The Deep Learning Image Reconstruction option is a deep learning based reconstruction method intended to produce cross-sectional images of the head and whole body by computer reconstruction of X-ray transmission data taken at different angles and planes, including Axial, Helical (Volumetric), and Cardiac acquisitions, for all ages.

Deep Learning Image Reconstruction can be used for head, whole body, cardiac, and vascular CT applications.

Device Description

Deep Learning Image Reconstruction is an image reconstruction method that uses a dedicated Deep Neural Network (DNN) that has been designed and trained specifically to generate CT Images to give an image appearance, as shown on axial NPS plots, similar to traditional FBP images while maintaining the performance of ASiR-V in the following areas: image noise (pixel standard deviation), low contrast detectability, high-contrast spatial resolution, and streak artifact suppression.

The images produced are branded as "TrueFidelity™ CT Images". Reconstruction times with Deep Learning Image Reconstruction support a normal throughput for routine CT.

Deep Learning Image Reconstruction was trained specifically on the Revolution CT/Apex platform (K163213, K133705, K19177). The deep learning technology is integrated into the scanner's existing raw data-based image reconstruction chain to produce DICOM compatible "TrueFidelity™ CT Images".

The system allows user selection of three strengths of Deep Learning Image Recon: Low, Medium or High. The strength selection will vary with individual users' preferences and experience for the specific clinical need.

The DLR algorithm is now being modified on the Revolution family CT systems (K133705, K163213, K19177) for improved reconstruction speed and image quality, thus triggering this premarket notification.

AI/ML Overview

The provided document describes the Deep Learning Image Reconstruction (DLIR) device, its acceptance criteria, and the study conducted to prove it meets these criteria.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state numerical acceptance criteria with pass/fail thresholds for each metric. Instead, it focuses on demonstrating non-inferiority or improvement compared to a predicate device (ASiR-V) and ensuring diagnostic quality. The reported device performance is qualitative, indicating "significantly better subjective image quality" and "diagnostic quality images."

However, based on the non-clinical and clinical testing sections, we can infer the performance metrics evaluated.

Acceptance Criteria (Inferred from tests)	Reported Device Performance (Qualitative)
Image Quality Metrics (Objective - Bench Testing):	DLIR maintains performance similar to ASiR-V, with potential for improvement in noise characteristics.
- Low Contrast Detectability (LCD)	Evaluated. Aim to be similar to ASiR-V.
- Image Noise (pixel standard deviation)	Evaluated. Aim to be similar to ASiR-V. DLIR is designed to "identify and remove the noise."
- High-Contrast Spatial Resolution (MTF)	Evaluated. Aim to be similar to ASiR-V.
- Streak Artifact Suppression	Evaluated. Aim to be similar to ASiR-V.
- Spatial Resolution, longitudinal (FWHM slice sensitivity profile)	Evaluated. Aim to be similar to ASiR-V.
- Noise Power Spectrum (NPS) and Standard Deviation of noise	Evaluated. NPS plots show similar appearance to traditional FBP images.
- CT Number Uniformity	Evaluated. Aims to ensure consistency.
- CT Number Accuracy	Evaluated. Aims to ensure measurement accuracy.
- Contrast to Noise (CNR) ratio	Evaluated. Aims to ensure adequate contrast.
- Artifact analysis (metal objects, unintended motion, truncation)	Evaluated. Aims to ensure reduction or absence of artifacts.
- Pediatric Phantom IQ Performance Evaluation	Evaluated. Specific to pediatric imaging.
- Low Dose Lung Cancer Screening Protocol IQ Performance Evaluation	Evaluated. Specific to low-dose imaging protocols.
Subjective Image Quality (Clinical Reader Study):	"produce diagnostic quality images and have significantly better subjective image quality than the corresponding images generated with the ASiR-V reconstruction algorithm."
- Diagnostic Usefulness	Diagnostic quality images produced.
- Image Noise Texture	"Significantly better" subjective image quality.
- Image Sharpness	"Significantly better" subjective image quality.
- Image Noise Texture Homogeneity	"Significantly better" subjective image quality.
Safety and Effectiveness:	No additional risks/hazards, warnings, or limitations introduced. Substantially equivalent to predicate.

2. Sample Size Used for the Test Set and Data Provenance

Sample Size: 40 retrospectively collected clinical cases.
Data Provenance: Retrospectively collected clinical cases. The country of origin is not specified in the provided text.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

Number of Experts: 6 board-certified radiologists.
Qualifications of Experts: Board-certified radiologists with "expecialty areas that align with the anatomical region of each case."

