(125 days)
The Tyto Insights for Crackles Detection is an over-the-counter artificial intelligence (AI) enabled decision support software system used in the evaluation of lung sounds in adults and pediatrics (2 years and older). It automatically analyzes the acoustic signal of the lung as recorded by the FDA 510k cleared compatible Tyto Stethoscope and identifies recordings where a specific abnormal lung sound suggestive of "Crackle" is suspected. It is not intended to detect other abnormal or normal lung sounds. A licensed health care professional’s advice is required to understand the meaning of the Tyto Insights for Crackles Detection result. Healthcare providers should consider the device result in conjunction with recording and other relevant patient data.
The Tyto Insights for Crackles Detection is a web-based (AI) enabled software system designed to aid in the clinical assessment of lungs auscultation sound data by analyzing recorded lung sounds to determine whether a Crackle is detected within the recorded sound data. The Tyto Insights for Crackles Detection Software is intended to process recordings from the FDA-cleared compatible Tyto Stethoscope (Tyto Stethoscope, K181612). The acquisition of the acoustic data (recordings) is carried out by a professional user in a clinical environment or by a lay- user in a non-medical environment, in compliance with the labeling of the Tyto Stethoscope. The system is composed of the following sub-systems:
- The Tyto Insights for Crackles Detection Application Server (APS) communicates with 1. the Tyto Insights for Crackles Detection Algorithm Server (ALS) and implements an application programming interface (API) for communication with the telehealth server.
- The Tyto Insights for Crackles Detection Algorithm Server (ALS) receives an audio file 2. as input and returns an analysis result of positive or negative regarding whether a Crackles was detected as output.
- The Tyto Insights for Crackles Detection Web Server (WBS) provides a graphic 3. indication whether a Crackles is detected in the recording. It can be utilized both in patient and clinician side.
All the software subsystems (servers and storage) are hosted in the cloud and communicate through IP network.
Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) summary for "Tyto Insights for Crackles Detection":
Product Information:
- Trade/Device Name: Tyto Insights for Crackles Detection
- Regulation Number: 21 CFR 868.1900
- Regulation Name: Diagnostic Pulmonary-Function Interpretation Calculator
- Regulatory Class: Class II
- Product Code: PHZ
- Intended Use/Indications for Use: An over-the-counter AI-enabled decision support software system for evaluating lung sounds (adults and pediatrics 2+ years) recorded by the compatible Tyto Stethoscope. It identifies recordings where "Crackle" is suspected. It is not intended to detect other abnormal/normal lung sounds. Requires a licensed healthcare professional's advice to interpret results, which should be considered with other patient data.
1. Table of Acceptance Criteria and Reported Device Performance
| Parameter | Acceptance Criteria (from PCCP) | Reported Device Performance (Stand-Alone) | Reported Device Performance (Clinical Accuracy vs. Readers) |
|---|---|---|---|
| Co-Primary Endpoints | |||
| Sensitivity (Se) | LCI > 0.6279 (for modifications) | 0.72 (0.63-0.79) | Not directly comparable (Clinical Readers AUC is a composite measure) |
| Specificity (Sp) | LCI > 0.9668 (for modifications) | 0.99 (0.97-1.00) | Not directly comparable (Clinical Readers AUC is a composite measure) |
| AUC (Area Under the Curve) | Not explicitly defined as a direct acceptance criterion for the initial submission, but non-inferiority margin used for comparison. | Not applicable (standalone metrics are Se, Sp, PPV, NPV) | Tyto Insights for Crackles Detection AUC: 0.97 (0.95–0.98) |
| Difference in AUC | Lower bound of 95% two-sided CI for (Device AUC - Clinical Readers AUC) > -0.05 (non-inferiority margin) | Not applicable | 0.2 (0.17–0.23). Meets criterion: 0.17 is > -0.05. |
| Secondary Endpoints | |||
| Positive Predictive Value (PPV) | (for modifications) | 0.63 (0.4-0.87) | Not applicable |
| Negative Predictive Value (NPV) | (for modifications) | 0.99 (0.98-0.99) | Not applicable |
| Repeatability (Software vs. Readers) | Not explicitly defined | Software kappa: 1.0, agreement: 100% | Readers kappa: 0.42 (0.35 -0.49) |
Note: The acceptance criteria for Sensitivity and Specificity (LCI > 0.6279 and LCI > 0.9668, respectively) are specifically laid out in the "Predetermined Change Control Plan (PCCP)" section for modifications to the device. For the initial submission, the primary endpoint focused on the non-inferiority of the device's AUC compared to clinical readers. The reported standalone sensitivity and specificity are actual performance metrics from the validation study.
