HepaFatSmart is intended for the quantitative measurement of volumetric liver fat fraction (VLFF), proton density fat fraction (PDFF) and steatosis grading.

HepaFatSmart is an application that is used for the non-invasive evaluation of liver tissue by utilising magnetic resonance images to evaluate the difference in resonance frequencies between hydrogen nuclei in water and triglyceride fat. The quantitative triglyceride fat fraction is based on the measurement of a magnetic resonance parameter that reflects the ratio of the proton density signal of triglyceride fat to the total proton density signal in the liver.

Indications for use:

Support clinical diagnoses in individuals with confirmed or suspected fatty liver disease;

Support the subsequent clinical decision making processes for patients under management for fatty liver related disease or metabolic syndromes;

Aid in the assessment and screening of living donors for liver transplant.

Results, when interpreted by a trained physician can be used to support clinical diagnoses about the status of liver fat content, the subsequent clinical decision making processes for the management of fatty liver related diseases, metabolic syndromes, liver donor screening and lifestyle change. HepaFatSmart can be used to analyse the MRI images of patients of all populations independent of age and gender, with suspected clinical conditions related to the level of liver fat.

Device Description

HepaFatSmart is an SaMD designed to automatically analyse magnetic resonance imaging (MRI) datasets for quantitative assessment of a patient's liver fat, in form of volumetric liver fat fraction (VLFF), proton density fat fraction (PDFF), and steatosis grade. It is an Al assisted, automated version of HepaFat-Scan (another SaMD of Resonance Health). To carry out an analysis, the user simply uploads DICOM images to FAST, Resonance Health's secured user portal and job management system. No other user input is required for the analysis thereby minimising the impact of human error on obtained results. HepaFatSmart requires DICOM images as input data that have been acquired according to the HepaFatSmart (same as HepaFat-Scan) protocol.

AI/ML Overview

The provided documentation describes the acceptance criteria and a study proving that HepaFatSmart (V2.0.0) meets these criteria, demonstrating its substantial equivalence to the predicate device HepaFat-Scan and improved performance over HepaFat-AI.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of predetermined acceptance criteria for the new device prior to testing. Instead, it demonstrates the device's performance by comparing it to a reference standard (HepaFat-Scan) and an existing predicate (HepaFat-AI). The "acceptance" is implied by demonstrating substantial equivalence to HepaFat-Scan and improvement over HepaFat-AI, primarily through quantitative metrics like bias, repeatability, and limits of agreement, as well as sensitivity and specificity for clinically relevant thresholds.

However, based on the provided data, we can infer the performance metrics used for comparison, which implicitly serve as the targets for acceptance.

Inferred Acceptance Criteria (Implicitly compared to HepaFat-Scan and better than HepaFat-AI):

Performance Metric	HepaFatSmart (v2.0.0) Reported Performance	Inferred Acceptance Criterion (based on HepaFat-Scan/HepaFat-AI comparison)
Repeatability (VLFF)	Bias: -0.1 (-0.14)Upper 95% Repeatability: 1.5Lower 95% Repeatability: -1.8100% reproducible (zero VLFF difference) on duplicate analysis of same datasets	Comparable to or better than HepaFat-Scan (Bias: -0.2; Upper: 1.9; Lower: -2.3)
Agreement with HepaFat-Scan (Validation Study VLFF)	Bias: 0.2 (0.19)Upper 95% Limits of Agreement: 1.7Lower 95% Limits of Agreement: -1.3	Bias small and clinically insignificant; Limits of agreement significantly better than HepaFat-AI (Bias: 0.4; Upper: 5.4; Lower: -4.6) and comparable to HepaFat-Scan repeatability.
Sensitivity for VLFF Detection (vs. HepaFat-Scan)	4.1% Threshold: 100.0% (97.3-100.0% CI)12.1% Threshold: 98.8% (93.6-99.8% CI)16.2% Threshold: 100.0% (93.8-100.0% CI)	Well above 90% (specifically 100% or close to 100%)
Specificity for VLFF Detection (vs. HepaFat-Scan)	4.1% Threshold: 98.6% (94.9-99.6% CI)12.1% Threshold: 98.0% (94.9-99.2% CI)16.2% Threshold: 99.6% (97.5-99.9% CI)	Well above 90% (specifically 100% or close to 100%)
Image Quality Control (IQC)	Applied to filter valid datasets (281 out of 300 passed)	Ensure input data meet quality standards for reliable analysis.

