(119 days)
Yes
The device description explicitly mentions "Al Platform 2.0" and describes "Al-assisted tools" for image analysis, quantification, and quality scoring. The document also refers to "Al modules" and "Algorithm and API" as components.
No
The device is intended for noninvasive processing of ultrasound images to detect, measure, and calculate medical parameters and provide quality score feedback; it does not directly treat or diagnose a disease.
Yes
Explanation: The "Intended Use/Indications for Use" states that the device is intended for "noninvasive processing of ultrasound images to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease." This directly indicates a diagnostic purpose. Additionally, the "Device Description" mentions that it "allows users to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease" and provides tools for "suggesting the presence of lung structures and artifacts" and "quantification of Left Ventricular Ejection Fraction (LVEF), Myocardium wall thickness... and IVC diameter," all of which are diagnostic activities.
Yes
The device description explicitly states that it is a "software as a medical device (SaMD)" and that the AI modules are provided as a "software component to be integrated by another computer programmer into their legally marketed ultrasound imaging device." This indicates that the device itself is solely software.
Based on the provided information, this device is not an IVD (In Vitro Diagnostic).
Here's why:
- IVD Definition: In Vitro Diagnostics are medical devices used to perform tests on samples taken from the human body, such as blood, urine, or tissue, to detect diseases, conditions, or infections. The tests are performed outside the body (in vitro).
- Device Function: The described device, Exo AI Platform 2.0, processes ultrasound images acquired from the patient's body. It analyzes these images to detect structures, measure parameters, and provide quality feedback. This is an in vivo process (within the living body), not an in vitro process.
- Input: The input is DICOM formatted ultrasound images, not biological samples.
- Output: The output is analysis and measurements derived from the images, not results from laboratory tests on biological samples.
Therefore, while it is a medical device that aids in diagnosis and assessment, it does not fit the definition of an In Vitro Diagnostic. It is a software as a medical device (SaMD) that assists healthcare professionals in interpreting medical imaging.
Yes
The letter explicitly states, "FDA's substantial equivalence determination also included the review and clearance of your Predetermined Change Control Plan (PCCP)." This language directly indicates that the PCCP for this specific device has been cleared by the FDA.
Intended Use / Indications for Use
AI Platform 2.0 is intended for noninvasive processing of ultrasound images to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease. In addition, it can provide Quality Score feedback to assist healthcare professionals, trained and qualified to conduct echocardiography and lung ultrasound scans in the current standard of care while acquiring ultrasound images. The device is intended to be used on images of adult patients.
Product codes
QIH
Device Description
Exo AI Platform 2.0 (AIP 2.0) is a software as a medical device (SaMD) that helps qualified users with image-based assessment of ultrasound examinations in adult patients. It is designed to simplify workflow by helping trained healthcare providers evaluate, quantify, and generate reports for ultrasound images. AIP 2.0 takes as an input in the Digital Imaging and Communications in Medicine (DICOM) format from ultrasound scanners of a specific range and allows users to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease. In addition, it provides frame and clip quality score in real-time for the Left Ventricle from the four-chamber apical and parasternal long axis views of the heart and lung scans. In addition, the AI modules are provided as a software component to be integrated by another computer programmer into their legally marketed ultrasound imaging device. Essentially, the Algorithm and API, which are modules, are medical device accessories.
Key features of the software are
Lung AI: An AI-assisted tool for suggesting the presence of lung structures and artifacts on ultrasound images, namely A-lines. Additionally, a per-frame and per-clip quality score is generated for each lung scan.
Cardiac AI: An AI-assisted tool for the quantification of Left Ventricular Ejection Fraction (LVEF), Myocardium wall thickness (Interventricular Septum (IVSd), Posterior wall (PWd)), and IVC diameter on cardiac ultrasound images. Additionally, a per-frame and per-clip quality score is generated for each Apical and PLAX cardiac scan.
