(91 days)
Sonio Detect-K240406
Yes
The intended use and device description explicitly state the use of "machine learning techniques" and a "Machine Learning-Based Algorithm" for identifying and characterizing abnormal fetal ultrasound findings.
No
The device is intended to assist interpreting physicians by identifying and characterizing abnormal findings and providing information for potential diagnoses; it explicitly states that "Patient management decisions should not be made solely on the results of the Sonio Suspect analysis," indicating it does not directly provide therapy.
Yes
Explanation: The device is described as assisting "interpreting physicians...by automatically identifying and characterizing abnormal fetal ultrasound findings." It also states, "The device provides information on abnormal findings that may be useful in rendering potential diagnosis." This indicates its role in the diagnostic process.
Yes
The device is described as a "Software as a Service (SaaS) solution" and a "web application accessible from any device connected to the internet." It processes uploaded or received images and provides analysis, without mentioning any specific hardware components included with the device itself. While it can connect to external devices (like ultrasound machines) via API, the core device functionality is software-based.
Based on the provided information, Sonio Suspect is not an In Vitro Diagnostic (IVD) device.
Here's why:
- IVD Definition: In Vitro Diagnostics are devices intended for use in the collection, preparation, and examination of specimens taken from the human body (such as blood, urine, tissue) to provide information for diagnostic purposes.
- Sonio Suspect's Function: Sonio Suspect analyzes images acquired from a fetal ultrasound examination. It does not analyze biological specimens taken from the patient.
- Intended Use: The intended use clearly states it's a "concurrent reading aid on acquired images, during and/or after fetal ultrasound examinations." This reinforces that it works with imaging data, not biological samples.
- Device Description: The description confirms it's a "Software as a Service (SaaS) solution" that processes "Ultrasound image file uploaded by the HCP."
While Sonio Suspect provides information that may be useful in rendering a potential diagnosis, it does so by analyzing imaging data, which falls outside the scope of an In Vitro Diagnostic device. It is a medical device, specifically a software medical device, but not an IVD.
No
The clearance letter does not explicitly state that the FDA has reviewed and approved or cleared a Predetermined Change Control Plan (PCCP) for this specific device.
Intended Use / Indications for Use
Sonio Suspect is intended to assist interpreting physicians, during or after fetal ultrasound examinations, by automatically identifying and characterizing abnormal fetal ultrasound findings on detected views, using machine learning techniques.
The device is intended for use as a concurrent reading aid on acquired images, during and/or after fetal ultrasound examinations.
The device provides information on abnormal findings that may be useful in rendering potential diagnosis.
Patient management decisions should not be made solely on the results of the Sonio Suspect analysis.
Product codes
POK
Device Description
Sonio Suspect is a Software as a Service (SaaS) solution that aims at helping interpreting physicians (designated as healthcare professionals i.e. HCP in the following) to identify abnormal fetal ultrasound findings during and/or after fetal ultrasound examinations.
Sonio Suspect is a web application accessible from any device connected to the internet. It can be accessed on a tablet, computer or any other support capable of providing access to a web application.
Sonio Suspect can be used by HCPs as a concurrent reading aid on acquired images, to assist them during and/or after fetal ultrasound examinations of gestational age (GA): from 11 weeks to 41 weeks. A concurrent read by the users means a read in which the device output is available during and/or after the fetal ultrasound examination.
The way Sonio Suspect is built allows the HCP to use it at any moment. The software can process any Ultrasound image file uploaded by the HCP, at any time.
Sonio Suspect can be connected through API to external devices (as an ultrasound machine) to receive images.
Sonio Suspect workflow goes through the following steps:
As soon as an image is automatically received, it is automatically detected and associated with a view (and can be manually re-associated by the HCP). Then abnormal fetal ultrasound findings linked to the view are evaluated and displayed, individually, with one of the following status:
- Suspected (abnormal findings identified on the image);
- . Not Suspected (abnormal findings not identified on the image);
- . Can't be analyzed (abnormal findings not evaluated due to one or several structures not detected or if the fetal position selected is "other or unknown" while it's required to evaluate the abnormal finding).
