EchoSolv AS by Echo IQ Ltd | FDA 510(k) Summary

EchoSolv AS is a machine learning (ML) and artificial intelligence (AI) based decision support software indicated for use as an adjunct to echocardiography for assessment of severe aortic stenosis (AS).

When utilized by an interpreting physician, this device provides information to facilitate rendering an accurate diagnosis of AS. Patient management decisions should not be made solely on the results of the EchoSolv AS analysis.

EchoSolv AS includes both the algorithm based AS phenotype analysis, and the application of recognized AS clinical practice quidelines.

Limitations: EchoSolv AS is not intended for patients under the age of 18 years or those who have previously undergone aortic valve replacement surgery

Device Description

EchoSolv AS is a standalone, cloud-based decision support software which is intended to be used certified cardiologist to aid in the diagnosis of Severe Aortic Stenosis. EchoSolv AS analyzes basic patient demographic data and measurements obtained from a transthoracic echo examination to provide a categorical assessment as to whether the data are suggestive of a high, medium or low probability of Severe AS. EchoSolv AS is intended for patients who 18 years or older who have an echocardiogram performed as part of routine clinical care (i.e., for the evaluation of structural heart disease).

Patient demographic and echo measurement data is automatically processed through the artificial intelligence algorithm which provides an output regarding the probability of a Severe AS phenotype to aid in the clinical diagnosis of Severe AS during the review of the patient echo study and generation of the final study report, according to current clinical practice guidelines. The software provides an output on the following assessments:

Severe AS Phenotype Probability

Whether the patient has a high, medium, or low probability of exhibiting a Severe AS phenotype, based on analysis by the EchoSolv AS proprietary Al algorithm, that the determined predicted AVA is ≤1.0cm². The Al probability score requires a minimum set of data inputs to provide a valid output but is based on all available echocardiographic measurement data and does not rely on the traditional LVOT measurements used to in the continuity equation.

Severe AS Guideline Assessment

Whether the patient meets the definition for Severe AS based on direct evaluation of provided echocardiogram data measurements (AV Peak Velocity, AV Mean Gradient and AV Area) with current clinical practice guidelines (2020 ACC/AHA Guideline for the Management of Patients with Valvular Heart Disease).

EchoSolv AS is intended to be used by board-certified cardiologists who review echocardiograms during the evaluation and diagnosis of structural heart disease, namely aortic stenosis. EchoSolv AS is intended to be used in conjunction with current clinical practices and workflows to improve the identification of Severe AS cases.

AI/ML Overview

Here's an analysis of the acceptance criteria and study detailed in the provided document for the EchoSolv AS device:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state "acceptance criteria" in a tabulated format. However, based on the performance data presented, the implicit acceptance criteria can be inferred from the reported performance and comparison to a predicate device. The performance metrics reported are AUROC, Sensitivity, Specificity, Diagnostic Likelihood Ratios (DLR), and improvement in reader AUROC and concordance in the MRMC study.

Performance Metric	Implicit Acceptance Criterion (Based on context/predicate)	Reported Device Performance (EchoSolv AS)
Standalone Performance
AUROC (Overall)	Expected to be high, comparable to or better than predicate (Predicate: 0.927 AUROC)	0.948 (95% CI: 0.943-0.952)
Sensitivity (at high probability)	High (No specific threshold given, but expected to detect a good proportion of true positive cases)	0.801 (95% CI: 0.786-0.818)
Specificity (at high probability)	High (No specific threshold given, but expected to correctly identify true negative cases)	0.923 (95% CI: 0.915-0.932)
DLR (Low Probability)	Low (Indicative of low probability of disease)	0.067 (95% CI: 0.057-0.080)
DLR (Medium Probability)	Close to 1 (Weakly indicative)	0.935 (95% CI: 0.829-1.05)
DLR (High Probability)	High (Strongly indicative of disease)	10.3 (95% CI: 9.22-11.50)
Cochran-Armitage Trend Test (p-value)	Statistically significant trend (p < 0.05)	<0.0001 (Statistic: 41.362)
AUROC across subgroups (Age, Sex, Race, LVEF, BMI, Inputs)	Consistent high performance across demographics and input completeness	Consistently high, ranging from 0.914 to 0.970, demonstrating consistency. Lowest for LVEF <30% but still strong.
Clinical Performance (MRMC Study)
Mean AUROC (assisted vs. unassisted)	Improvement expected with AI assistance (Predicate showed 0.054 improvement)	Improvement of 0.018 (95% CI: 0.037-0.001; p=0.064)
Reader Concordance (assisted vs. unassisted)	Improvement expected with AI assistance	Improvement of 0.027 (unassisted: 0.641; assisted: 0.667)