Mentions image processing
Yes
Mentions AI, DNN, or ML
Yes
Input Imaging Modality
ultrasound images
Anatomical Site
Heart, Lungs
Indicated Patient Age Range
adult patients
Intended User / Care Setting
healthcare professionals, trained and qualified to conduct echocardiography and lung ultrasound scans
Description of the training set, sample size, data source, and annotation protocol
Not Found
Description of the test set, sample size, data source, and annotation protocol
The clinical performance of the AI Platform 2.0 was successfully evaluated on test data encompassing diverse demographic variables, including gender, age, and ethnicity from multiple sites in metropolitan cities with diverse racial patient populations. The test data was entirely separated from the training/tuning datasets and was not used for any part of the training/tuning. We also established auditability measures, by assigning a unique identification number to each study and its corresponding images.
The left ventricle Wall thickness AI and IVC AI were each evaluated on scans of 100 subjects, on images acquired from cart-based and portable ultrasound devices. The ground truth for all measurements were obtained as the average measurement of three experts.
The validation of Quality AI consists of validating on:
a. Data previously acquired from 184 patients from various ultrasound devices and various cardiac pathologies, compared to quality rating by experienced sonographers on each frame and the entire clip. Total of 226 clips (29,732 frames) were used in this test.
b. Data acquired after using the image and clip Quality AI in real time while scanning Lung, Apical 4 chamber, and Parasternal Long Axis views of the heart. In total, 396 lung and cardiac scans were done by 26 users with a wide range of ultrasound experience in POC settings, including 18 novice users who received two hours of training.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Validation Performance Testing
The clinical performance of the AI Platform 2.0 was successfully evaluated on test data encompassing diverse demographic variables, including gender, age, and ethnicity from multiple sites in metropolitan cities with diverse racial patient populations. The test data was entirely separated from the training/tuning datasets and was not used for any part of the training/tuning. We also established auditability measures, by assigning a unique identification number to each study and its corresponding images.
The left ventricle Wall thickness AI and IVC AI were each evaluated on scans of 100 subjects, on images acquired from cart-based and portable ultrasound devices. The ground truth for all measurements were obtained as the average measurement of three experts. Performance was assessed by calculating Inter class correlation (ICC) between the AI and the ground truth.
Summary of AI Platform performance for left ventricle wall thickness and IVC measurement:
LV Wall thickness: InterVentricular Septum (IVSd), ICC (95% CI): 0.93 (0.89 – 0.96)
LV Wall thickness: Posterior Wall (PWd), ICC (95% CI): 0.94 (0.89 – 0.97)
Inferior vena cava (IVC): IVC Dmin, ICC (95% CI): 0.93 (0.90 – 0.95)
Inferior vena cava (IVC): IVC Dmax, ICC (95% CI): 0.94 (0.90 – 0.96)
The validation of Quality AI consists of validating on:
a. Data previously acquired from 184 patients from various ultrasound devices and various cardiac pathologies, compared to quality rating by experienced sonographers on each frame and the entire clip. Total of 226 clips (29,732 frames) were used in this test. The overall agreement between the Quality AI and quality rated by the experienced sonographers was ICC = 0.94 (95% Cl .94 - . 95) for frames and ICC = 0.94 (95% Cl .92 - . 95) for clips.
b. Data acquired after using the image and clip Quality AI in real time while scanning Lung, Apical 4 chamber, and Parasternal Long Axis views of the heart. In total, 396 lung and cardiac scans were done by 26 users with a wide range of ultrasound experience in POC settings, including 18 novice users who received two hours of training. 98.3% of the clips rated as ACEP quality of 3 or above by expert readers, also received at least "Minimum criteria met for diagnosis" image quality by Clip Quality AI. Additionally, 98.0% of scans that were considered as "Minimal criteria met for diagnosis" or "good" by Quality AI were also deemed diagnostic by experts (ACEP score of 3 or higher).