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

Test Set Sample Size:
- Repeatability Study: n = 42 subjects initially; n = 41 subjects for HepaFatSmart analysis (one case identified as high iron and excluded by the new algorithm).
- Validation Study: n = 300 initially; n = 281 datasets successfully passed the IQC rules and were used for analysis.
Data Provenance: The document states the data used for the validation study comprised "fully quarantined 300 validation subjects with different clinical conditions across a broad age range and fat level scanned from different MRI centres with different MRI makes and models." While explicit countries of origin are not specified, the mention of "different MRI centres" and "different MRI makes and models" suggests a diverse, real-world dataset. The data appears to be retrospective, as it refers to a "quarantined dataset" used for validation.

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

The ground truth for this study is established by the reference standard HepaFat-Scan. HepaFat-Scan is described as a "Standalone software application to facilitate the import and visualization of multi-slice, gradient-echo MRI data sets... to provide objective and reproducible determination of the triglyceride fat fraction in magnetic resonance images of the liver." It performs quantitative triglyceride fat fraction measurements.

The document states that HepaFat-Scan is "Algorithmic, with human interaction for Region of Interest (ROI) selection." The "User" for HepaFat-Scan is listed as "Resonance Health's trained analyst."

Therefore:

Number of Experts/Analysts: The document doesn't specify a number, but rather a type of user: "Resonance Health's trained analyst." This implies trained personnel, but not necessarily multiple independent experts in a consensus-building scenario.
Qualifications of Experts: "Resonance Health's trained analyst." No further specific qualifications (e.g., years of experience, radiologist vs. technician) are provided for the individuals performing the HepaFat-Scan ground truth ROI selection. The emphasis is on the software being the reference standard.

4. Adjudication Method for the Test Set

The primary ground truth is derived from the HepaFat-Scan software with human ROI selection. The document does not describe any specific adjudication method (e.g., 2+1, 3+1 consensus by multiple readers) for establishing the ground truth measurements. The HepaFat-Scan output, with its human-selected ROI, appears to be accepted as the reference.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done

No, a traditional MRMC comparative effectiveness study was not explicitly done to show human readers improve with AI vs. without AI assistance.

This study is focused on the standalone performance of HepaFatSmart and its equivalence/superiority to existing automated or semi-automated software solutions (HepaFat-Scan and HepaFat-AI). HepaFatSmart is an "AI-assisted, automated version" of HepaFat-Scan. HepaFat-Scan itself involves "human interaction for Region of Interest (ROI) selection." The document implies that HepaFatSmart's full automation (no user input for analysis, AI-predicted ROI) minimizes human error and makes it potentially better or comparable to the human-assisted HepaFat-Scan.

The statement: "Bias and both repeatability coefficients for the HepaFatSmart are slightly better than those obtained from the repeated scans of HepaFat-Scan, indicating the performance of the HepaFatSmart is comparable (no worse) than human for the repeatability data analysed. This does not suggest yet that the HepaFatSmart is better than human analyst as the original human analysis (HepaFat-Scan) historically used two small liver ROIs rather than a single large liver ROI used in the HepaFatSmart with potentially slightly larger sampling error in the original HepaFat-Scan analysis."
And: "HepaFatSmart demonstrated 100% repeatable (reproducible) in the repeatability study using the same datasets analysed twice with zero VLFF difference between the first and second analyses, which is better than human analysis with HepaFat-Scan."
These statements imply a comparison to "human analysis" (as part of HepaFat-Scan's workflow), but it's not a formal MRMC study as typically understood where human readers' diagnostic performance is measured with and without AI assistance on the same cases.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) was Done

Yes, a standalone performance study was foundational to this submission. HepaFatSmart is described as a "Standalone software platform" that "automatically analyse... MRI datasets... No other user input is required for the analysis thereby minimising the impact of human error on obtained results." The entire study evaluates the performance of this algorithm-only approach (HepaFatSmart) by comparing its output directly to the output of the human-assisted reference standard (HepaFat-Scan) and the previous AI version (HepaFat-AI).

7. The Type of Ground Truth Used

The ground truth used for the test set is expert-assisted software output (HepaFat-Scan).
Specifically, it's the "HepaFat-Scan" software's quantitative measurement of Volumetric Liver Fat Fraction (VLFF), which involves human interaction for Region of Interest (ROI) selection. This serves as the "reference standard" against which HepaFatSmart's automated results are compared.

8. The Sample Size for the Training Set

The document does not provide the sample size directly for the training set of the HepaFatSmart AI model. It notes that the system is "completely 'locked down' for final validation prior to release in commercial use to ensure reproducibility of the results" after training.