Each abnormal finding status can be manually overridden to Present or Not Present by the user.
Mentions image processing
Yes
Mentions AI, DNN, or ML
Yes, "machine learning techniques", "machine learning-based algorithms"
Input Imaging Modality
Fetal Ultrasound images
Anatomical Site
Fetal (Chest, Abdominal, Cephalic)
Indicated Patient Age Range
Gestational Age (GA): from 11 weeks to 41 weeks
Intended User / Care Setting
Interpreting physicians / Not specified, but accessible via web application on any device.
Description of the training set, sample size, data source, and annotation protocol
Not Found. The document states: "This global validation dataset was independent of the data used during model development (training/finternal validation) and the establishment of device operating points."
Description of the test set, sample size, data source, and annotation protocol
Standalone performance testing: "dataset of 8745 fetal ultrasound images from 1115 exams collected across 75 sites, 64 of which are located in the United States, representing the intended use population." Data source and annotation protocol not explicitly detailed beyond "collected".
Clinical performance testing: "750 fetal ultrasound images (between 11 and 41 weeks)" evaluated by "13 readers (5 MFM, 6 OB/GYN and 2 Diagnostic radiologists of 1-30+ years' experience)". "The dataset included 287 distinct exams from 47 sites, including 37 sites which are located in the United States." Data source and annotation protocol not explicitly detailed beyond "collected" and "For each image, each reader was required to provide a binary determination of the presence or absence of an abnormal finding and to provide a score representing their confidence in their annotation."
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Standalone Performance Testing:
- Study type: Standalone performance testing in accordance with 21 CFR §892.2060 special control 1(iv).
- Sample size: 8745 fetal ultrasound images from 1115 exams collected across 75 sites.
- Standalone performance: Sensitivity of 93.2% (95% CI: [91.6%-94.6%]), Specificity of 90.8% (95% CI: [89.5%-92.0%]).
- Key results: Automatically detects abnormal findings with high sensitivity and specificity. The abnormal finding "Abdominal Situs Inversus" had the highest performance (Sensitivity 99.3%, Specificity 99.3%). The lowest sensitivity (87.7%) was for "Malposition of the great vessels" and lowest specificity (81.5%) for "Absence or unusual size of at least one of the 3 vessels".
Clinical Performance Testing:
- Study type: Pre-market fully-crossed multiple case (MRMC) retrospective reader study in accordance with 21 CFR §892.2060 special control 1(ii) and 1(iii).
- Sample Size: 13 readers each evaluated 750 fetal ultrasound images (total 9750 reads). Dataset included 287 distinct exams.
- MRMC: The primary objective was to determine whether the performance of readers assisted by Sonio Suspect ("Assisted") was superior to the performance of readers when not assisted by Sonio Suspect ("Unassisted").
- Key Results: The accuracy of readers in identifying abnormal findings was superior when "Assisted" by Sonio Suspect than when "Unassisted". The AUC in the "Unassisted" reading setting is estimated at 68.9%, whilst the AUC in the "Assisted" reading setting is estimated at 90.0% which represents a significant difference of 21.9%. The "Assisted" reading curve consistently lies above the "Unassisted" curve across all 8 abnormal findings and all 13 readers.