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

Standalone Performance Test Set:
- Sample Size: 6,268 studies.
- Data Provenance: The document states the dataset for model development was randomly split into training and test sets. The standalone performance testing was performed on an "independent retrospective cohort study" meaning the data was collected from past records. The country of origin for this specific retrospective cohort is not explicitly stated, but it's implied to be within a clinical setting that would allow for US board-certified cardiologists to review and verify the data.
Clinical Performance (MRMC) Test Set:
- Sample Size: 200 retrospective transthoracic echocardiogram (TTE) studies (100 disease cases, 100 control studies).
- Data Provenance: Retrospective, performed at "one investigational site in the US."

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

Standalone Performance Test Set:
- Number of Experts: More than one, implied by "US board certified cardiologists" (plural).
- Qualifications: "US board certified cardiologists." No specific years of experience are mentioned.
Clinical Performance (MRMC) Test Set:
- Number of Experts: Two.
- Qualifications: "board certified cardiologists." No specific years of experience are mentioned.

4. Adjudication Method for the Test Set

Standalone Performance Test Set: The ground truth was established by "US board certified cardiologists, who reviewed and verified the echocardio data and hemodynamic profile... and were blinded to the device output." This implies a consensus or majority rule adjudication among the "cardiologists" if multiple were involved per case. There is no explicit "2+1" or "3+1" method mentioned, but rather a verification process.
Clinical Performance (MRMC) Test Set: "The total test dataset was reviewed by two board certified cardiologists to confirm the presence and severity of Severe AS." This suggests a 2-reader agreement for the ground truth. If there was disagreement, an implied adjudication process (e.g., a third reader or consensus discussion) would be needed, but it is not specified.

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 MRMC comparative effectiveness study was done.
Effect Size of Improvement:
- Mean AUROC Improvement: 0.018 (95% CI: 0.037-0.001; p=0.064).
- Reader Concordance Improvement: 0.027 (from 0.641 unassisted to 0.667 assisted).

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, a standalone performance testing was performed. The objective was to assess the "native system performance of the EchoSolv AS model in its ability to detect Severe AS."

7. The Type of Ground Truth Used

Standalone Performance Test Set: Expert consensus based on "the assessment of the presence of severe aortic stenosis (defined as an AVA≤1 cm²) by US board certified cardiologists, who reviewed and verified the echocardio data and hemodynamic profile." This combines clinical assessment using a specific echocardiographic measurement (AVA) and expert review.
Clinical Performance (MRMC) Test Set: Expert consensus. "The total test dataset was reviewed by two board certified cardiologists to confirm the presence and severity of Severe AS."

8. The Sample Size for the Training Set

442,276 individuals (from a total dataset of 631,824 individuals).

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

The document states: "The EchoSolv AS Al Model was developed on a dataset consisting of 631,824 individuals with 1,077,145 transthoracic echocardiograms (TTE)." It also notes that the model was trained "to detect severe AS cases." While implied that the training set also used ground truth related to severe AS detection, the specific methodology for establishing ground truth for the training set is not explicitly detailed in the provided text. It is reasonable to infer it would be similar to the test set ground truth (expert clinical assessment based on echocardiographic data), but this is not directly stated.

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 Department of Health & Human Services logo on the left and the FDA acronym along with the full name of the agency on the right. The Department of Health & Human Services logo features a stylized human figure, while the FDA part includes the acronym in a blue square and the words "U.S. FOOD & DRUG ADMINISTRATION" in blue text.

Echo IQ Ltd % J. David Giese CEO & Co-Founder Innolitics LLC 1101 West 34th St. #550 Austin, Texas 78705

October 4, 2024

Re: K241245

Trade/Device Name: EchoSolv AS Regulation Number: 21 CFR 892.2060 Regulation Name: Radiological computer-assisted diagnostic software for lesions suspicious of cancer Regulatory Class: Class II Product Code: POK Dated: May 3, 2024 Received: September 6, 2024

Dear J. David Giese:

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 (the 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 available 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.

Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device"

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

(https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).

Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30, Design controls; 21 CFR 820.90, Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the QS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).

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 Part 803) for devices or postmarketing safety reporting (21 CFR Part 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 Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.

All medical devices, including Class I and unclassified devices and combination product device constituent parts are required to be in compliance with the final Unique Device Identification System rule ("UDI Rule"). The UDI Rule requires, among other things, that a device bear a unique device identifier (UDI) on its label and package (21 CFR 801.20(a)) unless an exception or alternative applies (21 CFR 801.20(b)) and that the dates on the device label be formatted in accordance with 21 CFR 801.18. The UDI Rule (21 CFR 830.300(a) and 830.320(b)) also requires that certain information be submitted to the Global Unique Device Identification Database (GUDID) (21 CFR Part 830 Subpart E). For additional information on these requirements, please see the UDI System webpage at https://www.fda.gov/medical-device-advicecomprehensive-regulatory-assistance/unique-device-identification-system.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 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-regulatory

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

assistance/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,

Jessica Lamb

Jessica Lamb, Ph.D. Assistant Director Imaging Software Team DHT8B: Division of Radiological Imaging Devices and Electronic Products OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

Submission Number (if known)

K241245

Device Name

EchoSolv AS

Indications for Use (Describe)

EchoSolv AS includes both the algorithm based AS phenotype analysis, and the application of recognized AS clinical practice quidelines.

Limitations: EchoSolv AS is not intended for patients under the age of 18 years or those who have previously undergone aortic valve replacement surgery

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/2 description: The image shows the logo for Echo IQ. The word "Echo" is written in a dark teal color, while the "IQ" is written in a golden yellow color. The font is sans-serif and the letters are bolded. The logo is simple and modern.

1. SUBMITTER

Company	Echo IQ Ltd
Address	Suite 2.114, Level 1, 477 Pitt StreetSydney NSW 2000Australia
Phone	+61 (02) 9159 3719
Contact Person	Dane Brescacin
Date Prepared	October 2, 2024

2. SUBJECT DEVICE

Device Name	EchoSolv AS
Classification Name	Radiological Computer-Assisted Diagnostic Software (CADx) forLesions Suspicious for Cancer
Regulation	21 CFR 892.2060
Regulatory Class	Class II
Product Code	POK

3. PREDICATE DEVICE

Device Name	EchoGo Pro
Manufacturer	Ultromics Limited
510(k) Number	K201555

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

Image /page/5/Picture/1 description: The image shows the logo for Echo IQ. The word "Echo" is in a dark teal color, while the "IQ" is in a golden yellow color. The font is sans-serif and modern.

DEVICE DESCRIPTION 4.

1. Severe AS Phenotype Probability

2. Severe AS Guideline Assessment

The EchoSolv AS Al Model was developed on a dataset consisting of 631,824 individuals with 1,077,145 transthoracic echocardiograms (TTE). The dataset was randomly split (ratio 70:30 based on individuals) into two separate groups, training and test set. Data from 442,276 individuals (70%) were entered into the Al model to train the device to detect severe AS cases. The remaining 189,548 individuals (30%) were reserved for internal testing. Individual patients appeared only once in either the training or test dataset but not both.

INDICATIONS FOR USE 5.

EchoSolv AS is a machine learning (ML) and artificial intelligence (AI)-based decision support software indicated for use as an adjunct to echocardiography for assessment of severe aortic stenosis (AS).

EchoSolv AS includes both the algorithm based AS phenotype analysis, and the application of recognized Aortic Stenosis clinical practice guidelines.

Limitations: EchoSolv AS is not intended for patients under the age of 18 years or those who have previously undergone aortic valve replacement surgery.

6. SUBSTANTIAL EQUIVALENCE

COMPARISON OF INTENDED USE 6.1

This section summarizes the similarities and differences between EchoSolv AS and the predicated device in relation to intended use and indications for use, and a rationale on why the differences raise no new concerns for its safety and performance.

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

Image /page/6/Picture/1 description: The image shows the logo for Echo IQ. The word "Echo" is in a dark teal color, while the "IQ" is in a golden yellow color. The font is sans-serif and modern.