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

The document briefly mentions that HepaFatSmart uses "one (1) convolutional neural network (CNN) performing liver ROI detection... Following the training of the Al assisted device, the system is completely 'locked down' for final validation..."
Given that HepaFatSmart is an "AI-assisted, automated version of HepaFat-Scan," it is highly probable that the ground truth for training its CNN for ROI detection was established by using data annotated or measured previously by the HepaFat-Scan methodology, likely involving the "Resonance Health's trained analyst" for ROI selection. However, the specifics of this process (e.g., how many cases, who annotated, adjudication) are not detailed in the provided text.

Summary

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

Image /page/0/Picture/0 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: the symbol of the Department of Health & Human Services on the left and the FDA acronym with the full name of the agency on the right. The FDA part is in blue, with the acronym in a solid blue square and the full name written in a smaller font next to it.

June 20, 2023

Resonance Health Analysis Services Pty Ltd % Mitchell Wells Managing Director 141 Burswood Road Burswood, Western Australia 6100 Australia

Re: K231459

Trade/Device Name: HepaFatSmart (V2.0.0) Regulation Number: 21 CFR 892.1000 Regulation Name: Magnetic resonance diagnostic device Regulatory Class: Class II Product Code: LNH Dated: May 19, 2023 Received: May 19, 2023

Dear Mitchell Wells:

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.

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

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 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 (QS) 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 medical devices and radiation-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.

Daniel M. Krainak, Ph.D. Assistant Director DHT8C: Division of Radiological Imaging and Radiation Therapy Devices OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

Submission Number (if known)

K231459

Device Name

HepaFatSmart (V2.0.0)

Indications for Use (Describe)

Intended use:

HepaFatSmart is intended for the quantitative measurement of volumetric liver fat fraction (VLFF), proton density fat fraction (PDFF) and steatosis grading.

Indications for use:

Support clinical diagnoses in individuals with confirmed or suspected fatty liver disease;

Support the subsequent clinical decision making processes for patients under management for fatty liver related disease or metabolic syndromes;

Aid in the assessment and screening of living donors for liver transplant.

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."

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

510(K) SUMMARY

This Summary has been prepared in accordance with 21 CFR 807.92.

General Information

Date Prepared	07 June 2023
Submitted by	Resonance Health Analysis Service Pty Ltd141 Burswood RdBurswood 6100AUSTRALIA
Main Contact	Mitchell WellsManaging Director,Resonance Health Analysis Services Pty Ltdmitchellw@resonancehealth.comTel: +61 8 9286 5300Fax: +61 8 9286 5399
US Contact (US Agent)	Michael van der WoudeDirector & GMEmergo Global Representation LLC2500 Bee Cave Road, Building 1, Suite 300Austin, TX 78746Phone: 512 3279997Fax : 512 3279998Email : USAgent@ul.com

Device Information

Name of Device	HepaFatSmart
Trade/proprietary Name	HepaFatSmart (V2.0.0)
Classification	Class II
Product Code	LNH
CFR Section	892.1000 Magnetic Resonance Diagnostic Device
Panel	Radiology

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

Description of the Device

The key components for the HepaFatSmart are:

. MRI Protocol: A specific MRI protocol for acquisition of the raw image data. The MRI protocol is critical to ensure the quality of the end results. Its adherence is verified by the HepaFatSmart IQC module, an automated algorithm that checks the correctness of each parameter in the protocol.
. HepaFatSmart: An image analysis software predicting a suitable liver region of interest (ROI) utilizing Al-assisted SaMD technology then performing the Alpha measurement and anomaly (excessive iron) detection. It is composed of one (1) convolutional neural network (CNN) performing liver ROI detection with undesired components (artefacts and major blood vessels) considered/removed using a computer vision technique with machine learning technology. Background noise correction is not considered as there is no or very minimal impact on the analysis outcome. Following the training of the Al assisted device, the system is completely 'locked down' for final validation prior to release in commercial use to ensure reproducibility of the results. In principle, the HepaFatSmart v2.0.0 uses the same MRI data analysis approach as HepaFat-Scan.
. Volumetric Liver Fat Fraction Measurement (VLFF): A software module (algorithmic) that incorporates a conversion lookup table relating Alpha to VLFF is added to allow production of a VLFF report.
. Proton Density Fat Fraction Measurement (PDFF): A software module (algorithmic) that incorporates a conversion lookup table relating VLFF to PDFF is added to allow production of a PDFF report.
. Steatosis Grade Measurement: A software module (algorithmic) that incorporates a conversion lookup table relating VLFF to a steatosis grade.
. Excessive liver iron assessment: An additional software module (algorithmic) that estimates the impact of liver iron content as per the inclusion criteria based on an algorithm to determine the analysis outcome (accept or reject).