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size: 446 recordings (120 Crackles positive, 326 Crackles negative). This corresponded to a total of 445 patients.
- Data Provenance: Retrospective validation dataset. Recordings were obtained from the real-world use of the Tyto Care FDA-cleared compatible Tyto Stethoscope (K181612). The document does not specify the country of origin for the data.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
- Number of Experts: Three (3) blinded experienced Pulmonologists.
- Qualifications: "Experienced Pulmonologists" are specified. Further details on years of experience or specific board certifications are not provided in the summary.
4. Adjudication Method for the Test Set
- Adjudication Method: Binary ground truth was determined by a majority vote of the three blinded Pulmonologists. This is a form of 3+ Consensus.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
- MRMC Study Done?: A comparative effectiveness study was done comparing the device (AI) performance to "clinical readers" (Physicians non-Pulmonologists). However, this was an AI-only vs. human-only comparison, not a "human reading with AI assistance vs. human reading without AI assistance" MRMC study.
- Effect Size of Human Improvement with AI: This study did not assess how much human readers improve with AI assistance. It directly compared the AI algorithm's performance to human clinical readers.
- Device AUC: 0.97 (0.95–0.98)
- Clinical Readers AUC: 0.77 (0.73–0.8)
- Difference in AUC (Device - Clinical Readers): 0.2 (0.17–0.23). This indicates the AI algorithm significantly outperformed the clinical readers in this specific comparison, establishing non-inferiority (and superiority) based on the defined margin.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done
- Standalone Study Done?: Yes.
- Performance Metrics:
- Sensitivity: 0.72 (0.63-0.79)
- Specificity: 0.99 (0.97–1.00)
- Positive Predictive Value (PPV): 0.63 (0.4-0.87)
- Negative Predictive Value (NPV): 0.99 (0.98-0.99)
7. The Type of Ground Truth Used
- Type of Ground Truth: Expert Consensus. Specifically, a majority vote of three blinded experienced Pulmonologists on the presence or absence of "Crackle" in the lung sound recordings.
8. The Sample Size for the Training Set
- The document does not specify the exact sample size for the training set. It mentions that the device utilizes a CRNN (Convolutional Recurrent Neural Network) model and that "Each network is trained based on the target clinical class." However, it only provides details for the retrospective validation dataset.
9. How the Ground Truth for the Training Set Was Established
- The document does not explicitly describe how the ground truth for the training set was established. It primarily focuses on the ground truth establishment for the validation set used for performance evaluation. It's generally assumed that similar expert-driven annotation methods would be used for training data, but this is not detailed in the provided summary.
{0}------------------------------------------------
Image /page/0/Picture/0 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: the Department of Health & Human Services logo on the left and the FDA logo on the right. The FDA logo features the letters "FDA" in a blue square, followed by the words "U.S. FOOD & DRUG ADMINISTRATION" in blue text.
July 2, 2024
Tyto Care Ltd. Stella Raizelman Perry RA&QA Director 14 Beni Gaon Street Netanya, 4250803 Israel
Re: K240555
Trade/Device Name: Tyto Insights for Crackles Detection Regulation Number: 21 CFR 868.1900 Regulation Name: Diagnostic Pulmonary-Function Interpretation Calculator Regulatory Class: Class II Product Code: PHZ Dated: Mav 31, 2024 Received: May 31, 2024
Dear Stella Raizelman Perry:
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
{1}------------------------------------------------
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, 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 (OS) 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-regulatory
{2}------------------------------------------------
assistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).