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
Standalone Performance:
- Average Performance: Sensitivity 0.932 (0.916, 0.946), Specificity 0.908 (0.895, 0.920)
- Malposition of the Great Vessels: Sensitivity 0.877 (0.840, 0.908), Specificity 0.933 (0.905, 0.957)
- Absence or unusual size of at least one of the 3 vessels: Sensitivity 0.959 (0.926, 0.983), Specificity 0.815 (0.759, 0.872)
- Disequilibrium OR absence of at least one of the two ventricles: Sensitivity 0.900 (0.861, 0.934), Specificity 0.817 (0.782, 0.851)
- Thoracic Situs Inversus: Sensitivity 0.961 (0.930, 0.987), Specificity 0.958 (0.927, 0.982)
- Abdominal Situs Inversus: Sensitivity 0.993 (0.976, 1.000), Specificity 0.993 (0.984, 1.000)
- Non-visibility of a single stomach bubble OR abnormally big stomach: Sensitivity 0.880 (0.780, 0.946), Specificity 0.964 (0.950, 0.977)
- Absence of the Cavum Septum Pellucidum: Sensitivity 0.911 (0.861, 0.956), Specificity 0.925 (0.882, 0.962)
- Absence of the Corpus Callosum: Sensitivity 0.976 (0.952, 0.994), Specificity 0.859 (0.805, 0.907)
Clinical Performance (AUC):
- Overall AUCUnassisted: 0.689
- Overall AUCAssisted: 0.900
- Overall AUCDelta: 0.219
- Abdominal Situs Inversus: AUCUnassisted 0.696, AUCAssisted 0.953, AUCDelta 0.258
- Absence of at least one of the 2 ventricles OR disequilibrium of the 2 ventricles: AUCUnassisted 0.742, AUCAssisted 0.916, AUCDelta 0.174
- Absence of the cavum septum pellucidum: AUCUnassisted 0.567, AUCAssisted 0.884, AUCDelta 0.317
- Absence of the corpus callosum: AUCUnassisted 0.646, AUCAssisted 0.899, AUCDelta 0.253
- Absence of the stomach OR presence of two stomachs: AUCUnassisted 0.84, AUCAssisted 0.955, AUCDelta 0.114
- Absence or unusual size of at least one of the 3 vessels: AUCUnassisted 0.736, AUCAssisted 0.904, AUCDelta 0.168
- Great vessels malposition: AUCUnassisted 0.616, AUCAssisted 0.872, AUCDelta 0.256
- Thoracic situs inversus: AUCUnassisted 0.616, AUCAssisted 0.906, AUCDelta 0.29
Predicate Device(s)
Koios DS, K212616
Reference Device(s)
Sonio Detect-K240406
Predetermined Change Control Plan (PCCP) - All Relevant Information
Not Found
§ 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).
0
Image /page/0/Picture/0 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo, which is a blue square with the letters "FDA" in white. To the right of the blue square is the text "U.S. FOOD & DRUG ADMINISTRATION" in blue.
Sonio Florian Akpakpa Head of Regulatory Affairs and Quality Assurance 17 rue du Faubourg Montmartre Paris, 75009 France
February 21, 2025
Re: K243614
Trade/Device Name: Sonio Suspect 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: January 24, 2025 Received: January 24, 2025
Dear Florian Akpakpa:
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 (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.
1
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.
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 Rue"). 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-udi-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.
2
Page
3
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,
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)
Device Name
Sonio Suspect
Indications for Use (Describe)
Sonio Suspect is intended to assist interpreting physicians, during or after fetal ultrasound examinations, by automatically identifying and characterizing abnormal fetal ultrasound findings on detected views, using machine learning techniques.
The device is intended for use as a concurrent reading aid on acquired images, during and/or after fetal ultrasound examinations.
The device provides information on abnormal findings that may be useful in rendering potential diagnosis.
Patient management decisions should not be made solely on the results of the Sonio Suspect analysis.
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/1 description: The image contains the logo for Sonio. The logo consists of a stylized blue icon resembling a person with a circular head and curved body, followed by the word "sonio" in a sans-serif font, also in blue. The logo is simple and modern, with a clean design.
510(k) Summary
In accordance with 21 CFR 807.92 the 510(k) summary for Sonio Suspect is provided below.