The output of HepaFatSmart is the automatically generated reports in both PDF and DICOM (secondary captured) formats. Visually, the PDF and DICOM reports are identical except that the DICOM report also contains relevant header information. The HepaFatSmart report is populated with information stored in the DICOM header of the MRI images, the analysis result, where an Alpha value is converted into a VLFF value, a PDFF value, and a steatosis grade, and the associated confidence interval and normal range.

The HepaFatSmart report also contains pictures of two (2) echo times (TEs) (1st Out-of-Phase, 1st OP, and In-Phase, IP) of the analysed slice, a predicted liver ROI superimposed with one (IP) of the two TE images, and a fat distribution map. This is essential for the radiologist to check if the image

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

analysed is a liver image, the Al predicted ROI is placed correctly within the liver region, and the result provided is valid and consistent with other relevant clinical considerations.

HepaFatSmart SaMD can be accessed through a cloud-based or onsite platform. Resonance Health has developed its own cloud-based platform, called 'FAST'. Alternatively, HepaFatSmart can be offered on third parties' (channel partner) platforms.

It is important to note that HepaFatSmart as a medical device does not:

come into direct contact with patients or end-users;
. control any other device used on the patient;
. deliver any treatment or energy to the patient; or
. provide diagnostic information upon which inappropriate (or lack of) treatment likely to result in serious adverse events is based; as clinical judgment would be used in the patient's clinical management, based upon a range of other factors relating to the patient.

INTENDED USE

The intended use of HepaFatSmart is:

HepaFatSmart is intended for the quantitative measurement of volumetric liver fat fraction (VLFF), proton density fat fraction (PDFF) and steatosis grading.

INDICATIONS FOR USE

Support clinical diagnoses in individuals with confirmed or suspected fatty liver disease;
Support the subsequent clinical decision-making processes for patients under management for fatty liver related disease or metabolic syndromes;
Aid in the assessment and screening of living donors for liver transplant.
. Results, when interpreted by a trained physician can be used to support clinical diagnoses about the status of liver fat content, the subsequent clinical decision-making processes for the management of fatty liver related diseases, metabolic syndromes, liver donor screening and lifestyle change.

HepaFatSmart can be used to analyse the MRI images of patients of all population independent of age and gender, with suspected clinical conditions related to the level of liver fat.

PREDICATE INFORMATION

HepaFatSmart is substantially equivalent to the predicate device HepaFat-Scan (Resonance Health Analysis Services- K122035) and better than HepaFat-Al (Resonance Health Services - K201039).

SUBSTANTIAL EQUIVALENCE INFORMATION

The table below summarizes the main similarities and differences between HepaFat-Al and the predicate.