Both devices have the same intended use. Both devices are decision support software which are intended to, based on echocardiogram data, indicate whether there is evidence of a cardiovascular disease as an aid in diagnosis.

The subject device has a similar indication for use to the predicate devices are machine learning decision support software designed as adjuncts to echocardiography, aiding interpreting physicians in diagnosing specific cardiac conditions. The subject device is indicated for severe aortic stenosis (AS), while the predicate device is indicated for coronary artery disease (CAD). Neither device is meant for primary diagnosis, interpreting physicians retain responsibility for accurate diagnosis and patient management. EchoSolv AS is indicated for use in patients over 18 years old, while the predicate device does not specify age limitations within its intended patient population.

Any differences between the indications for use do not raise any new concerns with regards to safety and effectiveness. Both the subject and predicate device are intended to provide a categorical assessment as to whether the data is suggestive of a probability/possibility of the respective cardiac conditions and aid in the diagnosis.

COMPARISON OF TECHNICAL CHARACTERISTICS 6.2

This section summarizes the similarities and differences between EchoSolv AS and the predicated device in relation to technical characteristics, and a rationale on why the differences raise no new concerns for its safety and performance.

Form Factor and Algorithm Type: Both the subject and predicate device are Software as a Medical Device that incorporate a machine learning and artificial intelligence algorithm for their respective clinical decision support functionalities.

Device Input Modality: Both devices receive input from ultrasound (echocardiography) modalities. The subject device uses a resting transthoracic echocardiogram, the predicate device uses a stress transthoracic echocardiogram. Variations in the type of echocardiogram used are specific for the cardiac condition specified in the respective indications for use.

Device Output: Both devices provide a report with output analysis statements formatted in a highly similar manner. For both devices, the output is based on a machine learning and artificial intelligence algorithm which produces a categorical assessment of whether the data is suggestive of a probability/possibility of the intended cardiac disease to support the interpreting physician in the respective clinical workflow. These outputs are returned to the interpreting physician for review and to determine their applicability for use.

Software Integration: Both devices are standalone software applications that use the same method of integration (third-party systems) which allows for input datasets to be sent to their respective standalone software applications. In addition to the data ingest, the subject device allows users to upload csv files for assessment (retrospective analysis only) and also uses the same third-party integration systems to integrate outputs back into end user PACS or reporting software systems. However, this difference does not present any new issues related to safety and effectiveness as in output integration does not impact the clinical functionality or the intended use of the device.

Device Input: Both devices use echocardiogram data as a form of input; the subject device uses additional basic patient demographic inputs (age and body surface area (height and weight)). The subject input are measurements taken by a sonographer from an echocardiogram for algorithm analysis. The predicate device provides software functionalities to compute measurements semi-automatically from the echocardiographic image for subsequent analysis by the algorithm. In both devices, the measurements are derived from a similar image modality (i.e., echocardiography), and all measurements are subject to review and approval by a trained clinical user prior to the algorithm. The difference in how device input is acquired does not present new issues related to safety and effectiveness as both devices provide a categorical assessment as to whether the data indicated a possibility/probability to aid in the diagnosis of each device's respective cardiac conditions.

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

PERFORMANCE DATA 7.

SOFTWARE VERIFICATION AND VALIDATION TESTING 7.1

Software verification and validation activities were performed and documented in accordance with the FDA Guidance "Content of Premarket Submissions for Device Software Functions". Based on this guidance, EchoSolv AS was assessed to represent a "Basic Documentation Level", since a failure or latent flaw of the device software function(s) would not present a hazardous situation with a probable risk of death or serious injury to either a patient, user of the device or others in the environment of use, prior to the implementation of risk control measures.

7.2 STANDALONE PERFORMANCE TESTING

Standalone performance testing was performed in accordance with 21 CFR §892.2060 special control 1/iv). The objective of the standalone performance testing was to assess the native system performance of the EchoSolv AS model in its ability to detect Severe AS.

Standalone performance of EchoSolv AS was performed and evaluated on an independent retrospective cohort study. Overall, 6,268 studies were included in the analysis (mean age 74.95±13.98 years, inclusive of 3,172 men (50.61%) and 3,095 women (49.38%), who were predominately Caucasian (78.61%) followed by African American (5.47%), Hispanic (2.49%) and Asian (1.82%) and Other (11.60%). One study did not have gender or ethnicity disclosed.