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
ICC, agreement percentage
Predicate Device(s)
Reference Device(s)
Predetermined Change Control Plan (PCCP) - All Relevant Information
A Predetermined Change Control Plan (PCCP) is included with AI Platform 2.0. The PCCP specifies anticipated software modifications, implementation methods and validation criteria that will be used to implement the software modifications in a controlled manner and ensure the subject device remains as safe and effective as the predicate.
The PCCP includes a specific list of anticipated software modifications which are summarized in the table below, as well as a Modification protocol describing the verification and validation activities that will support these planned modifications. The modification protocol incorporates impact assessment considerations and specifies requirements for data management, including data sources, collection, storage, and sequestration, as well as documentation and data segregation/re-use practices.
Specific test methods are specified in the PCCP to establish substantial equivalence relative to AI Platform 2.0, and include sample size determination, and acceptance criteria. Original and additional validation data will be analyzed to ensure the datasets are representative of the intended use population. To ensure validation test datasets are representative of the intended use population, each will meet minimum demographic requirements.
Summary of software modifications, test methods, necessary performance requirements, and impact assessment:
Modification: (A) Modification to architecture, pre/post processing
Rationale: The AI models in AI Platform 2.0 may be modified in a focused and bound manner, to improve accuracy, efficiency, and adaptability while maintaining the safety and efficacy of the device.
Testing Methods: Substantial equivalence in performance matrices as compared to the AI Platform 2.0. Non-inferiority margins in alignment with FDA recommendations will be used to ensure the new device remains safe and effective without any such change compared with the original device and the last modified version of the device.
Impact Assessment: Improved accuracy and/or efficiency metrics for AI model performance. Benefit-Risk Analysis: Benefit: Enhanced performance; Improved efficiency and scalability. Risk: Reduction in clinical performance. Risk Mitigation: The risks of overfitting and generalizability are mitigated by conducting cross-validation, data augmentation, and hyper parameter tuning in internal training procedures. The final validation is performed using non-inferiority analysis according to the established clinical protocol to ensure baseline performance is maintained and improvements are realized before modifications are committed and released.
Modification: (B) Introduction of new training data
Rationale: The training dataset of the AI models may be augmented with a broader and more diverse range of imaging data to enhance model robustness, reduce bias, improve generalizability, and respond effectively to real-world feedback.
Testing Methods: Substantial equivalence in performance matrices as compared to the AI Platform 2.0. Non-inferiority margins in alignment with FDA recommendations will be used to ensure the new device remains safe and effective without any such change compared with the original device and the last modified version of the device.
Impact Assessment: Improved model generalizability for AI model outputs displayed to users. Benefit-Risk Analysis: Benefit: Improved model robustness, reduced bias, enhanced generalizability, and responding to real world feedback. Risk: Reduction in clinical performance. Risk Mitigation: The risks of overfitting and generalizability are mitigated by conducting cross-validation, data augmentation, and hyper parameter tuning in internal training procedures. The final validation is performed using non-inferiority analysis according to the established clinical protocol to ensure baseline performance is maintained and improvements are realized before modifications are committed and released.
The input and output data by the network will remain unchanged and model re-training will be conducted manually, not automatically.
Exo shall communicate AI Platform 2.0 software changes, performance changes, and labeling changes to end-users via customer and software update notifications.
The benefits and risks of each of the above modifications, independently and combined, including risks of social harm have been assessed, and the overall collective impact of implementing modifications remains low.
The activities proposed within the Modification Protocol will continue to reasonably ensure the safety and effectiveness of the device, and all modifications will maintain the AI Platform device within its stated intended use and indications for use.
§ 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).
0
August 5, 2024
Image /page/0/Picture/1 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 & Human Services logo on the left and the FDA logo on the right. The FDA logo includes the letters "FDA" in a blue square, followed by the words "U.S. FOOD & DRUG" and "ADMINISTRATION" in blue text.
