The Eko Murmur Analysis Software (EMAS) is intended to provide decision support to clinicians in their evaluation of patients' heart sounds. The software analyzes heart sounds and phonocardiograms (and ECG signals, when available). The software will automatically detect murmurs that may be present, and the murmur timing and character, including S1, S2, innocent heart murmurs, structural heart murmurs, and the absence of a heart murmur.

The Eko Murmur Analysis Software is not intended as a sole means of diagnosis and is for use in environments where health care is provided by clinicians. The interpretations of heart sounds offered by the software are meant only to provide decision support to the clinician, who may use the result in conjunction with their own evaluation and clinical judgment. The interpretations are not diagnoses. The Eko Murmur Analysis Software is intended for use on pediatric and adult patients.

Device Description

Eko Murmur Analysis Software (EMAS) is a cloud-based service that allows users to upload heart sound/phonocardiogram (PCG) and optional electrocardiogram (ECG) data via an application programming interface (API) for analysis. The software uses signal processing (such as waveform filtering), as well as algorithms derived from machine learning, to analyze the acquired data and generate clinical decision support output for clinicians. EMAS is designed to evaluate data derived by the company's two previously cleared devices, the Eko DUO (K170874) and Eko CORE (K151319, K200776). The heart sound data from those devices can be transmitted to the Eko Cloud using either the Eko mobile application or thirdparty applications that use a software development kit (SDK). The EMAS algorithm analyzes the heart sound data and outputs a JSON file with the algorithm results, which is passed down to the requesting application and displayed by the requesting application to the user in the humanreadable format.

The analysis will assess the signal quality of the phonocardiogram; detect heart murmurs and classify them as innocent or structural; determine the timing of S1 and S2 heart sounds; and distinguish between systolic and diastolic heart murmurs. As an integral part of a physical assessment, clinicians' interpretations of EMAS' output can help them rule in or out different pathological conditions in a patient.

The EMAS consists of the following algorithm components:

Signal Quality Detection Algorithm:
This pre-processing algorithm accepts as input the PCG sound from the API controller (e.g., a mobile smartphone application). The algorithm is used to classify PCG recordings based on their signal quality as good or poor.
Heart Sound Timing Algorithm:
This algorithm detects the presence and timing of specific heart sounds including S1, S2, the systole region, and the diastole region.
Murmur Detection & Classification Algorithm: This algorithm is used to identify and classify heart sounds as having "No Murmur", an "Innocent Murmur" (i.e., not pathologic), or a "Structural Murmur" (i.e., pathologic).
Murmur Timing Algorithm:

This algorithm is used to identify in which regions of the heart cycle (systole vs diastole) a heart murmur occurs if either an "Innocent Murmur" or "Structural Murmur" is identified by the Murmur Detection and Classification Algorithm.

AI/ML Overview

Here's an analysis of the Eko Murmur Analysis Software (EMAS) acceptance criteria and the study proving its performance, based on the provided FDA 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

Performance Metric	Acceptance Criteria	Reported Device Performance (EMAS)
Murmur Classification	Lower bound of 95% CI for Sensitivity > 75.0% (compared to primary predicate's lower bound of 72.9%)	Sensitivity: 85.6% (95% CI: 82.6 - 88.7)
	Lower bound of 95% CI for Specificity > 75.0% (compared to primary predicate's lower bound of 74.9%)	Specificity: 84.4% (95% CI: 81.3 - 87.5)
S1 Detection	Not explicitly stated as a separate acceptance criterion with a numerical threshold, but expected to demonstrate substantially equivalent performance to predicates.	Sensitivity: 96.2% (95% CI: 94.9 - 97.4)
		PPV: 97.1% (95% CI: 96.3 - 98.0)
S2 Detection	Not explicitly stated as a separate acceptance criterion with a numerical threshold, but expected to demonstrate substantially equivalent performance to predicates.	Sensitivity: 92.3% (95% CI: 90.3 - 94.3)
		PPV: 94.3% (95% CI: 93.4 - 95.1)

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

Test Set Size: The document does not explicitly state a separate "test set" size. However, it indicates that the clinical validation used a database of 2,380 unique heart sound recordings from 615 unique subjects.
- Of these, "recordings identified as being good signal by the expert cardiologists" (meaning suitable for analysis) included:
  - 45.8% (approx. 1090 recordings) with a confirmed structural murmur.
  - 54.2% (approx. 1290 recordings) with confirmed no murmur or innocent murmur.
- For heart sound timing, 299 heart sound recordings were annotated.
Data Provenance: Retrospective analysis on a proprietary database. The country of origin is not specified, but the applicant (Eko Devices, Inc.) is based in Oakland, California, USA.

