The Bay Labs, Inc. EchoMD Automated Ejection Fraction software is used to process previously acquired transthoracic cardiac ultrasound images, to store images, and to manipulate and make measurements on images using a personal computer or a compatible DICOM-compliant PACS system in order to provide automated estimation of left ventricular ejection. This measurement can be used to assist the clinician in a cardiac evaluation.

The EchoMD Automated Ejection Fraction Software is indicated for use in adult patients.

Device Description

The EchoMD Automated Ejection (AutoEF) software applies machine learning algorithms to process echocardiography images in order to calculate left ventricular ejection fraction. The software operates in between the DICOM source and the DICOM destination. EchoMD AutoEF performs left ventricular ejection measurements using both the apical four chamber and apical two chamber cardiac ultrasound views.

The software selects the image clips to be used, performs the AutoEF calculation, and forwards the results to a destination PACS server for clinician viewing. The output of the Ejection Fraction estimate stated as a percentage, which is displayed via the destination PACS system. The software applies machine learning algorithms to assess image quality and provides this information as qualitative and quantitative user feedback. By automating the estimation of ejection fraction, the EchoMD software is designed to streamline the clinician's calculation of the measurement.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study details for the Bay Labs, Inc. EchoMD Automated Ejection Fraction Software, based on the provided text:

Acceptance Criteria and Reported Device Performance

Acceptance Criteria	Reported Device Performance
Root Mean Square Deviation (RMSD) below a set threshold as compared to the reference ground truth EF.	8.290% RMSD

1. Sample Size and Data Provenance

Test Set Sample Size: Over 300 previously-acquired studies.
Data Provenance: Retrospective, non-interventional validation study. The country of origin is not explicitly stated, but given the FDA submission, it's highly likely to be within the US or compliant with US standards.

2. Number of Experts and Qualifications

Number of Experts: Not explicitly stated for the primary validation study's ground truth establishment. However, an additional study was performed "with a different set of cardiologists" on a subset of the validation patient studies. The specific number of cardiologists for this additional study is also not provided.
Qualifications of Experts: The experts involved in the "additional study" were "cardiologists." No further details on their experience (e.g., years of experience) are given.

3. Adjudication Method for the Test Set

The text does not describe an explicit adjudication method (e.g., 2+1, 3+1) for establishing the ground truth EF for the test set. The ground truth was established by "the biplane method of disks ejection fraction reported." It implies that the reported EF values were considered the reference.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

Was an MRMC study done? The text states, "An additional study was performed with a different set of cardiologists on a subset of the validation patient studies to further demonstrate the generalizability of the software." This suggests a form of multi-reader study, but it's not explicitly framed as an MRMC comparative effectiveness study in the sense of comparing human readers with AI vs. without AI assistance. The focus seems to be on verifying the generalizability of the software's performance against different cardiologists.
Effect Size: No effect size or improvement for human readers with AI assistance versus without AI assistance is reported. The study's purpose was to demonstrate the generalizability of the software's performance, not to quantify human reader improvement.

5. Standalone (Algorithm Only) Performance

Was a standalone performance study done? Yes. The primary endpoint measured the EchoMD AutoEF ejection fraction measurements compared to the biplane method ejection fraction, with a root mean square deviation calculated. This directly assesses the algorithm's performance independent of a human-in-the-loop scenario.
Performance Metric: Root Mean Square Deviation (RMSD).
Performance Result: 8.290% RMSD (p-value 0.00052).

6. Type of Ground Truth Used

Ground Truth: "the biplane method of disks ejection fraction reported." This is a clinical standard for Left Ventricular Ejection Fraction (LVEF) measurement based on echocardiography. While it relies on expert interpretation and measurement, it is not explicitly described as "expert consensus" in the sense of multiple independent readers reaching a consensus. It's more aligned with a established clinical measurement method.

7. Sample Size for the Training Set

The sample size for the training set is not provided in the document. The text only states that "Test datasets were strictly segregated from algorithm training datasets."

8. How Ground Truth for Training Set was Established

The method for establishing the ground truth for the training set is not specified. The document only mentions that the software "applies machine learning algorithms to process echocardiography images."

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: the Department of Health and Human Services seal on the left and the FDA acronym and name on the right. The FDA acronym is in a blue box, and the full name is in blue text. The text reads "FDA U.S. FOOD & DRUG ADMINISTRATION".