Exo Imaging Jacqueline Murray Senior Regulatory Affairs Specialist 4201 Burton Drive Santa Clara, California 95054
Re: K240953
Trade/Device Name: AI Platform 2.0 (AIP002) Regulation Number: 21 CFR 892.2050 Regulation Name: Medical Image Management And Processing System Regulatory Class: Class II Product Code: QIH Dated: July 2, 2024 Received: July 3, 2024
Dear Jacqueline Murray:
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.
FDA's substantial equivalence determination also included the review and clearance of your Predetermined Change Control Plan (PCCP). Under section 515C(b)(1) of the Act, a new premarket notification is not required for a change to a device cleared under section 510(k) of the Act, if such change is consistent with an established PCCP granted pursuant to section 515C(b)(2) of the Act. Under 21 CFR 807.81(a)(3), a new premarket notification is required if there is a major change or modification in the intended use of a device,
1
or if there is a change or modification in a device that could significantly affect the safety or effectiveness of the device, e.g., a significant change or modification in design, material, chemical composition, energy source, or manufacturing process. Accordingly, if deviations from the established PCCP result in a major change or modification in the intended use of the device, or result in a change or modification in the device that could significantly affect the safety or effectiveness of the a new premarket notification would be required consistent with section 515C(b)(1) of the Act and 21 CFR 807.81(a)(3). Failure to submit such a premarket submission would constitute adulteration and misbranding under sections 501(f)(1)(B) and 502(o) of the Act, respectively.
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" (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 (QS) 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.
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-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).
2
Sincerely,
Saml for
Jessica Lamb, Ph.D. Assistant Director 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
510(k) Number (if known) K240953
Device Name AI Platform 2.0 (AIP002)
Indications for Use (Describe)
Al Platform 2.0 is intended for noninvasive processing of ultrasound images to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease. In addition, it can provide Quality Score feedback to assist healthcare professionals, trained and qualified to conduct echocardiography and lung ultrasound scans in the current standard of care while acquiring ultrasound images. The device is intended to be used on images of adult patients.
| Type of Use (Select one or both, as applicable) | |
|---|---|
| ☒ Prescription Use (Part 21 CER 801 Subpart D) | □ Over-The-Counter Use (21 CER 801 Subpart C) |
CONTINUE ON A SEPARATE PAGE IF NEEDED.
This section applies only to requirements of the Paperwork Reduction Act of 1995.
DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.
The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:
Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov
"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."
4
Image /page/4/Picture/1 description: The image shows the logo for EXO. On the left side of the logo is a graphic of 12 circles arranged in a pattern. The circles are colored with a blue-green gradient. To the right of the graphic is the word "EXO" in a sans-serif font. The letters are dark gray.
K240953
510(k) Summary
General Information
| 510(k) Sponsor | Exo Imaging |
|---|---|
| Address | 4201 Burton Drive |
| Santa Clara, CA 95054 | |
| Correspondence Person | Jacqueline Murray |
| Contact Information | jmurray@exo.inc |
| Cell: +1 236 838-5056 | |
| Date Prepared | April 08, 2024 |
Proposed Device
| Proprietary Name | Al Platform 2.0 (AIP002) |
|---|---|
| Common Name | Al Platform 2.0 |
| Classification Name | Automated Radiological Image Processing Software |
| Regulation Number | 21 CFR 892.2050 |
| Product Code | QIH |
| Regulatory Class | II |
Predicate Device
| Proprietary Name | LVivo IQS |
|---|---|
| Premarket Notification | K222970 |
| Classification Name | Automated Radiological Image Processing Software |
| Regulation Number | 21 CFR 892.2050 |
| Product Code | QIH |
| Regulatory Class | II |
Reference Device 1
| Proprietary Name | Caption Guidance |
|---|---|
| Premarket Notification | DEN190040 |
| Classification Name | Image Acquisition And/Or Optimization Guided by Artificial |
| Intelligence | |
| Regulation Number | 21 CFR 892.2100 |
5
Image /page/5/Picture/1 description: The image shows the logo for EXO. On the left side of the logo, there is a cluster of blue and green dots arranged in a pattern. To the right of the dots, the word "EXO" is written in a dark gray sans-serif font. The letters are bold and spaced closely together.