Sincerely,
Image /page/2/Figure/3 description: The image shows a digital signature. The signature is for Binoy J. Mathews -S. The date of the signature is 2024.07.02 14:14:26 -04'00'.
For
Rachana Visaria Assistant Director DHT1C: Division of Sleep Disordered Breathing, Respiratory and Anesthesia Devices OHT1: Office of Ophthalmic, Anesthesia, Respiratory, ENT and Dental Devices Office of Product Evaluation and Quality Center for Devices and Radiological Health
Enclosure
{3}------------------------------------------------
Indications for Use
510(k) Number (if known) K240555
Device Name Tyto Insights for Crackles Detection
Indications for Use (Describe)
The Tyto Insights for Crackles Detection is an over-the-counter artificial intelligence (AI) enabled decision support software system used in the evaluation of lung sounds in adults and pediatrics (2 years and older). It automatically analyzes the acoustic signal of the lung as recorded by the FDA 510k cleared compatible Tyto Stethoscope and identifies recordings where a specific abnormal lung sound suggestive of "Crackle" is suspected. It is not intended to detect other abnormal or normal lung sounds. A licensed health care professional's advice is required to understand the meaning of the Tyto Insights for Crackles Detection result. Healthcare providers should consider the device result in conjunction with recording and other relevant patient data.
| Type of Use (Select one or both, as applicable) | |
|---|---|
| ------------------------------------------------- | -- |
Prescription Use (Part 21 CFR 801 Subpart D)
|X | 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/0 description: The image shows the logo for TytoCare. The logo features a stylized owl icon to the left of the word "tytocare" in a lowercase, sans-serif font. A trademark symbol is located to the upper right of the word "tytocare".
510(k) Summary
| Submitter Name andAddress: | Tyto Care Ltd.14 Beni Gaon Street Netanya, Israel,4250803 |
|---|---|
| Contact Person: | Stella Raizelman Perry RA & QADirectorEmail: stellar@tytocare.comPhone Number: +972 72-2210750Fax Number: +972 72-2210752 |
| EstablishmentRegistrationNumber: | 3012678246 |
| Date Prepared: | July 02, 2024 |
| Device TradeName(s): | Tyto Insights for Crackles Detection |
| Device CommonName: | Tyto Insights for Crackles Detection |
| Classification: | Name: Diagnostic pulmonary-function interpretationcalculatorProduct code: PHZRegulation No: 21 CFR 868.1900Class: IIPanel: Anesthesiology |
| Primary Predicate Device(s): | ||
|---|---|---|
| Device name | 510(k) No. | Date of Clearance |
| Tyto Insights for WheezeDetection | K232237 | December 13, 2023 |
| Reference Device(s): | ||
| Device name | 510(k) No. | Date of Clearance |
| VRIxp | K091732 | March 04, 2010 |
{5}------------------------------------------------
Intended use / indication for use statement
The Tyto Insights for Crackles Detection is an over-the-counter artificial intelligence (AI) enabled decision support software system used in the evaluation of lung sounds in adults and pediatrics (2 years and older). It automatically analyzes the acoustic signal of the lung as recorded by the FDA 510k cleared compatible Tyto Stethoscope and identifies recordings where a specific abnormal lung sound suggestive of "Crackle" is suspected. It is not intended to detect other abnormal or normal lung sounds. A licensed health care professional's advice is required to understand the meaning of the Tyto Insights for Crackles Detection result. Healthcare providers should consider the device result in conjunction with recording and other relevant patient data.
Device description
The Tyto Insights for Crackles Detection is a web-based (AI) enabled software system designed to aid in the clinical assessment of lungs auscultation sound data by analyzing recorded lung sounds to determine whether a Crackle is detected within the recorded sound data. The Tyto Insights for Crackles Detection Software is intended to process recordings from the FDA-cleared compatible Tyto Stethoscope (Tyto Stethoscope, K181612). The acquisition of the acoustic data (recordings) is carried out by a professional user in a clinical environment or by a lay- user in a non-medical environment, in compliance with the labeling of the Tyto Stethoscope. The system is composed of the following sub-systems:
- The Tyto Insights for Crackles Detection Application Server (APS) communicates with 1. the Tyto Insights for Crackles Detection Algorithm Server (ALS) and implements an application programming interface (API) for communication with the telehealth server.