June 14, 2018

Bay Labs, Inc. % Yarmela Pavlovic Partner Hogan Lovells US LLP 3 Embarcadero Center, Suite 1500 SAN FRANCISCO CA 94111

Re: K173780

Trade/Device Name: EchoMD Automated Ejection Fraction Software Regulation Number: 21 CFR 892.2050 Regulation Name: Picture archiving and communications system Regulatory Class: II Product Code: LLZ Dated: May 22, 2018 Received: May 22, 2018

Dear Yarmela Pavlovic:

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. 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 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); good manufacturing practice requirements as set forth in the quality systems (OS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

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

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 http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm.

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/DeviceRegulationandGuidance/) and CDRH Learn (http://www.fda.gov/Training/CDRHLearn). 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 (http://www.fda.gov/DICE) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Michael D. O'Hara For

Robert Ochs, Ph.D. Director Division of Radiological Health Office of In Vitro Diagnostics and Radiological Health Center for Devices and Radiological Health

Enclosure

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

DEPARTMENT OF HEALTH AND HUMAN SERVICES
Food and Drug Administration

Indications for Use

Form Approved: OMB No. 0910-0120 Expiration Date: 06/30/2020 See PRA Statement below

510(k) Number (if known)

K173780

Device Name

Bay Labs, Inc. EchoMD Automated Ejection Fraction Software

Indications for Use (Describe)

The EchoMD Automated Ejection Fraction Software is indicated for use in adult patients.

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 NEEDEDThis 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

Bay Labs, Inc. EchoMD Automated Ejection Fraction Software

Submitter's Name, Address, Telephone Number, Contact Person

Bay Labs, Inc. 290 King Street, Suite 9 San Francisco, CA 94107

Contact Person: Charles Cadieu, CEO Phone: 415 671 4711 email: charles@baylabs.io

Date Prepared: May 21, 2018

Name of Device

Common or Usual Name: Picture Archival and Communications Systems Workstation

Proprietary Name: EchoMD Automated Ejection Fraction Software

Classification Name: 21 CFR § 892.2050

Regulatory Class: II

Product Code: LLZ, System, Image processing, Radiological

Predicate Device

Diacardio, Ltd. LVivo EF Software (K130779)

Intended Use / Indications for Use

The Bay Labs, Inc. EchoMD Automated Ejection Fraction software is used to process previously acquired transthoracic cardiac ultrasound images, and to manipulate and make measurements on images using a personal computer or a compatible DICOM-compliant PACS system in order to provide automated estimation of left ventricular ejection. This measurement can be used to assist the clinician in a cardiac evaluation.

The EchoMD Automated Ejection Fraction Software Application is indicated for use in adult patients.

The intended use/indications for use statement is worded slightly differently than that of the predicate device, but the minor differences do not alter the intended diagnostic effect of the subject device; both products are intended to process echocardiogram images and produce ejection fraction calculations. The minor differences also do not raise different questions of safety or effectiveness, as in both instances the key question is whether the software can both produce ejection fraction calculations acceptable for clinical use.

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

Device Description

Summary of Technological Characteristics

The technological principle underlying both the EchoMD AutoEF and the predicate device is the calculation of ejection fraction (EF) on previously acquired cardiac scans using machine learning-based algorithms and biplane apical echocardiographic images. Ultimately, the key question for both products is whether the EF calculation provides adequate information for the clinician.

While the two products use slightly different approaches to achieving their shared purpose, these differences do not raise different questions of safety or effectiveness. EchoMD AutoEF operates as a microservice background function communicating with ultrasound systems and PACS workstation systems, which provide the user interface. The predicate device may operate in this manner but also is configurable to provide its own user interface. EchoMD AutoEF automates the selection of image video clips on which to operate. The predicate device includes manual user selection of image video clips. EchoMD AutoEF utilizes a convolutional network design which operates on the image data in order to estimate ejection fraction without creating endocardial traces. The predicate generates endocardial traces. Other minor differences arise due to the design of the EchoMD running as part of a PACS system, rather than as a fully stand-alone application like the predicate.

A tabular comparison of the key features of the subject and predicate devices is provided at the end of this Summary.