The reference standard was established using the assessment of the presence of severe aortic stenosis (defined as an AVA≤1 cm²) by US board certified cardiologists, who reviewed and verified the echocardio data and hemodynamic profile of the study cohort, and were blinded to the device output. Of the 6,262 studies, 2,483 (39.6%) had an AVA ≤1.0cm² and 3,779 (60.4%) had an AVA >1.0cm². A verification analysis of the reference standard was performed on an additional cohort, with results provided in the device labelling.

The primary endpoint of the study, a standalone Receiver Operating Characteristic (ROC) curve was generated. Area under the receiver operating characteristic (AUROC) curve were computed via the trapezoidal approximation with 95% confidence intervals (Cls). Diagnostic likelihood ratios (DLR) were calculated for each device output (low, medium and high), with 95% Cls. Cochrane-Armitage Trend Test for a trend in probability across the device outputs. AUROC with 95% Cis were generated for the following subgroups: age, sex, ethnicity, LVEF and BM.

The EchoSolv AS model achieved a native system performance of 0.948 (95% CJ: 0.943-0.952) AUROC. Sensitivity and specificity at the high probability threshold were 0.801(95%C): 0.786-0.818) and 0.923 (95%Cl: 0.915-0.932), respectively At the low, medium and high probability outputs, the DLR were 0.067 (95% Cl: 0.057-0.080), 0.935 (95%C): 0.829-1.05) and 10.3 (95%Cl: 9.22-11.50), respectively. The Cochran-Armitage Trend Test statistic of 41.362 p: <0.0001.

Subgroups	Variations	N	AUROC (95% CI)
Age	18-65 years	1295	0.954 (0.941 - 0.966)
	≥65 years	4973	0.942 (0.936 - 0.948)
Sex	Male	3172	0.945 (0.937 - 0.952)
	Female	3095	0.954 (0.947 - 0.961)
Race	White	4927	0.949 (0.943 - 0.954)
	Black	343	0.953 (0.924 - 0.977)
	Asian	114	0.970 (0.938 - 0.992)
	Hispanic	156	0.965 (0.937 - 0.986)
	Other	727	0.921 (0.901 - 0.937)
LVEF	<30%	421	0.914 (0.883 - 0.941)
	>30 to <50%	939	0.939 (0.925 - 0.953)

4887

1866

EchoSolv AS performed consistently across all subgroups, refer to the results in the table below.

50%

18-25 kg/m²

BMI

0.950 (0.945 - 0.956)

0.952 (0.943 - 0.961)

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

	>25 to 30 kg/m²	2075	0.951 (0.942 - 0.959)
	>30 to 35 kg/m²	1121	0.936 (0.922 - 0.949)
	>35 kg/m²	888	0.947 (0.932 - 0.960)
Inputs	Minimum inputs	6,268	0.931 (0.925 - 0.937)
	All available inputs	6,268	0.948 (0.942 - 0.952)

The predicate device achieved a native system performance of 0.927 AUROC. Based on the standalone performance testing, the EchoSolv AS model achieved a greater native system performance than that of the predicate device.

CLINICAL PERFORMANCE TESTING 7.3

Clinical performance testing was performed in accordance with 21 CFR $892.2060 special control 1(ii) and 1(ii). The objective of the performance testing was to evaluate the diagnostic performance of readers when interpreting TTE studies, with and without the assistance of EchoSolv AS. Clinical performance testing was performed at one investigational site in the US, to ensure the test data was independent from the training dataset.

Clinical performance of EchoSolv AS was evaluated in a fully-crossed, multi-reader multi case (MRMC) study. The study evaluated the performance of five readers (board-certified cardiologists) in their ability to identify severe AS in a dataset of 200 retrospective transthoracic echocardiogram (TTE) studies. The total test dataset was reviewed by two board certified cardiologists to confirm the presence and severity of Severe AS. The MRMC dataset consisted of 100 disease cases (confirmed Severe AS) and 100 control studies (confirmed no Severe AS). The dataset was inclusive of 101 women (50.5%) and 99 men (49.5%) with a mean age of 73.55±12.5 years, who were predominately Caucasian (86.5%), followed by African American (10.5%) and Hispanic or Latino (0.5%) did not have ethnicity disclosed.