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

Number of Experts: "Multiple cardiologists" were used. The exact number is not specified.
Qualifications of Experts: "Cardiologists." No further details on their years of experience or specific subspecialties are provided.

4. Adjudication Method for the Test Set

Recordings were "annotated by multiple cardiologists."
There's no explicit mention of an adjudication method like 2+1 or 3+1. However, the ground truth for murmur classification was obtained via "pairing cardiologist annotations with gold standard echocardiogram," suggesting that the echocardiogram served as the definitive ground truth reference alongside expert opinion.

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

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not reported. The study focuses on the standalone performance of the EMAS algorithm against a ground truth. There is no information provided about human readers improving with or without AI assistance.

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

Yes, a standalone study was done. The reported performance metrics (Sensitivity, Specificity, PPV) are directly attributed to the "EMAS algorithm testing" and represent the algorithm's performance against the established ground truth. The device is intended as "decision support" and "not intended as a sole means of diagnosis," indicating it operates standalone and then informs a clinician.

7. The Type of Ground Truth Used

For Murmur Classification: Ground truth was established by pairing cardiologist annotations with gold standard echocardiogram.
For S1/S2 Timing: Ground truth was established via expert cardiologist annotations.

8. The Sample Size for the Training Set

The document explicitly states: "No study subjects included in the training datasets were included in the test database." However, it does not provide the sample size for the training set. It only mentions that the algorithms were validated using "retrospective analysis on a proprietary database."

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

The document does not explicitly describe how the ground truth for the training set was established. It only refers to a "proprietary database" used for training and then tested on a separate, distinct set of subjects. Assuming a consistent approach, it's likely similar methods (expert annotations, potentially with echocardiogram correlation) were used, but this is not stated.

Summary

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

Image /page/0/Picture/0 description: The image shows the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: a symbol on the left and the FDA name on the right. The symbol on the left is a stylized image of a human figure, while the FDA name on the right is written in blue letters. The words "U.S. FOOD & DRUG ADMINISTRATION" are written in a clear, sans-serif font.

June 29, 2022

Eko Devices, Inc. Sam Huang Director of Regulatory Affairs 1212 Broadway, Suite 100 Oakland, California 94612

Re: K213794

Trade/Device Name: Eko Murmur Analysis Software (EMAS) Regulation Number: 21 CFR 870.1875 Regulation Name: Stethoscope Regulatory Class: Class II Product Code: DQD, DQC, DPS Dated: Mav 28, 2022 Received: June 2, 2022

Dear Sam Huang:

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,

LCDR Stephen Browning Assistant Director Division of Cardiac Electrophysiology, Diagnostics and Monitoring Devices Office of Cardiovascular Devices Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K213794

Device Name

Eko Murmur Analysis Software (EMAS)

Indications for Use (Describe)

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}------------------------------------------------

GENERAL INFORMATION

Applicant: Eko Devices, Inc. 1210 Broadway, Suite 100, Oakland, CA 94612 Phone: 844-356-3384

Contact Person: Sam Huang, Ph.D. Director of Regulatory Affairs Eko Devices, Inc. 1210 Broadway, Suite 100, Oakland, CA 94612

Date Prepared: Dec 03, 2021

DEVICE INFORMATION

Trade/Proprietary Name: Eko Murmur Analysis Software (EMAS) Regulation number: 21 CFR 870.1875 Classification Name: Electronic Stethoscope Regulatory Class: Class II Product Code: DOD, DOC, DPS

PREDICATE DEVICES

CSD Labs GmbH, eMurmur ID (K181988) Eko Devices, Inc., Eko Analysis Software (K192004)

DEVICE DESCRIPTION

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

The EMAS consists of the following algorithm components:

Signal Quality Detection Algorithm: .
This pre-processing algorithm accepts as input the PCG sound from the API controller (e.g., a mobile smartphone application). The algorithm is used to classify PCG recordings based on their signal quality as good or poor.
Heart Sound Timing Algorithm: ● This algorithm detects the presence and timing of specific heart sounds including S1, S2, the systole region, and the diastole region.
Murmur Detection & Classification Algorithm: This algorithm is used to identify and classify heart sounds as having "No Murmur", an "Innocent Murmur" (i.e., not pathologic), or a "Structural Murmur" (i.e., pathologic).
Murmur Timing Algorithm:

INDICATIONS FOR USE

The Eko Murmur Analysis Software is intended to provide decision support to clinicians in their evaluation of patients' heart sounds. The software analyzes heart sounds and phonocardiograms (and ECG signals, when available). The software will automatically detect murmurs that may be present, and the murmur timing and character, including S1, S2, innocent heart murmurs, structural heart murmurs, and the absence of a heart murmur.