I. Submitter
| Applicant: | Sonio |
|---|---|
| 147 Rue d'Aboukir, | |
| 75002, Paris France | |
| Primary Contact Person: | Florian Akpakpa |
| Director Regulatory Affairs and Quality Assurance | |
| Sonio | |
| Phone: +33 6 19 38 71 45 | |
| Email: florian.akpakpa@sonio.ai | |
| Date Prepared: | February 19, 2025 |
II. Device
| Device Trade | Sonio Suspect | ||
|---|---|---|---|
| Name: | |||
| Classification | 21 CFR 892.2060 - Radiological computer-assisted diagnostic | ||
| Name: | software for lesions suspicious of cancer | ||
| Regulatory Class: | Class II | ||
| Product Code: | POK (primary) |
III. Predicate Device
Koios DS from the manufacturer Koios Medical, Inc. cleared in K212616.
This predicate has not been subject to a design-related recall.
The following reference device was used: Sonio Detect-K240406.
5
Image /page/5/Picture/1 description: The image contains the logo for "sonio". The logo consists of a stylized, abstract symbol to the left of the word "sonio". The symbol is a blue, curved shape with a dot above it. The word "sonio" is written in a sans-serif font and is also blue. The logo is simple and modern.
IV. Device Description
Sonio Suspect is a Software as a Service (SaaS) solution that aims at helping interpreting physicians (designated as healthcare professionals i.e. HCP in the following) to identify abnormal fetal ultrasound findings during and/or after fetal ultrasound examinations.
Sonio Suspect is a web application accessible from any device connected to the internet. It can be accessed on a tablet, computer or any other support capable of providing access to a web application.
Sonio Suspect can be used by HCPs as a concurrent reading aid on acquired images, to assist them during and/or after fetal ultrasound examinations of gestational age (GA): from 11 weeks to 41 weeks. A concurrent read by the users means a read in which the device output is available during and/or after the fetal ultrasound examination.
The way Sonio Suspect is built allows the HCP to use it at any moment. The software can process any Ultrasound image file uploaded by the HCP, at any time.
Sonio Suspect can be connected through API to external devices (as an ultrasound machine) to receive images.
Sonio Suspect workflow goes through the following steps:
As soon as an image is automatically received, it is automatically detected and associated with a view (and can be manually re-associated by the HCP). Then abnormal fetal ultrasound findings linked to the view are evaluated and displayed, individually, with one of the following status:
- Suspected (abnormal findings identified on the image);
- . Not Suspected (abnormal findings not identified on the image);
- . Can't be analyzed (abnormal findings not evaluated due to one or several structures not detected or if the fetal position selected is "other or unknown" while it's required to evaluate the abnormal finding).
Each abnormal finding status can be manually overridden to Present or Not Present by the user.
6
Image /page/6/Picture/0 description: The image shows the logo for Sonio. The logo consists of a stylized blue icon on the left and the word "sonio" in blue text on the right. The icon appears to be a stylized sound wave or a curved shape with a dot above it. The text is in a sans-serif font and is all lowercase.
The list of abnormal findings that Sonio Suspect can automatically detect is detailed in Table 1 below.
Table 1: list of Fetal Abnormal Findings in Sonio Suspect scope and their associated View, Fetal Anatomy and Gestational Age (GA)
| Fetal
| Anatomy | Abnormal Finding | View | GA | |
|---|---|---|---|---|
| Chest | Absence or unusual size of at least one of the 3 | |||
| vessels | 3 vessels | T2/T3 | ||
| Malposition of the great vessels | LVOT/RVOT | T2/T3 | ||
| Disequilibrium OR absence of at least one of the | ||||
| two ventricles | 4 chambers | T1/T2/T3 | ||
| Thoracic situs inversus | ||||
| Abdominal | Abdominal situs inversus | Abdominal | ||
| Non-visibility of a single stomach bubble OR | ||||
| circumference | ||||
| abnormally big stomach | T1/T2/T3 | |||
| Cephalic | Absence of the cavum septum pellucidum | Transthalamic view | T2/T3 | |
| Absence of the Corpus Callosum | Corpus callosum | |||
| view* | T2/T3 |
*Note: The views are automatically detected by Sonio Suspect besides the Corpus callosum view on which abnormal finding identification and characterization is done by the software when it is manually associated by the users.