The primary endpoint of the study, ROC curves were generated and compared between paired with and without EchoSolv AS. All AUROCs were computed via the trapezoidal approximation with 95% Cls. AUROC for unassisted and assisted reads were 0.865 (95%C): 0.837-0.893) and 0.883 (95%C1: 0.857-0.909), respectively. When cardiologist readers were provided with EchoSolv AS to assist with their interpretation of a TTE, there was an improvement in all study endpoints: mean AUROC (0.018±0.010, 95%C!: 0.037-0.001; p=0.064). The predicate device showed a mean improvement of 0.054; both devices showed an improvement in reader AUROC and accuracy when assisted. Reader concordance (agreement) was evaluated using Fleiss' Kappa for unassisted and assisted reads were 0.641 (95%CI: 0.597-0.685) and 0.667 (95%CI: 0.623-0.711), respectively. When assisted with EchoSolv AS, there was an improvement in reader concordance of 0.027.

The results of the standalone and reader performance testing demonstrated that the EchoSolv AS device meets established specifications necessary for consistent performance to achieve its intended use and confirmed that the technological difference do not raise any new questions of safety and effectiveness.

CONCLUSION OF SUBSTANTIAL EQUILAVENCE DETERMINATION 8.

Non-clinical and clinical performance data demonstrates that the EchoSolv AS device is substantially equivalent to the predicate device Echo Go Pro (K201555). EchoSolv AS has the same intended use, and similar indications for use and technological characteristics as its predicate device. Any differences identified in the indications for use and technical characteristics do not impact the intended use of the device and does not raise any new questions relating to its safety and effectiveness when used as intended.

Therefore, it can be concluded that EchoSolv AS is substantially equivalent to the predicate device, EchoGo Pro.

Regulation Number and Section

§ 892.2060 Radiological computer-assisted diagnostic software for lesions suspicious of cancer.

(a)
Identification. A radiological computer-assisted diagnostic software for lesions suspicious of cancer is an image processing prescription device intended to aid in the characterization of lesions as suspicious for cancer identified on acquired medical images such as magnetic resonance, mammography, radiography, or computed tomography. The device characterizes lesions based on features or information extracted from the images and provides information about the lesion(s) to the user. Diagnostic and patient management decisions are made by the clinical user.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of the image analysis algorithms including, but not limited to, a detailed description of the algorithm inputs and outputs, each major component or block, and algorithm limitations.
(ii) A detailed description of pre-specified performance testing protocols and dataset(s) used to assess whether the device will improve reader performance as intended.
(iii) Results from performance testing protocols that demonstrate that the device improves reader performance in the intended use population when used in accordance with the instructions for use. The performance assessment must be based on appropriate diagnostic accuracy measures (
e.g., receiver operator characteristic plot, sensitivity, specificity, predictive value, and diagnostic likelihood ratio). The test dataset must contain sufficient numbers of cases from important cohorts (e.g., subsets defined by clinically relevant confounders, effect modifiers, concomitant diseases, and subsets defined by image acquisition characteristics) such that the performance estimates and confidence intervals of the device for these individual subsets can be characterized for the intended use population and imaging equipment.(iv) Standalone performance testing protocols and results of the device.
(v) Appropriate software documentation (
e.g., device hazard analysis; software requirements specification document; software design specification document; traceability analysis; and description of verification and validation activities including system level test protocol, pass/fail criteria, results, and cybersecurity).(2) Labeling must include:
(i) A detailed description of the patient population for which the device is indicated for use.
(ii) A detailed description of the intended reading protocol.
(iii) A detailed description of the intended user and recommended user training.
(iv) A detailed description of the device inputs and outputs.
(v) A detailed description of compatible imaging hardware and imaging protocols.
(vi) Warnings, precautions, and limitations, including situations in which the device may fail or may not operate at its expected performance level (
e.g., poor image quality or for certain subpopulations), as applicable.(vii) Detailed instructions for use.
(viii) A detailed summary of the performance testing, including: Test methods, dataset characteristics, results, and a summary of sub-analyses on case distributions stratified by relevant confounders (
e.g., lesion and organ characteristics, disease stages, and imaging equipment).