4. Adjudication Method for the Test Set

The document describes a reader study where each of the 40 cases (reconstructed with both ASiR-V and DLIR) was read by 3 different radiologists independently. They provided an assessment of image quality using a 5-point Likert scale. There's no explicit mention of an adjudication process (e.g., 2+1, 3+1) if there were disagreements among the three readers, as the focus seems to be on independent assessment and overall subjective preference comparison.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

Yes, a multi-reader multi-case (MRMC) comparative effectiveness study was done. Human readers compared images reconstructed with DLIR (AI-assisted reconstruction) against images reconstructed with ASiR-V (without DLIR).

Effect Size: The study confirmed that DLIR (the subject device) produced diagnostic quality images and "have significantly better subjective image quality" than the corresponding images generated with the ASiR-V reconstruction algorithm. The text doesn't provide a specific numerical effect size (e.g., a specific improvement percentage or statistical metric), but it qualitatively states a "significant" improvement based on reader preference for image noise texture, image sharpness, and image noise texture homogeneity.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, extensive standalone (algorithm-only) performance testing was conducted. This is detailed in the "Additional Non-Clinical Testing" section, where DLIR and ASiR-V reconstructions of identical raw datasets were compared for various objective image quality metrics without human interpretation during these specific tests.

7. The Type of Ground Truth Used

The ground truth for the clinical reader study was established through expert consensus/assessment of image quality and preference by the participating radiologists. For the non-clinical bench testing, the ground truth was based on objective physical measurements and established phantom data with known properties.

8. The Sample Size for the Training Set

The document mentions that the Deep Neural Network (DNN) for DLIR was "trained specifically on the Revolution CT/Apex platform." However, it does not specify the sample size (number of images or cases) used for the training set.

9. How the Ground Truth for the Training Set was Established

The text states that the DNN was "trained specifically to generate CT Images to give an image appearance, as shown on axial NPS plots, similar to traditional FBP images while maintaining the performance of ASiR-V." It also notes that the DNN "models the scanned object using information obtained from extensive phantom and clinical data."

While the exact method for establishing ground truth for training isn't explicitly detailed, it implies a process where:

Reference Images: Traditional FBP (Filtered Back Projection) and ASiR-V images likely served as reference or target outputs for the DNN, specifically regarding image appearance, noise characteristics, and spatial resolution.
"Extensive phantom and clinical data": This data, likely corresponding to various anatomical regions, pathologies, and dose levels, was fed into the training process. The ground truth would involve teaching the network to reconstruct images that, when compared to conventionally reconstructed images (FBP/ASiR-V), exhibit desired image quality attributes (e.g., reduced noise while preserving detail).
Noise Modeling: The training process characterized "the propagation of noise through the system" to identify and remove it, suggesting a ground truth related to accurate noise modeling and reduction.

Summary

{0}------------------------------------------------

Image /page/0/Picture/0 description: The image shows the logos of the Department of Health & Human Services and the Food and Drug Administration (FDA). The Department of Health & Human Services logo is on the left, and the FDA logo is on the right. The FDA logo is a blue square with the letters "FDA" in white, followed by the words "U.S. FOOD & DRUG ADMINISTRATION" in blue.

February 18, 2022

GE Medical Systems, LLC. % Amy Yang Regulatory Affairs Manager 3000 N. Grandview Blvd. WAUKESHA WI 53188

Re: K213999

Trade/Device Name: Deep Learning Image Reconstruction Regulation Number: 21 CFR 892.1750 Regulation Name: Computed Tomography X-Ray System Regulatory Class: Class II Product Code: JAK Dated: December 20, 2021 Received: December 21, 2021

Dear Amy Yang:

We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database located at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.

If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's

{1}------------------------------------------------

requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR 803) for devices or postmarketing safety reporting (21 CFR 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (OS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.

For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Laurel Burk Assistant Director Diagnostic X-ray Systems Team Division of Radiological Health OHT7: Office of In Vitro Diagnostics and Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

{2}------------------------------------------------

Image /page/2/Picture/1 description: The image shows the logo for General Electric (GE). The logo consists of the letters "GE" in a stylized script, enclosed within a blue circle. There are three white swirls around the outside of the circle.

K213999

Section 4: Indications for Use Statement

Deep Learning Image Reconstruction (DLIR)

{3}------------------------------------------------

Indications for Use

510(k) Number (if known)

K213999

Device Name Deep Learning Image Reconstruction

Indications for Use (Describe)

The Deep Learning Image Reconstruction option is a deep learning based reconstruction method intended to produce cross-sectional images of the head and whole body by computer reconstruction of X-ray transmission data taken at different angles and planes, including Axial, (Volumetric), and Cardiac acquisitions, for all ages.