- The Tyto Insights for Crackles Detection Algorithm Server (ALS) receives an audio file 2. as input and returns an analysis result of positive or negative regarding whether a Crackles was detected as output.
- The Tyto Insights for Crackles Detection Web Server (WBS) provides a graphic 3. indication whether a Crackles is detected in the recording. It can be utilized both in patient and clinician side.
All the software subsystems (servers and storage) are hosted in the cloud and communicate through IP network.
{6}------------------------------------------------
Substantial Equivalence to Predicate Devices
The following table compares the Tyto Insights for Crackles Detection to the predicate and reference device.
Table 1. Substantial Equivalence Summary
| Device | Primary Predicate device | Reference device | Summary | |
|---|---|---|---|---|
| Device Name | Tyto Insights for CracklesDetection | Tyto Insights for WheezeDetection | VRIxp | NA |
| DeviceManufacturer | Tyto Care Ltd. | Tyto Care Ltd. | Deep Breeze Ltd. | NA |
| 510(k) Number | TBD | K232237 | K091732 | NA |
| Device Class | Class II | Class II | Class II | Same as the predicatedevice. |
| Review Panel | Anesthesiology | Anesthesiology | CardiovascularDiagnostic Devices | Same as the primarypredicate device. |
| Product code | PHZ | PHZ | OCR | Same as the primarypredicate device. |
| Regulation number | 21 CFR 868.1900 | 21 CFR 868.1900 | 21 CFR 870.1875 | Same as the primarypredicate device. |
| DeviceClassification Name | Abnormal breath sounddevice | Abnormal breath sounddevice | Lung sound monitor | Same as the primarypredicate device. |
| Intended use andindication for use | Device | Primary Predicate device | Reference device | Summary |
| The “Tyto Insights forCrackles Detection” is anover-the-counter artificialintelligence (AI) enableddecision support softwaresystem used in theevaluation of lung soundsin adults and pediatrics (2years and older). Itautomatically analyses theacoustic signal of the lungas recorded by the FDAcleared compatible TytoStethoscope and identifiesrecordings where aspecific abnormal lungsound suggestive of“Crackle” is suspected. Itis not intended to detectother abnormal or normallung sounds. A licensedhealth care professional’sadvice is required tounderstand the meaning ofthe Tyto Insights forCrackles Detection result.Healthcare providersshould consider the deviceresult in conjunction withrecording and otherrelevant patient data. | The “Tyto Insights forWheeze Detection” is anover-the-counter artificialintelligence (AI) enableddecision support softwaresystem used in theevaluation of lung soundsin adults and pediatrics (2years and older). Itautomatically analyses theacoustic signal of the lungas recorded by the FDAcleared compatible TytoStethoscope and identifiesrecordings where a specificabnormal lung soundsuggestive of “Wheeze” issuspected. It is notintended to detect otherabnormal or normal lungsounds. A licensed healthcare professional’s adviceis required to understandthe meaning of the TytoInsights for WheezeDetection result.Healthcare providersshould consider the device | The VRIxp is intendedfor monitoring andrecording lung soundsand automatic detectionof crackles and wheezes.When interpreted byphysicians with generalmedical training andexperience, the VRIxpaids in diagnosis andpatient management. TheVRIxp is intended to beused in healthcarefacilities on adults,adolescents, and/orchildren over the heightof 2 feet 9 inches. | Same intended use asthe primary predicatedevice and similarindication for use as thereference device.Both the predicatedevice and the subjectdevice have the sameintended use in thatboth are intended todetect specific andabnormal breath soundsin the same intendedpatient population(adult and pediatric) bythe same user [HealthCare Professional(HCP)] when self-administered by patientand/or the HCPs. Bothdevices are onlyintended to beinterpreted by HCP andHCP advice is requiredfor the patient tounderstand their result.Both are labeled OTC.The subject device isindicated to detect | |
| Device | Primary Predicate device | Reference device | Summary | |
| result in conjunction withrecording and otherrelevant patient data. | crackles similar to thereference device. | |||
| Type of use | Over-The-Counter Use | Over-The-Counter Use | Prescription use | Same as the primarypredicate device. |