7
Image /page/7/Picture/1 description: The image shows the logo for Sonio. The logo consists of a stylized blue symbol to the left of the word "sonio" in blue lowercase letters. The symbol appears to be a stylized wave or abstract shape, and the overall design is clean and modern.
V. Indications for Use
Sonio Suspect is intended to assist interpreting physicians, during or after fetal ultrasound examinations, by automatically identifying and characterizing abnormal fetal ultrasound findings on detected views, using machine learning techniques.
The device is intended for use as a concurrent reading aid on acquired images, during and/or after fetal ultrasound examinations.
The device provides information on abnormal findings that may be useful in rendering potential diagnosis.
Patient management decisions should not be made solely on the results of the Sonio Suspect analysis.
Sonio Suspect and the predicate device, Koios DS, have similar intended use in assisting interpreting physicians in analyzing ultrasound images and characterizing ultrasound image items using machine learning techniques. Both devices are used as an aid to diagnosis and provide information that may be useful in rendering potential diagnosis.
The indication for use of Sonio Suspect and the predicate device differ in the following:
- The target population: Sonio Suspect is indicated for pregnant women undergoing fetal ultrasound examinations of the fetus while Koios DS is indicated for adult female patients with soft tissue breast lesions and/or all adult patients with thyroid nodules suspicious for cancer
- . Feature scope: While the predicate Koios DS, requires the user to select or confirm regions of interests (ROIs) within an image to be analyzed, Sonio Suspect automatically detects fetal ultrasound views on which abnormal fetal ultrasound findings will be identified.
- . Feature scope: Koios DS allows the users to adjust, measure, and document images and output into a structured report while Sonio Suspect does not.
However, these differences should not raise new questions regarding the safety and effectiveness of the device when used as labeled.
8
VI. Comparison of Technological Characteristics with the Predicate Device
Table 2 provides a comparison of the Technological Characteristics of Sonio Suspect to the predicate Koios DS cleared in K212616.
Table 2 - comparison of the Technological Characteristics of Sonio Suspect to the predicate
| Items | Predicate: Koios DS
Proposed device: Sonio Suspect | | |
|--------------------------|-----------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------|--|
| Manufacturer
name | Koios Medical | Sonio | |
| Device name | Koios DS | Sonio Suspect | |
| Regulation
Number | 21 CFR 892.2060
21 CFR 892.2050 | 21 CFR 892.2060 | |
| Product code | POK
OIH | POK | |
| Image modality | Breast Ultrasound Data
Thyroid Ultrasound
DataUltrasound images | Fetal Ultrasound images | |
| Algorithm
Methodology | Computer vision
Machine Learning Techniques | Computer vision
Machine Learning-Based
Algorithm | |
| Platform | ASP.NET web application
deployed to a Microsoft IIS web
server inside a Windows
operating system environment | Secure cloud-based and stand-alone
software compatible with an
ultrasound system | |
The technical principle of both Sonio Suspect and the predicate Koios DS is the characterization of ultrasound images items using computer vision and machine learningbased algorithms. Both devices use ultrasound images modality for analysis.
Sonio Suspect differs from Koios DS in the following:
- Operating platform: Sonio Suspect is a secure cloud-based and standalone software compatible with ultrasound systems while Koios DS the predicate is a ASP.NET web application deployed to a Microsoft IIS web server inside a Windows operating system environment.
However, these differences should not raise new questions regarding the safety and effectiveness of the device when used as labeled.
9
VII. Performance Data
The following performance data were provided in support of the substantial equivalence determination.
Software Verification and Validation Testing
Software verification and validation testing were conducted, and documentation was provided as recommended by FDA's Guidance for Industry and FDA Staff, "Content of Premarket Submissions for Device Software Functions."