Deep Learning Image Reconstruction can be used for head, whole body, cardiac, and vascular CT applications.

Type of Use (Select one or both, as applicable)
☑ Prescription Use (Part 21 CFR 801 Subpart D)	☐ Over-The-Counter Use (21 CFR 801 Subpart C)

CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

{4}------------------------------------------------

Image /page/4/Picture/1 description: The image shows the logo for General Electric (GE). The logo consists of the letters "GE" in a stylized script, enclosed within a circle. The circle is surrounded by a series of curved lines, resembling water droplets or waves. The logo is colored in blue.

510(k) SUMMARY OF SAFETY AND EFFECTIVNESS K213999

This 510(k) summary of Safety and Effectiveness information is submitted in accordance with the requirement of 21 CFR Part 807.92:

Date:	December 20, 2021
Submitter:	GE Medical Systems, LLC3000 North Grandview BlvdWaukesha, WI 53188
Primary Contact:	Amy YangRegulatory Affairs ManagerGE HealthcarePhone: 1-414-514-3904Email: amy.yang@ge.com
Secondary Contacts:	Helen PengSenior Regulatory Affairs DirectorGE HealthcarePhone: 1-262-424-8222Email: hong.peng@med.ge.comNikolina MavrodievaRegulatory Affairs LeaderGE HealthcarePhone: 1-905-302-3913Email: nikolina.mavrodieva@ge.com
Subject Device Name:	Deep Learning Image Reconstruction
Device Classification	Class II
Regulation Number/Product Code:	21 CFR 892.1750 Computed tomography x-ray system / JAK
Predicate Device Information
Device Name:	Deep Learning Image Reconstruction
Manufacturer:	GE Medical Systems, LLC
510(k) Number:	K183202 cleared on April 12, 2019
Regulation Number/Product Code:	21 CFR 892.1750 Computed tomography x-ray system / JAK

{5}------------------------------------------------

Image /page/5/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized script, enclosed within a circle. The circle is surrounded by what appear to be stylized water droplets or splashes, giving the logo a dynamic and fluid appearance. The color of the logo is a bright, solid blue.

Reference Devices Information
Device Name:	ASiR-V
Manufacturer:	GE Medical Systems, LLC
510(k) Number:	K134640 cleared on March 25, 2014
Regulation Number/Product Code:	21 CFR 892.1750 Computed tomography x-ray system / JAK

. Devices Inform Dofa

Device Description

The images produced are branded as "TrueFidelity™ CT Images". Reconstruction times with Deep Learning Image Reconstruction support a normal throughput for routine CT.

Compared to the predicate device, the intended use and indications for use of Deep Learning Image Reconstruction are identical.

Intended Use

The Deep Learning Image Reconstruction option is intended for head, whole body, cardiac, and vascular CT scans.

Indications for Use

Deep Learning Image Reconstruction can be used for head, whole body, cardiac, and vascular CT applications.

{6}------------------------------------------------

Image /page/6/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized, cursive font, enclosed within a blue circle. There are decorative swirls or flourishes surrounding the letters within the circle, giving the logo a classic and recognizable appearance. The logo is simple, yet distinctive, and is widely recognized as the symbol of the General Electric company.

Comparisons

The modified Deep Learning Image Reconstruction (DLIR) option is substantially equivalent to the predicate device K183202. The modified DLIR is based on the same fundamental technology as the predicate device and is implemented on the Revolution CT/Apex platform (K133705, K163213, K19177). They utilize the same hardware and software platform technology on which substantial equivalence is demonstrated. The table below summarizes the substantive feature/technological similarities and differences between the predicate device and the proposed device:

Specification/ Attribute	Predicate DeviceDLIR for Revolution CT (K183202)	Proposed DeviceModified DLIR forRevolution CT/Apexplatform
Technology	DLIR uses a dedicated Deep NeuralNetwork (DNN) which is trained on theCT scanner and therefore models thepropagation of noise through the systemto identify and remove the noise	Same DNN technology withrevised networkarchitecture with retrainingand inferencing techniques
System statistics - Noisemodeling of the datacollection imaging chain(photon noise andelectronic noise)	Characterization of the photon statisticsas it propagates through thepreprocessing and calibration imagingchain	Same
System statistics - Noisecharacteristics of thereconstructed images	DLIR uses a trained DNN which modelsthe scanned object using informationobtained from extensive phantom andclinical data to identify the noisecharacteristics and remove it	Same
Clinical Workflow	Select recon type and strength (Low,Medium, High).	Same

The subject device Deep Learning Image Reconstruction does not introduce any additional risks/hazards, warnings, or limitations.