| Product Code | QJU |
|---|---|
| Regulatory Class | II |
Reference Device 2
| Proprietary Name | AI Platform |
|---|---|
| Premarket Notification | K232501 |
| Classification Name | Automated Radiological Image Processing Software |
| Regulation Number | 21 CFR 892.2050 |
| Product Code | QIH |
| Regulatory Class | II |
Device Description
Exo Al Platform 2.0 (AIP 2.0) is a software as a medical device (SaMD) that helps qualified users with image-based assessment of ultrasound examinations in adult patients. It is designed to simplify workflow by helping trained healthcare providers evaluate, quantify, and generate reports for ultrasound images. AIP 2.0 takes as an input in the Digital Imaging and Communications in Medicine (DICOM) format from ultrasound scanners of a specific range and allows users to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease. In addition, it provides frame and clip quality score in real-time for the Left Ventricle from the four-chamber apical and parasternal long axis views of the heart and lung scans. In addition, the Al modules are provided as a software component to be integrated by another computer programmer into their legally marketed ultrasound imaging device. Essentially, the Algorithm and API, which are modules, are medical device accessories.
Key features of the software are
- Lung Al: An Al-assisted tool for suggesting the presence of lung structures and artifacts on ● ultrasound images, namely A-lines. Additionally, a per-frame and per-clip quality score is generated for each lung scan.
- . Cardiac Al: An Al-assisted tool for the quantification of Left Ventricular Ejection Fraction (LVEF), Myocardium wall thickness (Interventricular Septum (IVSd), Posterior wall (PWd)), and IVC diameter on cardiac ultrasound images. Additionally, a per-frame and per-clip quality score is generated for each Apical and PLAX cardiac scan.
6
Image /page/6/Picture/1 description: The image contains the logo for EXO. On the left side of the logo, there is a cluster of blue circles arranged in a pattern. To the right of the circles, the word "EXO" is written in a bold, sans-serif font. The color of the text is a dark gray or black.
Warning:
It's important to note that patient management decisions should not be made solely on the Al Platform 2.0 analysis results. Users are fully responsible for ensuring scan quality and making diagnoses. Images obtained through the use of Quality Al should be interpreted exclusively by certified healthcare professionals. It is also imperative that a qualified healthcare professional reviews the data, ensuring its adequacy and appropriateness for the intended diagnosis or management.
Indications for Use
Al Platform 2.0 is intended for noninvasive processing of ultrasound images to detect, measure, and calculate relevant medical parameters of structures and function of patients with suspected disease. In addition, it can provide Quality Score feedback to assist healthcare professionals, trained and qualified to conduct echocardiography and lung ultrasound scans in the current standard of care while acquiring ultrasound images. The device is intended to be used on images of adult patients.