The following quality assurance measures were applied to the development of the system:
- Risk Analysis
- Design Reviews ●
- Software Development Lifecycle ●
- Algorithm Verification (Algorithm internal validation) ●
- Software units verification ●
- Software verification
- Simulated use testing (Validation)
- Performance testing
- Cybersecurity testing .
Performance testing - Bench
Sonio conducted a standalone performance testing in accordance with 21 CFR §892.2060 special control 1(iv). The testing was conducted on a dataset of 8745 fetal ultrasound images from 1115 exams collected across 75 sites, 64 of which are located in the United States, representing the intended use population. This global validation dataset was independent of the data used during model development (training/finternal validation) and the establishment of device operating points.
The results of the standalone performance testing demonstrated that Sonio Suspect automatically detects abnormal findings with a sensitivity of 93.2% (Confidence Interval of [91.6%-94.6%]) and a specificity of 90.8% (Confidence Interval of [89.5%-92.0%]).
Table 3 below summarizes Sonio Suspect's performance results.
The abnormal finding "Abdominal Situs Inversus" has the highest performance in terms of both sensitivity (99.3%) and specificity (99.3%). The lowest sensitivity (87.7%) is associated with the abnormal finding "Malposition of the great vessels" and the lowest specificity (81.5%) with the abnormal finding "Absence or unusual size of at least one of the 3 vessels".
10
Image /page/10/Picture/0 description: The image shows the logo for Sonio. The logo consists of a stylized blue icon on the left and the word "sonio" in blue letters on the right. The icon appears to be a stylized representation of sound waves or a similar concept.
Sonio 510(k) Premarket Notification Submission
| Table 3: Performance of Sonio Suspect for abnormal finding detection. | |
|---|---|
| ----------------------------------------------------------------------- | -- |
| Fetal Anatomy | Gestational
Age | Abnormal Finding | Sensitivity | Specificity |
|------------------------|--------------------|------------------------------------------------------------------------|--------------------------------------|--------------------------------------|
| | | | Point Estimate
(95% bootstrap CI) | Point Estimate
(95% bootstrap CI) |
| Average
Performance | T1/T2/T3 | Average Performance | 0.932
(0.916, 0.946) | 0.908
(0.895, 0.920) |
| Chest | T2/T3 | Malposition of the Great Vessels | 0.877
(0.840, 0.908) | 0.933
(0.905, 0.957) |
| | | Absence or unusual size of at least one of the 3 vessels | 0.959
(0.926, 0.983) | 0.815
(0.759, 0.872) |
| | T1/T2/T3 | Disequilibrium OR absence of at least one of the two
ventricles | 0.900
(0.861, 0.934) | 0.817
(0.782, 0.851) |
| | | Thoracic Situs Inversus | 0.961
(0.930, 0.987) | 0.958
(0.927, 0.982) |
| Abdominal | T1/T2/T3 | Abdominal Situs Inversus | 0.993
(0.976, 1.000) | 0.993
(0.984, 1.000) |
| | | Non-visibility of a single stomach bubble OR abnormally
big stomach | 0.880
(0.780, 0.946) | 0.964
(0.950, 0.977) |
| Cephalic | T2/T3 | Absence of the Cavum Septum Pellucidum | 0.911
(0.861, 0.956) | 0.925
(0.882, 0.962) |
| | | Absence of the Corpus Callosum | 0.976
(0.952, 0.994) | 0.859
(0.805, 0.907) |
11
Image /page/11/Picture/0 description: The image shows the logo for "sonio". The logo consists of a stylized blue icon resembling a sound wave or a stylized letter "S", followed by the word "sonio" in a sans-serif font, also in blue. The logo is simple and modern, with a clean design.