SUBSTANTIAL EQUIVALENCE

Eko Murmur Analysis Software (EMAS) has the same intended use as the predicate devices for providing support to the physician in the evaluation of heart-related physiological data. Additionally, EMAS combines the indications of the predicate devices into a single product: algorithms to classify suspected murmurs as either innocent or structural, similar to that of the eMurmur ID; and algorithms that analyze heart sound data to identify the presence of suspected murmurs, similar to that of the Eko Analysis Software. Any differences in the Eko Murmur

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

Analysis Software's indication statement. i.e., omission of certain outputs and compatible device information, do not alter the intended diagnostic effect of the device.

Eko Murmur Analysis Software also has similar technological characteristics as the predicate devices in regard to it being a software-only analysis tool that analyzes and classifies heart sound data. EMAS's new technological characteristics, i.e., proprietary analysis algorithms, could affect its safety or effectiveness, but do not raise any different questions of safety or effectiveness, because they analyze the same data (input) to determine the same parameters (output) as those of the predicate devices. The performance testing data shown in this application demonstrate that the Eko Murmur Analysis Software is as safe and as effective as the predicate device(s).

A table comparing the key features of the subject and predicate devices is provided below.

Parameters	Eko Murmur AnalysisSoftware	eMurmur ID(Primary Predicate)	Eko AnalysisSoftware(Secondary Predicate)
FDA Number	K213794	K181988	K192004
Indications forUse	The Eko Murmur AnalysisSoftware is intended toprovide decision supportto clinicians in theirevaluation of patients'heart sounds. The softwareanalyzes heart sounds andphonocardiograms (andECG signals, whenavailable). The softwarewill automatically detectmurmurs that may bepresent, and the murmurtiming and character,including S1, S2, innocentheart murmurs, structuralheart murmurs, and theabsence of a heartmurmur.The Eko Murmur AnalysisSoftware is not intendedas a sole means ofdiagnosis and is for use in	The eMurmur ID softwaresystem is a decisionsupport device for thehealthcare provider (theuser) in the evaluation ofpatient heart sounds.eMurmur ID is used torecord, display, analyze,and store the acousticsignal of the heart,recorded by means of anelectronic stethoscope. Theautomated analysis willidentify specific heartsounds that may be present,including S1, S2,physiological heartmurmurs, pathologicalheart murmurs and absenceof a heart murmur.eMurmur ID is indicatedfor use in a setting whereauscultation would	The Eko AnalysisSoftware is intended toprovide support to thephysician in theevaluation of patients'heart sounds andECG's. The softwareanalyzes simultaneousECG and heart sounds.The software willdetect the presence ofsuspected murmurs inthe heart sounds. Thesoftware also detectsthe presence of atrialfibrillation and normalsinus rhythm from theECG signal. Inaddition, it calculatescertain cardiac timeintervals such as heartrate, ORS duration andEMAT. The softwaredoes not distinguish
Parameters	Eko Murmur AnalysisSoftware	eMurmur ID(Primary Predicate)	Eko AnalysisSoftware(Secondary Predicate)
environments wherehealth care is provided byclinicians. Theinterpretations of heartsounds offered by thesoftware are meant only toprovide decision supportto the clinician, who mayuse the result inconjunction with theirown evaluation andclinical judgment. Theinterpretations are notdiagnoses. The EkoMurmur AnalysisSoftware is intended foruse on pediatric and adultpatients.	typically be performed by ahealthcare provider. It isnot intended as a solemeans of diagnosis. Theheart sound interpretationsoffered by eMurmur ID areonly significant whenconsidered in conjunctionwith healthcare providerover-read and including allother relevant patient data.	between different kindsof murmurs and doesnot identify otherarrhythmias.It is not intended as asole means ofdiagnosis. Theinterpretations of heartsounds and ECGoffered by the softwareare only significantwhen used inconjunction withphysician over-read andis for use on adults (>18 years).
PatientPopulation	Adult and pediatricpatients	Adult and pediatric patients	Adult patients
DeviceClassification	DQD, DQC, DPS	DQD, DQC	MWI, DQD, DPS
Prescribed	Prescription Only	Prescription Only	Prescription Only
Components	Software Only	Software Only	Software Only
Interface	Callable applicationprogramming interface(API)	N/A (Self-containedsoftware application)	Callable applicationprogramming interface(API)
Display	No primary display	No primary display	No primary display
MurmurDetection	Yes	Yes	Yes
MurmurClassification	Yes	Yes	No

