low-resolution images and gradually adds layers to learn finer details. This makes the network more stable and better at handling the complex, high-resolution information needed for medical imaging.
What it does not cover
- Does not cover medical image reconstruction methods that do not specifically use a generative adversarial network (GAN) as a regularizer.
- Does not cover GANs that are not progressively trained, meaning they don't start at low resolution and add layers for higher detail (Claim 7, 8).
- Does not cover reconstruction algorithms that are not an 'unrolled iterative reconstruction' (Claim 1, 5).
- Does not cover generators that are not an 'image-to-image U-net type network' (Claim 1).
- Does not cover the reconstruction of images from medical modalities other than Magnetic Resonance (MR) imaging (Claim 1).
Patent timeline
Application submitted to the patent office
Application published, typically 18 months after filing
Patent officially issued
Patent enters public domain
PatentBrief Score
Impact Score
Strong
Citation count
6/40
Early citations
Claim breadth
14/20
Broad claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →
Recency
20/20
Granted within 5 years
Assignee scale
20/20
Major company or institution
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$59K – $187K
Midpoint $117K · 13.6 yr remaining · industry ×1.5
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
Claim text not yet imported for this patent
The original legal language
Original claims
21 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Arberet, S., Mailhe, B., Nadar, M. S., & Moreau, A. L. J. (2022). How AI Reconstructs Clearer MRI Scans Using Progressive Networks (U.S. Patent No. 11,422,217). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/11422217/progressive-generative-adversarial-network-in-medical-image-reconstruction
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US11422217"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
Semantically similar
You might also find these interesting
US 11756161 · 2023 · Shenzhen Institutes of Advanced Technology
AI Method for Sharpening Low-Dose PET Scans with Multi-Task Learning
US 11580646 · 2023 · Nanjing University of Posts and Telecommunications
Improving Medical Image Segmentation with AI and Fake Data
US 12141700 · 2024 · Naver
How AI Generates Images Based on Style and Content Cues
US 12737843 · 2026
Upscaling Images with AI and Sub-Pixel Information
More to explore
More in AI & Machine Learning
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11615208 · 2023 · Capital One Services LLC
How Cloud Systems Automatically Create and Train AI Data Models
US 10402750 · 2019 · Facebook Inc
How Facebook Uses Deep Learning to Predict What You Might Like
New to patents?
Common Questions
Frequently Asked Questions
What does How AI Reconstructs Clearer MRI Scans Using Progressive Networks cover?
This patent describes using a special type of artificial intelligence called a progressive generative adversarial network to reconstruct high-resolution magnetic resonance images, making them clearer and more detailed for doctors.
Who owns patent US 11422217?
Siemens Healthcare owns this patent, granted in 2022.
When does this patent expire?
This patent is expected to expire on June 10, 2040, when the invention enters the public domain.
What is patent US 11422217 cited by?
This patent has been cited by 1 later patents that build on its ideas.
What problem does this patent solve?
Clearer and more detailed MR images are crucial for accurate medical diagnoses. This technology aims to improve image quality, potentially allowing doctors to detect subtle abnormalities more easily. It could also lead to faster scan times, making the MR experience more comfortable for patients and increasing patient throughput in clinics.
What does this patent NOT cover?
Does not cover medical image reconstruction methods that do not specifically use a generative adversarial network (GAN) as a regularizer.
Patent monitoring