| Intended users | Intended to be used byprofessional users and layusers (18-65 years old). | Intended to be used byprofessional users and layusers (18-65 years old). | Professional users | Same as the primarypredicate device. |
| Intended patientpopulation | Intended for patients of 2years and older | Intended for patients of 2years and older | Adults, adolescents,and/or children over theheight of 2 feet 9 inches. | Same as the primarypredicate device. |
| Intendedenvironment | Non-clinical (home) andclinical | Non-clinical (home) andclinical | Clinical setting | Same as the primarypredicate device |
| Form | Stand-alone softwaresystem | Stand-alone softwaresystem | Hardware (sensors) andsoftware.Use sound sensors tocollect lung sounds viadermal contact, which isthen converted to a visualdisplay. | Same as the primarypredicate device. |
| Device composition | The following modulescompose the Tyto Insightsfor Crackles Detection:• The Tyto Insights forCrackles DetectionApplication Server(APS) | The following modulescompose the Tyto Insightsfor Wheeze Detection:• The Tyto Insights forWheeze DetectionApplication Server(APS) | The system is composedof:1. Sound sensorsdesigned to collectlung sounds viadermal contact withhuman skin2. Digital Collection | The subject deviceshares the samestructural sub-systemsas the primary predicatedevice: ALS, APS, andWBS.The analyticalcomponent of the ALS |
| Device | Primary Predicate device | Reference device | Summary | |
| • The Tyto Insights forCrackles DetectionAlgorithm Server(ALS)• The Tyto Insights forCrackles Detection WebServer (WBS) providesa graphic indicationwhether a crackle isdetected in the recordingIt can be utilized both inpatient and clinicianside. | • The Tyto Insights forWheeze DetectionAlgorithm Server (ALS)• The Tyto Insights forWheeze Detection WebServer (WBS) providesa graphic indicationwhether a wheeze isdetected in therecording It can beutilized both in patientand clinician side. | conversion of analogdata to digital data3. a mobile computerworkstation to assist inprocessing, displaying,and/or storingrecording information | algorithm, differsbetween the proposeddevice and the primarypredicate device.The differences in thesub-systems don't raisenew questions of safetyor effectiveness. | |
| Input | Lung sounds recorded bycompatible TytoStethoscope | Lung sounds recorded bycompatible TytoStethoscope | Sound sensors designedto collect lung sounds viadermal contact withhuman skin | Same as the primarypredicate device |
| Device technologyand operatingprinciple | The recordings are createdby the compatible TytoStethoscope (K181612)and are sent by the third-party point of care app tothe clinician app throughthe telehealth server.The telehealth serversends the set of the lungsound recordings to theTyto Insights for CracklesDetection web serverusing its dedicated API.The telehealth serversubsequently sends the | The recordings are createdby the compatible TytoStethoscope (K181612)and are sent by the third-party point of care app tothe clinician app throughthe telehealth server.The telehealth server sendsthe set of the lung soundrecordings to the TytoInsights for WheezeDetection web server usingits dedicated API.The telehealth serversubsequently sends the link | VRIxv uses sound sensorsto collect lung sounds viadermal contact, which isthen converted to a visualdisplay.During the breathingprocess, the VRlxpdetects lung sounds (i.e.,acoustic energy) andconverts them into avisual display, which canbe viewed via a personalcomputer (PC) monitorand stored for future | The analyticalcomponent of the AIalgorithm differsbetween the proposeddevice and the primarypredicate device.Both the proposeddevice and the primarypredicate device utilizethe CRNN(ConvolutionalRecurrent NeuralNetwork) models, |
{7}------------------------------------------------
{8}------------------------------------------------
{9}------------------------------------------------
{10}------------------------------------------------
{11}------------------------------------------------
| Device | Primary Predicate device | Reference device | Summary | |
|---|---|---|---|---|