Additionally, the performance for the detection of abnormal fetal ultrasound findings was also validated for subgroups including Indication of examination, Race, Ethnicity, BMI, Gestational Age (GA), Maternal age, Manufacturer, Geography (OUS and US), Finding Status (Findingpositive and Finding-negative images) and Fetal Anatomy, and detailed subgroup analysis results are reported in the User Manual.
Sonio Suspect was validated only on GE, Philips, Samsung and Canon Ultrasound devices.
The results of verification and performance testing demonstrate the safe and effective use of Sonio Suspect.
Performance testing - Clinical
Sonio conducted a clinical performance testing in accordance with 21 CFR §892.2060 special control 1(ii) and 1(iii). A pre-market fully-crossed multiple case (MRMC) retrospective reader study was conducted to determine the impact of Sonio Suspect on reader accuracy in identifying abnormal fetal ultrasound findings. The MRMC study consisted of two independent reading sessions separated by a washout period of at least 28 days in order to avoid memory bias.
The primary objective of this study was to determine whether the performance of readers assisted by Sonio Suspect ("Assisted") was superior to the performance of readers when not assisted by Sonio Suspect ("Unassisted").
13 readers (5 MFM, 6 OB/GYN and 2 Diagnostic radiologists of 1-30+ years' experience) each evaluated 750 fetal ultrasound images (between 11 and 41 weeks) under both "Assisted" and "Unassisted" reading settings. The dataset included 287 distinct exams from 47 sites, including 37 sites which are located in the United States. For each image, each reader was required to provide a binary determination of the presence or absence of an abnormal finding and to provide a score representing their confidence in their annotation.
The results of the study demonstrated that the accuracy of readers in identifying abnormal findings was superior when "Assisted" by Sonio Suspect than when "Unassisted". Particularly, the AUC in the "Unassisted" reading setting is estimated at 68.9%, whilst the AUC in the "Assisted" reading setting is estimated at 90.0% which represents a significant difference of 21.9%, as shown in Figure 1 and Table 4 below.
12
Image /page/12/Picture/0 description: The image contains the logo for "sonio". The logo consists of a stylized blue icon resembling a water droplet or a sound wave, followed by the word "sonio" in a rounded, sans-serif font, also in blue. The logo is simple and modern, with a clean and professional appearance.
Figure 1: ROC curves averaged over all readers and over all abnormal findings for ''Assisted'' (red) and ''Unassisted'' (blue) reading settings.
Image /page/12/Figure/3 description: The image is a plot comparing the true positive rate (sensitivity) and false positive rate (1 - specificity) for two reading settings: assisted and unassisted. The y-axis represents the true positive rate, ranging from 0% to 100%, while the x-axis represents the false positive rate, also ranging from 0% to 100%. The plot shows two curves, one for the assisted reading setting and one for the unassisted reading setting, along with a dashed diagonal line.
13
Image /page/13/Picture/0 description: The image shows the logo for Sonio. The logo consists of a blue abstract shape resembling a water droplet or a stylized letter 'S' on the left. To the right of the shape is the word "sonio" in a sans-serif font, also in blue. The logo has a clean and modern design.