Table 1 Substantial Equivalence Summary Comparison

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

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

PERFORMANCE DATA - NONCLINICAL TESTING SUMMARY

The Eko Murmur Analysis Software was the subject of software verification and validation testing, consistent with the principles outlined in FDA's General Principles of Software Validation: Final Guidance for Industry and FDA Staff issued May 11, 2005.

PERFORMANCE DATA - CLINICAL TESTING SUMMARY

The algorithms in this submission have been validated using retrospective analysis on a proprietary database. The database contains 2.380 unique heart sound recordings from 615 unique subjects. In the proprietary database, all heart sound recordings were collected using the Eko CORE (67.6%) or Eko DUO (32.4%) digital stethoscopes. The subject population was 42.6% female and had an age range of 0 to 95 years old, race was recorded in 45% of the subjects and out of that subset, 98% were white, 1.5% were African American and 0.7% were Asian. No complications or adverse events were experienced during the use of the EMAS device.

All recordings were annotated by multiple cardiologists in respect to their quality and the presence of any murmur. Of the recordings identified as being good signal by the expert cardiologists, 45.8% had a confirmed structural murmur and 54,2% had a confirmed no murmur or innocent murmur. Ground truth for murmur classification was obtained via pairing cardiologist annotations with gold standard echocardiogram. Additionally, 299 heart sound recordings were annotated by expert cardiologists to obtain the timing of the S1 and S2 sounds. Cardiologists were blinded to all subject demographic data and echocardiogram findings during annotation.

No study subjects included in the training datasets were included in the test database. All algorithm testing was performed once the final EMAS version was locked.

The following tables demonstrate the results of the primary performance analyses (Murmur Classification and Heart Sound Timing).

Table 2 EMAS Murmur Classification Performance

	Sensitivity (%)	Specificity (%)
Murmur Classification	85.6(95% CI: 82.6 - 88.7)	84.4(95% CI: 81.3 - 87.5)

The lower bounds of the 95% confidence intervals of the primary predicate device are 72.9% and 74.9% for sensitivity and specificity, respectively. Through clinical validation, the lower bound of EMAS' calculated 95% confidence interval is found to be above 75.0%, thus demonstrating substantially equivalent performance (effectiveness) to the predicate device (eMurmur).

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

		Sensitivity (%)	PPV (%)
Heart SoundTiming	S1 Detection	96.2(95% CI: 94.9 - 97.4)	97.1(95% CI: 96.3 - 98.0)
	S2 Detection	92.3(95% CI: 90.3 - 94.3)	94.3(95% CI: 93.4 - 95.1)

Table 3 EMAS Heart Sound Timing Performance

Given the data described above, the EMAS algorithm testing shows substantially equivalent performance to the predicate devices where applicable. Based on the clinical performance, the Eko Murmur Analysis Software has a safety and effectiveness profile that is similar to the predicate devices.

CONCLUSIONS

The Eko Murmur Analysis Software is as safe and as effective as the predicate devices. The Eko Murmur Analysis Software has the same intended uses and similar indications, technological characteristics, and principles of operation as its primary predicate device. The minor differences in indications do not alter the intended diagnostic use of the device and do not raise different questions of safety and effectiveness when used as labeled. In addition, the minor technological differences between the Eko Murmur Analysis Software and its predicate devices raise no new issues of safety or effectiveness. Performance data demonstrate that the Eko Murmur Analysis Software is as safe and effective as the predicate devices. Thus, the Eko Murmur Analysis Software is substantially equivalent.

Regulation Number and Section

§ 870.1875 Stethoscope.

(a)
Manual stethoscope —(1)Identification. A manual stethoscope is a mechanical device used to project the sounds associated with the heart, arteries, and veins and other internal organs.(2)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 870.9.(b)
Electronic stethoscope —(1)Identification. An electronic stethoscope is an electrically amplified device used to project the sounds associated with the heart, arteries, and veins and other internal organs.(2)
Classification. Class II (special controls). The device, when it is a lung sound monitor, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 870.9.