Comparison of Technological Characteristics with the Predicate Device
| Feature/Function | Subject Device Exo Al Platform 2.0 | Predicate Device: LVivo IQS (K222970) | Reference Device 1: Caption Guidance (DEN190040) | Reference Device 2: Al Platform (K232501) |
|---|---|---|---|---|
| Scan type | Single and Multi-frame ultrasound images | Same as subject device | Same as subject device | Same as subject device |
| Principle of Operation and Technology | Ultrasound image processing software implementing artificial intelligence, including non-adaptive machine learning algorithms trained with clinical data intended for non-invasive analysis of ultrasound data | Same as subject device | Same as subject device | Same as subject device |
| Feature/ | ||||
| Function | Subject Device | |||
| Exo Al Platform 2.0 | Predicate Device: | |||
| LVivo IQS | ||||
| (K222970) | Reference Device | |||
| 1: | ||||
| Caption Guidance | ||||
| (DEN190040) | Reference Device | |||
| 2: | ||||
| Al Platform | ||||
| (K232501) | ||||
| Al Algorithm | Deep Convolutional | |||
| Neural Networks for | ||||
| Segmentation, | ||||
| Landmark Detection | ||||
| and Classification | Same as subject | |||
| device | Same as subject | |||
| device | Same as subject | |||
| device | ||||
| Anatomical | ||||
| Sites | Heart, Lungs | Heart, Bladder | Heart | Heart, Lungs |
| Cardiac | ||||
| Measurements | LVEF | |||
| IVC Minimum | ||||
| diameter on | ||||
| inspiration and | ||||
| Maximum diameter | ||||
| on expiration | ||||
| Myocardium wall | ||||
| thickness | ||||
| (Interventricular | ||||
| Septum and | ||||
| Posterior wall) from | ||||
| Plax view | LVEF, GLS | |||
| RV size and | ||||
| function: Fractional | ||||
| Area Change | ||||
| (FAC), Free Wall | ||||
| Strain (FWS), | ||||
| Tricuspid annular | ||||
| plane systolic | ||||
| excursion (TAPSE) | No | LVEF | ||
| Non Cardiac Al | ||||
| modules | Absence/presence of | |||
| A-lines | ||||
| B-lines count | Bladder Volume | No | Absence / presence of | |
| A-lines | ||||
| B-lines count | ||||
| Real time | ||||
| feedback on | ||||
| quality | Yes | Same as subject | ||
| device | Same as subject | |||
| device | No | |||
| Retrospectively | ||||
| recording of | ||||
| Diagnostic | ||||
| quality clip | Yes | No | Same as subject | |
| Device | No | |||
| Feature/ | ||||
| Function | Subject Device | |||
| Exo Al Platform 2.0 | Predicate Device: | |||
| LVivo IQS | ||||
| (K222970) | Reference Device | |||
| 1: | ||||
| Caption Guidance | ||||
| (DEN190040) | Reference Device | |||
| 2: | ||||
| Al Platform | ||||
| (K232501) | ||||
| Al modules are | ||||
| an accessory to | ||||
| compatible | ||||
| general | ||||
| purpose | ||||
| diagnostic | ||||
| ultrasound | ||||
| systems | Yes | Same as subject | ||
| device | Same as subject | |||
| device | No |
7
Image /page/7/Picture/1 description: The image shows the logo for EXO. On the left side of the logo, there is a cluster of blue and cyan circles arranged in a pattern. To the right of the circles, the word "EXO" is written in a dark gray sans-serif font.
8
Image /page/8/Picture/1 description: The image shows the logo for EXO. On the left side of the logo, there is a cluster of blue and cyan circles arranged in a circular pattern. To the right of the circles, the word "EXO" is written in a dark gray sans-serif font. The letters are bold and spaced closely together.
Performance Data
Safety and performance of the Al Platform 2.0 has been evaluated and verified in accordance with software specifications and applicable performance standards through software verification and validation testing. Additionally, the software validation activities were performed in accordance with IEC 62304:2006/AC:2015 - Medical device software - Software life cycle processes, FDA's 'Content of Premarket Submissions for Device Software Functions'' Guidance for Industry and Food and Drug Administration Staff Document issued on June 14, 2023 and FDA Guidance (June 2022) "Technical performance assessment of quantitative imaging in radiological device premarket submissions".
Validation Performance Testing
The clinical performance of the Al Platform 2.0 was successfully evaluated on test data encompassing diverse demographic variables, including gender, age, and ethnicity from multiple sites in metropolitan cities with diverse racial patient populations. The test data was entirely separated from the training/tuning datasets and was not used for any part of the training/tuning. We also established auditability measures, by assigning a unique identification number to each study and its corresponding images.
The left ventricle Wall thickness Al and IVC Al were each evaluated on scans of 100 subjects, on images acquired from cart-based and portable ultrasound devices. The ground truth for all measurements were obtained as the average measurement of three experts. Performance was assessed by calculating Inter class correlation (ICC) between the AI and the ground truth.