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

	HepaFatSmart	HepaFat-Al	HepaFat-Scan
Regulatory Class	ll	ll	ll
510(k) number	K231459	K201039	K122035
Classification Name	System, NuclearMagnetic ResonanceImaging, System, ImageProcessing Radiological	System, Nuclear MagneticResonance Imaging,System, Image ProcessingRadiological	System, Nuclear MagneticResonance Imaging,System, Image ProcessingRadiological
CFR Section	892.1000	892.1000	892.1000
Product Code andClassification Panel	LNH	LNH	LNH
Device Name	HepaFatSmart	HepaFat-Al	HepaFat-Scan
Trade/CommonName	HepaFatSmart	HepaFat-Al	HepaFat-Scan
Description	Standalone softwareplatform designed toautomatically analysewithin seconds magneticresonance imaging (MRI)datasets using themethod of HepaFat-Scanwith the liver ROIpredicted to generate anestimate of the patient'svolumetric liver fatfraction (VLFF),converted into protondensity fat fraction(PDFF) and steatosisgrade. No user input isrequired for the analysisthus minimising theimpact of human erroron obtained results.	Standalone softwareplatform designed toautomatically analysewithin seconds magneticresonance imaging (MRI)datasets to generate anestimate of the patient'svolumetric liver fat fraction(VLFF), converted intoproton density fat fraction(PDFF) and steatosis grade.No user input is requiredfor the analysis thusminimising the impact ofhuman error on obtainedresults.	Standalone softwareapplication to facilitate theimport and visualization ofmulti-slice, gradient-echoMRI data setsencompassing theabdomen, withfunctionality independentof the MRI equipment, toprovide objective andreproducibledetermination of thetriglyceride fat fraction inmagnetic resonanceimages of the liver. Itutilises magneticresonance images thatexploit the difference inresonance frequenciesbetween hydrogen nucleiin water and triglyceridefat. The quantitativetriglyceride fat fraction isbased on themeasurement of amagnetic resonanceparameter that reflectsthe ratio of the protondensity signal oftriglyceride fat to the totalproton density signal inthe liver.
	HepaFatSmart	HepaFat-Al	HepaFat-Scan
Technology	Convolutional neuralnetworks for theprediction of the liverregion of interest (ROI).Algorithmic for theimage quality checkingand calculated Alphaconversion into VLFF,PDFF and Steatosisgrade.Algorithms for themeasurement andcalculation of AlphaVLFF, PDFF and Steatosisgrade.	Convolutional neuralnetworks for the imageanalysis.Algorithmic for the imagequality checking and Alphaconversion into VLFF.	Algorithmic, with humaninteraction for Region ofInterest (ROI) selection.
Intended Use	HepaFatSmart isintended for thequantitativemeasurement ofvolumetric liver fatfraction (VLFF), protondensity fat fraction(PDFF) and steatosisgrading.HepaFatSmart is anapplication that is usedfor the non-invasiveevaluation of liver tissueby utilising magneticresonance images toevaluate the differencein resonance frequenciesbetween hydrogennuclei in water andtriglyceride fat. Thequantitative triglyceridefat fraction is based onthe measurement of amagnetic resonanceparameter that reflectsthe ratio of the protondensity signal oftriglyceride fat to thetotal proton densitysignal in the liver.	HepaFat-Al is intended forquantitative measurementof the triglyceride fatfraction in magneticresonance images of theliver, also known asvolumetric liver fat fraction(VLFF).It utilises magneticresonance images thatexploit the difference inresonance frequenciesbetween hydrogen nucleiin water and triglyceridefat. The quantitativetriglyceride fat fraction isbased on themeasurement of amagnetic resonanceparameter that reflects theratio of the proton densitysignal of triglyceride fat tothe total proton densitysignal in the liver.When interpreted by atrained physician, theresults provide informationthat can aid in diagnosis.	HepaFat-Scan is a softwaredevice intended forquantitative measurementof the triglyceride fatfraction in magneticresonance images of theliver. It utilises magneticresonance images thatexploit the difference inresonance frequenciesbetween hydrogen nucleiin water and triglyceridefat. The quantitativetriglyceride fat fraction isbased on themeasurement of amagnetic resonanceparameter that reflectsthe ratio of the protondensity signal oftriglyceride fat to the totalproton density signal inthe liver.When interpreted by atrained physician, theresults provideinformation that can aid indiagnosis.
Indications	Support clinical• diagnoses inindividuals with	HepaFat-Al is indicated to:• Assess the volumetricliver fat fraction	HepaFat-Scan is a softwaredevice intended forquantitative measurement
	HepaFatSmart	HepaFat-Al	HepaFat-Scan
	confirmed or suspected fatty liver disease;Support the subsequent clinical decision-making processes for patients under management for fatty liver related disease or metabolic syndromes; Aid in the assessment and screening of living donors for liver transplant. Results, when interpreted by a trained physician can be used to support clinical diagnoses about the status of liver fat content, the subsequent clinical decision-making processes for the management of fatty liver related diseases, metabolic syndromes, liver donor screening and lifestyle change. HepaFatSmart can be used to analyse the MRI images of patients of all population independent of age and gender, with suspected clinical conditions related to the level of liver fat.	proton density fat fraction and steatosis grade in individuals with confirmed or suspected fatty liver disease;Monitor liver fat content in patients undergoing weight loss management; Aid in the assessment and screening of living donors for liver transplant.	of the triglyceride fat fraction in magnetic resonance images of the liver. It utilises magnetic resonance images that exploit the difference in resonance frequencies between hydrogen nuclei in water and triglyceride fat. The quantitative triglyceride fat fraction is based on the measurement of a magnetic resonance parameter that reflects the ratio of the proton density signal of triglyceride fat to the total proton density signal in the liver.When interpreted by a trained physician, the results provide information that can aid in diagnosis.
User	Radiologist	Radiologist	Resonance Health's trained analyst
Hosting platform	Cloud-based or onsite platform	Cloud-based or onsite platform	Resonance Health's internal server
Image-type utilized	Magnetic Resonance	Magnetic Resonance	Magnetic Resonance
Image format	DICOM	DICOM	DICOM
Data Acquisition method	Gradient Recalled Echo (GRE)	Gradient Recalled Echo (GRE)	Gradient Recalled Echo (GRE)
	HepaFatSmart	HepaFat-Al	HepaFat-Scan
Anatomical Sites	Liver	Liver	Liver
Result reportcontent	Unique Report ID Patient ID, patient name, and date of birth for full identification of the patient. Scan date, and analysis date. Referrer and MRI centre. Results displayed: VLFF (%), PDFF (%) and Steatosis grade, associated with confidence intervals and normal range. Pictures of the 2 TEs of the analysed slice and analysis liver ROI placed on 1 TE. Liver colour map (for illustration purpose only, not for diagnostic)	Unique Report ID Patient ID, patient name, and date of birth for full identification of the patient. Scan date, and analysis date. Referrer and MRI centre. Results displayed: VLFF (%), PDFF (%) and Steatosis grade, associated with confidence intervals and normal range. Pictures of the 3 TEs of the analysed slice. Liver colour map (for illustration purpose only, not for diagnostic)	Unique Report ID Patient ID, patient name, and date of birth for full identification of the patient. Scan date, and analysis date. Referrer and MRI centre. Results displayed: VLFF (%) associated with confidence intervals and normal range. Picture of the analysed slice.
Result reportformat	PDF (encrypted) andsecondary capture(DICOM)	HTML and PDF	PDF