Performance Data

EchoMD EF software was developed and tested in accordance with Bay Labs' Design Control processes and has been subjected to extensive safety and performance testing. Non-clinical verification and validation test results established that the device meets its design requirements and intended use. Specifically, software verification was conducted at unit, module, and system integration levels. Risk management analysis generated multiple risk mitigation measures and verification activities. A Cybersecurity Analysis and data security testing were conducted to verify that data and patient protected health information security measures are included in the software.

Verification and validation testing was conducted to demonstrate the substantial equivalence of EchoMD AutoEF. In both tests, the success criterion was that the EchoMD AutoEF would produce an

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

ejection fraction number with a Root Mean Square Deviation below a set threshold as compared to the reference ground truth EF.

A formal retrospective, non-interventional validation study was conducted using over 300 previously-acquired studies where the biplane method of disks ejection fraction was reported. This patient dataset was constructed to provide a balanced range of gender, ejection fraction values, and body mass index levels. Testing included a wide array of ultrasound system manufacturers to verify that EchoMD AutoEF performs acceptably across multiple scanner platforms. Variability testing was also performed to demonstrate that EchoMD AutoEF performs acceptably with a variety of image clips and frames from the same patient. Test datasets were strictly segregated from algorithm training datasets. EchoMD AutoEF ejection fraction measurements were compared to the biplane method ejection fraction, and a root mean square deviation was calculated and used as the primary endpoint. The primary endpoint was met (results: 8.290% RMSD, p-value 0.00052). An additional study was performed with a different set of cardiologists on a subset of the validation patient studies to further demonstrate the generalizability of the software. Based on the clinical performance as documented in this pivotal clinical study, the device has a safety and effectiveness profile that is similar to the predicate Diacardio LVivo EF Software device.

Conclusion

The EchoMD AutoEF Software is as safe and effective as the Diacardio LVivo EF Software. The subject device has the same intended uses and similar indications, technological characteristics, and principles of operation as its predicate device. The minor differences in indications do not alter the intended diagnostic use of the device and do not affect its safety and effectiveness when used as labeled. In addition, the minor technological differences between the EchoMD AutoEF and its predicate device do not raise different questions of safety or effectiveness. Performance data demonstrate that the subject device is as safe and effective as the identified predicate. Thus, the EchoMD Automated Ejection Fraction Software is substantially equivalent.

	Bay Labs EchoMD Automated EjectionFraction Software Application	DiaCardio, Ltd, LVivo EF SoftwareApplication (K130779, Predicate)
Product Code	LLZ	LLZ
Intended Use	The Bay Labs, Inc. EchoMD AutomatedEjection Fraction software is used toprocess previously acquired transthoraciccardiac ultrasound images, to store images,and to manipulate and makemeasurements on images using a personalcomputer or a compatible DICOM-compliant PACS system in order to provideautomated estimation of left ventricularejection fraction. This measurement can beused to assist the clinician in a cardiacevaluation.	DiaCardio's L Vivo EF SoftwareApplication is intended for non-invasiveprocessing of already acquiredechocardiographic images in order todetect, measure, and calculate the leftventricular wall for left ventricularfunction evaluation. This measurementcan be used to assist the clinician in acardiac evaluation.
Machine Learning-Based Algorithm	Yes	Yes

Comparison Table: EchoMD AutoEF versus Predicate

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

	Bay Labs EchoMD Automated EjectionFraction Software Application	DiaCardio, Ltd, LVivo EF SoftwareApplication (K130779, Predicate)
Operates on DICOMclips	Yes	Yes
Automation level	Fully automated, including clip selection	Fully automated (with manually-selected clips)
EF Method	Biplane (non-segmentation/non-endocardial trace)	Biplane Method of DisksSegmentation (endocardial trace)
Offline EF evaluationusing clips from multipleultrasound scanners	Yes	Yes
Image Clip Selection	Automated	Manual
Automated EjectionFraction Calculation	Yes	Yes
Ejection Fractionreported	Whole number estimate (percentage)	Whole number estimate(percentage)
Algorithm Confidence	Qualitative and quantitative userfeedback on transthoracic cardiacultrasound image quality	Display of endocardial border tracing
EF Result shown withvideo clip	Yes	Yes
User confirmation/rejection of result	Yes	Yes
Manual editing ofautomated result byuser	Yes (on PACS workstation)	Yes (in application)
Left ventricular volumesmeasurements	Not internal. Uses manual trace optionPACS workstation	Yes

Regulation Number and Section

§ 892.2050 Medical image management and processing system.

(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).