{7}------------------------------------------------

Image /page/7/Picture/1 description: The image shows the logo for General Electric (GE). The logo consists of the letters "GE" in a stylized script, enclosed within a blue circle. The circle has a design of swirling lines around the letters, giving the impression of movement or energy.

Determination of Substantial Equivalence

Summary of Non-Clinical Testing

Deep Learning Image Reconstruction has successfully completed the design control testing per our quality system. No additional hazards were identified, and no unexpected test results were observed. Deep Learning Image Reconstruction was designed under the Quality System Regulations of 21CFR 820 and ISO 13485. GE believes that the extensive bench testing and the physician evaluation are sufficient for FDA's substantial equivalence determination.

The following quality assurance measures have been applied to the development of the system:

. Requirement Definition
. Risk Analysis and Control
. Technical Design Reviews
Formal Design Reviews
. Software Development Lifecycle
- Code Review O
- O Software Unit Implementation
- Software Integrations and Integration Testing O
System Testing
- o Safety Testing (Verification)
- O Image Performance Testing (Verification)
- Simulating Use Testing (Validation) O
Software Release

The testing and results did not raise different questions of safety and effectiveness than associated with predicate device. We consider the proposed device is substantially equivalent to the predicate device, DLIR.

The substantial equivalence is also based on the software documentation for a "Moderate" level of concern.

Additional Non-Clinical Testing

Engineering bench testing was performed to support substantial equivalence and the product performance claims. The evaluation and analysis used the same test methodologies and acceptance criteria with the identical raw datasets obtained on GE's Revolution CT/Apex platform and then applying the Deep Learning Image Reconstruction or ASiR-V reconstruction (hence the dose (CTDIvol) is identical for both). The resultant images were then compared for:

Low Contrast Detectability (LCD)
. Image Noise (pixel standard deviation)
. High-Contrast Spatial Resolution (MTF)
Streak Artifact Suppression
Spatial Resolution, longitudinal (FWHM slice sensitivity profile)
Noise Power Spectrum (NPS) and Standard Deviation of noise
CT Number Uniformity

{8}------------------------------------------------

Image /page/8/Picture/1 description: The image shows the logo for General Electric (GE). The logo consists of the letters 'GE' in a stylized script, enclosed within a blue circle. There are also some white swirls around the circle, giving it a dynamic and fluid appearance.

CT Number Accuracy
Contrast to Noise (CNR) ratio
Artifact analysis – metal objects, unintended motion, truncation
Pediatric Phantom IQ Performance Evaluation
Low Dose Lung Cancer Screening Protocol IQ Performance Evaluation

Clinical Testing

The reader study used a total of 40 retrospectively collected clinical cases. The raw data from each of these cases was reconstructed with both ASiR-V and Deep Learning Image Reconstruction and presented side by side to each reader independently. The results of the study support substantial equivalence and performance claims.

These images were read by 6 board certified radiologists with expecialty areas that align with the anatomical region of each case. Each image was read by 3 different radiologists who provided an assessment of image quality related to diagnostic use according to a 5-point Likert scale. Three readers read the cases primarily covering body and extremity anatomy, and three different read the cases primarily covering neuro anatomy.

Additionally, the readers were asked to compare directly the ASIR-V and Deep Learning Image Reconstruction images according to three key metrics of image quality preference – image noise texture, image sharpness, and image noise texture homogeneity.

The result of this reader study confirmed that the DLIR (the subject device) produce diagnostic quality images and have significantly better subjective image quality than the corresponding images generated with the ASiR-V reconstruction algorithm.

Substantial Equivalence

The changes associated with Deep Learning Image Reconstruction do not change the Intended Use or indications for use from the predicate, and represent equivalent technological characteristics including the dedicated neural network, with no impact on control mechanism or operating principle.

Deep Learning Image Reconstruction was developed under GE Healthcare's quality system. Design verification, along with bench testing and the clinical reader study provided in this submission demonstrates that modified Deep Learning Image Reconstruction is substantially equivalent and hence as safe and as effective as the legally marketed predicate device. GE's quality system's design, verification, and risk management processes did not identify any additional hazards, unexpected results, or adverse effects stemming from the changes to the predicate.

GE Healthcare believes that Deep Learning Image Reconstruction is substantially equivalent to the predicate device and hence is safe and effective for its intended use.

Regulation Number and Section

§ 892.1750 Computed tomography x-ray system.

(a)
Identification. A computed tomography x-ray system is a diagnostic x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data from the same axial plane taken at different angles. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.