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

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

{9}------------------------------------------------

{10}------------------------------------------------

SUMMARY OF HEPAFATSMART PERFORMANCE - SUBSTANTIAL EQUIVALENCE

The device HepaFatSmart is in principle a new version of the predicate HepaFat-Al but with the name changed. As indicated in the Software Validation Report, three different technical analyses were performed to indicate the technical equivalency of HepaFatSmart compared with the reference standard HepaFat-Scan: Repeatability Study, Linear Regression Analysis and Bland Altman Analysis.

Repeatability Study (n = 42)

Repeatability study using a dataset with two different MRI scans for each subject to evaluate the device performance is a self-proven process as each paired data for a subject is supposed to produce the same analysis outcome. Briefly, the results are summarised as following:

Image /page/10/Figure/4 description: The image contains two scatter plots, labeled A and B, comparing two sets of measurements. Plot A compares 'HepaFat-Scan VLFF Scan 1 (%)' on the x-axis with 'HepaFat-Scan VLFF Scan 2 (%)' on the y-axis, showing a positive correlation. Plot B compares 'HepaFatSmart VLFF Scan 1 (%)' on the x-axis with 'HepaFatSmart VLFF Scan 2 (%)' on the y-axis, also showing a positive correlation. Both plots feature a diagonal line, indicating a perfect agreement between the two scans, and data points clustered closely around this line.

Figure 1. Plot A of HepaFat-Scan 2 against HepaFat-Scan VLFF measured at scan 1 for the 42 subjects in the repeatability study. Plot B is for HepaFatSmart (41 subjects). The solid line is the line of equivalence. Note, 41 instead of 42 subjects were used in the HepaFatSmart related analysis as a single case was identified as a high iron case with the newly introduced excessive iron assessment algorithm. All the results are closely scattered around the equivalency line, indicating good performance and substantial equivalence for both the predicate HepaFatSmart,

From the linear regression analysis shown in Figure 1 for both the reference standard HepaFat-Scan (plot A) and HepaFatSmart (plot B), all the results are closely scattered around the equivalency line and difficult to tell visually which one is better.

Image /page/10/Figure/7 description: The image contains two Bland-Altman plots comparing HepaFat-Scan and HepaFatSmart measurements. The left plot, labeled "A. HepaFat-Scan (reference standard)," displays the difference between two HepaFat-Scan measurements against their mean, with the y axis labeled as "HepaFat-Scan VLFF Scan 2 - HepaFat-Scan VLFF Scan 1" and the x axis labeled as "Mean HepaFat-Scan VLFF (%)". The right plot, labeled "B. HepaFatSmart," shows the difference between two HepaFatSmart measurements against their mean, with the y axis labeled as "HepaFatSmart VLFF Scan 2 - HepaFatSmart VLFF Scan 1" and the x axis labeled as "Mean HepaFatSmart VLFF (%)".

{11}------------------------------------------------

Figure 2. Bland Altman analysis of the repeatability study: plot A - HepaFat-Scan VLFF measured scan 1 and 2 for the 42 subjects; plot B is for HepaFatSmart (41 subjects). Bias and both repeatability coefficients are slightly better for HepaFatSmart compared with the reference standard HepaFat-Scan.