| Per Abnormal Finding | N of
finding-
positive
images | N of
finding-
negative
images | AUCDelta | AUCUnassisted | AUCAssisted |
|--------------------------------------------------------------------------------------------|----------------------------------------|----------------------------------------|--------------------------------|--------------------------------------|--------------------------------|
| | | | Point Estimate
(95% BDG CI) | Point
Estimate
(95% BDG
CI) | Point Estimate
(95% BDG CI) |
| | | | Overall | 250 | 500 |
| Abdominal Situs
Inversus | 32 | 143 | 0.258
(0.177, 0.339) | 0.696
(0.602, 0.789) | 0.953
(0.913, 0.994) |
| Absence of at least one
of the 2 ventricles OR
disequilibrium of the 2
ventricles | 31 | 145 | 0.174
(0.1, 0.248) | 0.742
(0.668, 0.816) | 0.916
(0.877, 0.954) |
| Absence of the cavum
septum pellucidum | 31 | 55 | 0.317
(0.249, 0.384) | 0.567
(0.52, 0.614) | 0.884
(0.81, 0.957) |
| Absence of the corpus
callosum | 31 | 55 | 0.253
(0.206, 0.3) | 0.646
(0.599, 0.694) | 0.899
(0.851, 0.948) |
| Absence of the stomach
OR presence of two
stomachs | 31 | 144 | 0.114
(0.071, 0.158) | 0.84
(0.785, 0.895) | 0.955
(0.916, 0.993) |
| Absence or unusual size
of at least one of the 3
vessels | 31 | 55 | 0.168
(0.097, 0.239) | 0.736
(0.674, 0.798) | 0.904
(0.852, 0.956) |
| Great vessels
malposition | 31 | 110 | 0.256
(0.183, 0.329) | 0.616
(0.555, 0.676) | 0.872
(0.809, 0.934) |
| Thoracic situs inversus | 32 | 144 | 0.29
(0.231, 0.349) | 0.616
(0.559, 0.672) | 0.906
(0.855, 0.957) |
Table 4: AUCDelta, AUCUnassisted, AUCAssisted over all Readers per Abnormal Finding. (C1 = Confidence Interval)
14
Image /page/14/Picture/0 description: The image contains the logo for Sonio. The logo consists of a stylized blue icon resembling a sound wave or a stylized letter 'S', followed by the word "sonio" in a sans-serif font, also in blue. The logo is simple and modern in design.
Figure 2 and 3 below show that the "Assisted" reading curve consistently lies above the "Unassisted" curve across all 8 abnormal findings and all 13 readers, indicating the improved performance of reader accuracy in identifying abnormal findings with Sonio Suspect. This consistent pattern highlights Sonio Suspect's effectiveness in improving reader accuracy across a variety of abnormal findings and regardless of individual variability.
Figure 2: "Assisted" (red) and "Unassisted" (blue) ROC curves for each abnormal finding
Reading setting - Assisted - Unassisted
False Positive Rate (1 - Specificity)
15
Image /page/15/Picture/0 description: The image shows the logo for Sonio. The logo consists of a stylized blue icon on the left and the word "sonio" in blue on the right. The icon appears to be a stylized representation of a sound wave or a droplet with a smaller circle above it.
Image /page/15/Figure/2 description: The image contains 13 ROC curves comparing the performance of assisted and unassisted reading settings for different readers. Each subplot represents a different reader, labeled from Reader 1 to Reader 13. The y-axis represents the true positive rate (sensitivity), ranging from 0% to 100%, while the x-axis represents the false positive rate, also ranging from 0% to 100%. The red lines represent the assisted reading setting, and the teal lines represent the unassisted reading setting.
Figure 3: "Assisted" (red) and "Unassisted" (blue) ROC curves for each reader Reading setting — Assisted — Unassisted
False Positive Rate (1 - Specificity)
16
Image /page/16/Picture/1 description: The image contains the logo for Sonio. The logo consists of a stylized, abstract symbol to the left of the word "sonio". The symbol is blue and appears to be a rounded shape with a dot above it. The word "sonio" is also in blue and is written in a sans-serif font.
VIII. Conclusions
The conclusions drawn from the standalone and clinical studies demonstrate that Sonio Suspect is substantially equivalent to the predicate device Koios DS cleared in K212616.
The special controls for 21 CFR 892.2060 regulation are satisfied by demonstrating the effectiveness of the device in both the standalone testing and the clinical testing, showing the superiority of "Assisted" versus "Unassisted" readings in the clinical testing and communicating testing results in the labeling.
Sonio Suspect intended use, clinical outcome, and clinical applications are similar to those of the predicate device. The technological characteristics differences identified and discussed in Section VI do not raise any different questions of safety and effectiveness of the device.
Thus, Sonio Suspect is substantially equivalent to its predicate Koios DS cleared in K212616.