The performance of Al Platform 2.0 for wall thickness and IVC compared with reference data is summarized in Table 1 below:
9
Image /page/9/Picture/1 description: The image shows the logo for EXO. The logo consists of a pattern of blue and teal dots arranged in a grid-like formation on the left side. To the right of the dots is the word "EXO" in a bold, sans-serif font. The color of the text is a dark gray or black.
| Measurement | ICC (95% CI) | |
|---|---|---|
| LV Wall thickness | InterVentricular Septum (IVSd) | 0.93 (0.89 – 0.96) |
| Posterior Wall (PWd) | 0.94 (0.89 – 0.97) | |
| Inferior vena cava (IVC) | IVC Dmin | 0.93 (0.90 – 0.95) |
| IVC Dmax | 0.94 (0.90 – 0.96) |
Table 1: Summary of Al Platform performance for left ventricle wall thickness and IVC measurement
The validation of Quality Al consists of validating on:
- a. Data previously acquired from 184 patients from various ultrasound devices and various cardiac pathologies, compared to quality rating by experienced sonographers on each frame and the entire clip. Total of 226 clips (29,732 frames) were used in this test. The overall agreement between the Quality Al and quality rated by the experienced sonographers was ICC = 0.94 (95% Cl .94 - . 95) for frames and ICC = 0.94 (95% Cl .92 - . 95) for clips.
- b. Data acquired after using the image and clip Quality Al in real time while scanning Lung, Apical 4 chamber, and Parasternal Long Axis views of the heart. In total, 396 lung and cardiac scans were done by 26 users with a wide range of ultrasound experience in POC settings, including 18 novice users who received two hours of training. 98.3% of the clips rated as ACEP quality of 3 or above by expert readers, also received at least "Minimum criteria met for diagnosis" image quality by Clip Quality Al. Additionally, 98.0% of scans that were considered as "Minimal criteria met for diagnosis" or "good" by Quality Al were also deemed diagnostic by experts (ACEP score of 3 or higher).
Predetermined Change Control Plan
A Predetermined Change Control Plan (PCCP) is included with Al Platform 2.0. The PCCP specifies anticipated software modifications, implementation methods and validation criteria that will be used to implement the software modifications in a controlled manner and ensure the subject device remains as safe and effective as the predicate.
The PCCP includes a specific list of anticipated software modifications which are summarized in the table below, as well as a Modification protocol describing the verification and validation activities that will support these planned modifications. The modification protocol incorporates impact assessment considerations and specifies requirements for data management, including data sources, collection, storage, and sequestration, as well as documentation and data segregation/re-use practices.
10
Image /page/10/Picture/1 description: The image shows a logo with two distinct parts. On the left, there is a cluster of blue-green gradient dots arranged in a circular or network-like pattern. To the right of the dots, the letters 'EXO' are displayed in a bold, sans-serif font, with a dark gray color. The overall design is clean and modern, suggesting a technology or innovation-related brand.
Specific test methods are specified in the PCCP to establish substantial equivalence relative to Al Platform 2.0, and include sample size determination, and acceptance criteria. Original and additional validation data will be analyzed to ensure the datasets are representative of the intended use population. To ensure validation test datasets are representative of the intended use population, each will meet minimum demographic requirements.