From the Bland Altman analysis shown in Figure 2 for both the reference standard HepaFat-Scan (plot A) and HepaFatSmart (plot B), bias and both repeatability coefficients for the HepaFatSmart are slightly better than those obtained from the repeated scans of HepaFat-Scan, indicating the performance of the HepaFatSmart is comparable (no worse) than human for the repeatability data analysed. This does not suggest yet that the HepaFatSmart is better than human analyst as the original human analysis (HepaFat-Scan) historically used two small liver ROIs rather than a single large liver ROI used in the HepaFatSmart with potentially slightly larger sampling error in the original HepaFat-Scan analysis. From the repeatability data, there was no statistically significant bias and tight repeatability coefficients for the HepaFatSmart, indicating the substantial equivalence to the predicate HepaFat-Scan and a possibility that the results from HepaFatSmart and HepaFat-Scan could be interchangeable.

In addition, HepaFatSmart demonstrated 100% repeatable (reproducible) in the repeatability study using the same datasets analysed twice with zero VLFF difference between the first and second analyses, which is better than human analysis with HepaFat-Scan.

Validation Study (n=300)

Validation study using a dataset with two different MRI scans for each subject to evaluate the device performance is a self-proven process as each paired data for a subject is supposed to produce the same analysis outcome. Briefly, the results are summarised as following (note the study population size is smaller than 300 as per the IQC module in place):

Image /page/11/Figure/5 description: The image contains two scatter plots, labeled A and B, comparing HepaFatSmart VLF and HepaFat-Scan VLFF. Plot A shows a broader range of values, extending up to 45% on both axes, while plot B focuses on a smaller range, up to 10%. Both plots show a positive correlation between the two methods, with data points clustered around a diagonal line, indicating agreement between HepaFatSmart and HepaFat-Scan VLFF measurements.

Figure 3. Linear regression analysis of the validation study (n = 281) by comparing HepaFatSmart with the reference standard HepaFat-Scan: plot A shows the full VLFF range and plot B shows the VLFF range between 0 – 10%. The solid line is the line of equivalence. 281 datasets out of 300 passed the IQC rules. All the results are closely scattered around the equivalency line, indicating good performance and substantial equivalence for both the predicate HepaFat-Scan and the device HepaFatSmart.

From the linear regression analysis shown in Figure 3 for both the reference standard HepaFat-Scan (plot A) and HepaFatSmart (plot B), all the results are closely scattered around the equivalency line and difficult to tell visually which one is better, indicating good performance and substantial equivalence for both the predicate HepaFat-Scan and the device HepaFatSmart.

From the Bland Altman analysis shown in Figure 4 for both the device HepaFatSmart (plot A) and another predicate HepaFat-Al (plot B) by comparing with the reference standard HepaFat-Scan, bias

{12}------------------------------------------------

and both 95% limits for the HepaFatSmart are significantly better than those obtained from the predicate HepaFat-Al, indicating the much improved and safer performance of the HepaFatSmart, which is as expected as the analysis outcomes are solely dependent on the liver ROI.. Despite the Bias of 0.2% is statistically significant (due to small variability), it is small and unlikely clinically significant. Both upper and lower 95% limits are small and close to the repeatability coefficients found in the repeatability study but significantly smaller than the predicate HepaFat-Al, indicating the new device HepaFatSmart is substantially equivalence to the predicate HepaFat-Scan and better than the predicate HepaFat-Al.

Image /page/12/Figure/1 description: The image contains two scatter plots, labeled A and B, comparing different methods of measuring Very Low-Density Lipoprotein Fraction (VLFF). Plot A compares 'HepaFatSmart VLFF' against 'HepaFat-Scan', while plot B compares 'HepaFat-AI VLFF' against 'HepaFat-Scan'. Both plots show the difference in VLFF measurements on the y-axis versus the mean VLFF on the x-axis, with horizontal lines indicating the mean difference and limits of agreement. In plot A, the mean difference is around 0.2, with limits of agreement at 1.7 and -1.3, while in plot B, the mean difference is around 0.4, with limits of agreement at 5.4 and -4.6.

Figure 4. Bland Altman analysis of the validation study (n = 281) of the HepaFatSmart VLFF (plot A) and HepaFat-Al (plot B) by comparing with the reference standard HepaFat-Scan VLFF. For the HepaFatSmart, the Bias of 0.2% is statistically significant (due to small variability) but small and unlikely significant. Both upper and lower 95% limits are small and close to the repeatability coefficients found in the repeatability study. Compared with the predicate HepaFat-Al, significant improvement is indicated by the small bias and 95% confident limits.

Table 1 summarises the Bland Altman analysis results performed in the repeatability study and validation study for the device HepaFatSmart and two predicates HepaFat-Scan and HepaFat-Al. In both studies, the performance of the device HepaFatSmart stands out and demonstrates the substantial equivalence to the predicate HepaFat-Scan and better outcomes than the predicate HepaFat-Al.