A summary of these software modifications, test methods, necessary performance requirements, and impact assessment, are presented in the table below:
| Modification | Rationale | Testing Methods | Impact Assessment |
|---|---|---|---|
| (A) Modification | |||
| to architecture, | |||
| pre/post | |||
| processing | The Al models in | ||
| Al Platform 2.0 | |||
| may be modified | |||
| in a focused and | |||
| bound manner, to | |||
| improve accuracy, | |||
| efficiency, and | |||
| adaptability while | |||
| maintaining the | |||
| safety and efficacy | |||
| of the device. | Substantial equivalence | ||
| in performance matrices | |||
| as compared to the Al | |||
| Platform 2.0. | |||
| Non-inferiority margins in | |||
| alignment with FDA | |||
| recommendations will be | |||
| used to ensure the new | |||
| device remains safe and | |||
| effective without any | |||
| such change compared | |||
| with the original device | |||
| and the last modified | |||
| version of the device. | Improved accuracy and/or efficiency metrics for | ||
| Al model performance. | |||
| Benefit-Risk Analysis: | |||
| Benefit: Enhanced performance; Improved | |||
| efficiency and scalability. | |||
| Risk: Reduction in clinical performance | |||
| Risk Mitigation: | |||
| The risks of overfitting and generalizability are | |||
| mitigated by conducting cross-validation, data | |||
| augmentation, and hyper parameter tuning in | |||
| internal training procedures. | |||
| The final validation is performed using | |||
| non-inferiority analysis according to the | |||
| established clinical protocol to ensure baseline | |||
| performance is maintained and improvements | |||
| are realized before modifications are committed | |||
| and released. | |||
| (B) Introduction | |||
| of new training | |||
| data | The training | ||
| dataset of the Al | |||
| models may be | |||
| augmented with a | |||
| broader and more | |||
| diverse range of | |||
| imaging data to | |||
| enhance model | |||
| robustness, | |||
| reduce bias, | |||
| improve | |||
| generalizability, | |||
| and respond | |||
| effectively to | |||
| real-world | |||
| feedback. | Substantial equivalence | ||
| in performance matrices | |||
| as compared to the Al | |||
| Platform 2.0. | |||
| Non-inferiority margins in | |||
| alignment with FDA | |||
| recommendations will be | |||
| used to ensure the new | |||
| device remains safe and | |||
| effective without any | |||
| such change compared | |||
| with the original device | |||
| and the last modified | |||
| version of the device | Improved model generalizability for Al model | ||
| outputs displayed to users. | |||
| Benefit-Risk Analysis: | |||
| Benefit: Improved model robustness, reduced | |||
| bias, enhanced generalizability, and responding | |||
| to real world feedback. | |||
| Risk: Reduction in clinical performance | |||
| Risk Mitigation: | |||
| The risks of overfitting and generalizability are | |||
| mitigated by conducting cross-validation, data | |||
| augmentation, and hyper parameter tuning in | |||
| internal training procedures. | |||
| The final validation is performed using | |||
| non-inferiority analysis according to the | |||
| established clinical protocol to ensure baseline | |||
| performance is maintained and improvements | |||
| are realized before modifications are committed | |||
| and released. |
11
Image /page/11/Picture/1 description: The image shows the logo for EXO. The logo consists of a pattern of blue circles on the left and the word "EXO" in dark gray on the right. The circles are arranged in a grid-like pattern, with some circles closer together than others. The word "EXO" is written in a sans-serif font.
The input and output data by the network will remain unchanged and model re-training will be conducted manually, not automatically.
Exo shall communicate Al Platform 2.0 software changes, performance changes, and labeling changes to end-users via customer and software update notifications.
The benefits and risks of each of the above modifications, independently and combined, including risks of social harm have been assessed, and the overall collective impact of implementing modifications remains low.
The activities proposed within the Modification Protocol will continue to reasonably ensure the safety and effectiveness of the device, and all modifications will maintain the Al Platform device within its stated intended use and indications for use.
Conclusions
Exo's Al Platform is substantially equivalent in intended use, design, principles of operation, technological characteristics, and safety features to the predicate device. There are no different questions of safety and/or effectiveness introduced by the Al Platform when used as intended.
The clinical performance of the Al Platform 2.0 product was successfully evaluated on test data encompassing diverse demographic variables, including gender, age and ethnicity from multiple clinical sites in metropolitan cities with diverse racial patient populations.
The A-Line, B-line and LVEF algorithms did not change during the development of the Al Platform 2.0 product and remain as per the cleared algorithms in K232501.
The activities proposed within the Modification Protocol will continue to reasonably ensure the safety and effectiveness of the product, and all modifications will maintain the Al Platform device within its stated intended use and indications for use.