	Bias	95% Cl	Upper 95%LA/Repeatability	95% Cl	Lower 95%LA/Repeatability	95% Cl
HepaFat-ScanRepeatability	-0.2 (-0.19)	-0.5 to 0.1	1.9	1.3 to 2.5	-2.3	-1.7 to -2.9
HepaFat-Al v1.2.17Repeatability	-0.2 (-0.22)	-0.8 to 0.3	3.2	2.3 to 4.2	-3.6	-2.7 to -4.6
HepaFatSmart v2.0.0Repeatability	-0.1 (-0.14)	-0.4 to 0.1	1.5	1.1 to 2.0	-1.8	-1.4 to -2.4
HepaFat-Al v1.2.17 Limitsof Agreement (fullquarantine dataset)	0.4 (0.41)	0.1 to 0.7	5.4	4.9 to 6.0	-4.6	- 4.1 to -5.2
HepaFatSmart v2.0.0Limits of Agreement (fullquarantine dataset)	0.2 (0.19)	0.1 to 0.3	1.7	1.5 to 1.8	-1.3	-1.1 to -1.4

Table 1. Upper and lower 95% limits of repeatability and upper and lower 95% limits of agreement between HepaFat-Al v1.2.17, HepaFatSmart and HepaFat-Scan VLFF measurements.

{13}------------------------------------------------

The fully quarantined 300 validation subjects with different clinical conditions across a broad age range and fat level scanned from different MRI centres with different MRI makes and models were used to validate the whole functionalities including IQC and to evaluate the technical and clinical performances. The sensitivities and specificities of HepaFatSmart for predicting HepaFat-Scan VLFF values in the fully quarantined dataset (n=281, successfully passed the IQC rules) above clinically relevant VLFF thresholds are given in Table 2 below, which demonstrate excellent performances and substantial equivalency of the new device HepaFatSmart compared with the reference standard and predicate HepaFat-Scan with both sensitivity and specificity well above 90% and close to 100%. A few very limited subjects with miscategorised outcomes were identified and discussed with minimal or non-clinical relevant impact. No adverse effects and complications have been identified for any of the miscategorised subjects.

VLFF threshold	Clinical relevance	Sensitivity(95% CI) (%)	Specificity(95% CI) (%)
4.1 %	• Boundary between grade 0 (<5%) and grade1 (5 - 33%) steatosis by histologicalinspection. Used to define the absence (0) orpresence (1) of NAFLD.	100.0 (97.3 –100.0)	98.6 (94.9 –99.6)
12.1 %	• Boundary between grade 1 (5-33%) andgrade 2 (33-66%) steatosis by histologicalinspection.	98.8 (93.6 –99.8)	98.0 (94.9 -99.2)
16.2 %	• Boundary between grade 2 (33-66%) andgrade 3 (> 66%) steatosis by histologicalinspection.	100.0 (93.8 -100.0)	99.6 (97.5 -99.9)

Table 2. Sensitivities and specificities of HepaFatSmart for predicting HepaFat-Scan VLFF values greater than three clinically relevant thresholds.

As shown in both technical and clinical tests from above, the new device HepaFatSmart produces almost the same analysis results as the reference standard and predicate HepaFat-Scan with the liver ROI approach, which is as expected because the analysis outcomes are purely governed by the liver ROI. In the meantime, both technical and clinical tests have demonstrated significant improvement compared with the predicate HepaFat-Al with safer and more effective outcomes.

CONCLUSION

The Special 510(k) submission for HepaFatSmart contains adequate information and data to enable the FDA-CDRH to determine substantial equivalence to the predicate devices. Resonance Health Analysis Services Pty Ltd believes that enough evidence has been presented in this submission to conclude that HepaFatSmart is safe, effective and performs as well as the predicate HepaFat-Scan and safer, more effective, and better than the predicate HepaFat-Al.

Regulation Number and Section

§ 892.1000 Magnetic resonance diagnostic device.

(a)
Identification. A magnetic resonance diagnostic device is intended for general diagnostic use to present images which reflect the spatial distribution and/or magnetic resonance spectra which reflect frequency and distribution of nuclei exhibiting nuclear magnetic resonance. Other physical parameters derived from the images and/or spectra may also be produced. The device includes hydrogen-1 (proton) imaging, sodium-23 imaging, hydrogen-1 spectroscopy, phosphorus-31 spectroscopy, and chemical shift imaging (preserving simultaneous frequency and spatial information).(b)
Classification. Class II (special controls). A magnetic resonance imaging disposable kit intended for use with a magnetic resonance diagnostic device only is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.