| Signal length | The length of the signal isdictated by the recordingprocess of the compatibleStethoscope. The subjectdevice processes therecordings in segments ofup to 12 seconds whilesignals shorter than 6seconds will not beprocessed. | The length of the signal isdictated by the recordingprocess of the compatibleStethoscope. The subjectdevice processes therecordings in segments ofup to 12 seconds whilesignals shorter than 6seconds will not beprocessed. | Was not specified. | Same as the primarypredicate device |
| Data transfer andstorage | The telehealth serversends the list of therecordings (identified by aunique identifier and timestamp) to the TytoInsights for CracklesDetection web serverusing its dedicated API.The server executes theTyto Insights for CracklesDetection which runs thealgorithm and providesthe results. Then the TytoInsights for CracklesDetection web serverinitiates the web userinterface.All the softwaresubsystems (server andstorage) are hosted in thecloud and communicatethrough IP network | The telehealth server sendsthe list of the recordings(identified by a uniqueidentifier and time stamp)to the Tyto Insights forWheeze Detection webserver using its dedicatedAPI. The server executesthe Tyto Insights forWheeze Detection whichruns the algorithm andprovides the results. Thenthe Tyto Insights forWheeze Detection webserver initiates the webuser interface.All the softwaresubsystems (server andstorage) are hosted in thecloud and communicatethrough IP network. | the VRIxp contains asoftware application thatwas designed to providecomputer-aidedrecordings with theelectronic stethoscopeand to store theserecordings along withother appropriate patientinformation. | Same as the primarypredicate |
{12}------------------------------------------------
| Device | Primary Predicate device | Reference device | Summary | |
|---|---|---|---|---|
| Output | • Positive (Crackles suspected),• Negative (Crackles not suspected),• The Tyto Insights for Crackles Detection was not able to analyze the recording | • Positive (wheeze suspected),• Negative (Wheeze not suspected),• The Tyto Insights for Wheeze Detection was not able to analyze the recording | Lung sounds can be viewed collectively as a grayscale image, as well as audibly by sensor. Additionally, the VRlxp has an automated feature for detecting sounds consistent with crackles and wheezes for further clinical evaluation. | Same as the primary predicate device. Indication whether the abnormal lung sound was detected or not and indication in case the device was not able to analyze the recording. |
| Performance | The primary study endpoints and hypotheses were met for the intended patient population, the difference in AUC was in favor of Tyto Insights for Crackles Detection compared to the clinical readers.AUC Tyto Insights for Crackles Detection: 0.97 (0.95–0.98).Sensitivity and Specificity were evaluated:Estimate two-sided 95% Cl):Sensitivity: 0.72 (0.63-0.79)Specificity: 0.99 (0.97 - 1.00) | The primary study endpoints and hypotheses were met for the intended patient population, the difference in AUCs was in favor of Tyto Insights for Wheeze Detection compared to the predicate device and in favor of the non-inferiority claim.AUC Tyto Insights for Wheeze Detection: 0.96 (0.94-0.97)Sensitivity and specificity were evaluated:Estimate two-sided 95% Cl):Sensitivity: 0.5465 [0.4304 - 0.6549] | Unknown | The clinical validation path of the predecessor of the predicate device (TytoCare Lung Sounds Analyzer K221614) that established non inferiority with clinical readers was followed.The study endpoints and hypotheses were met for the intended patient population.The difference in AUC was in favor of Tyto Insights for Crackles Detection compared to the clinical readers. |
{13}------------------------------------------------
| Device | Primary Predicate device | Reference device | Summary | |
|---|---|---|---|---|
| Specificity: 0.9895 [0.9684 – 0.9966]. | to be substantially equivalent when compared with the primary predicate and reference devices. | |||
| User interface for point of care and clinician apps | Web view | Mobile computer workstation is used for displaying the recording information to the clinician | Same as the primary predicate device | |
| Predetermined Change Control Plan (PCCP) | PCCP is included in the 510k submission | PCCP was not included in the 510k submission | PCCP was not included in the 510k submission | The subject device is substantially equivalent to the predicate device, other than the implementation of an authorized Predetermined Change Control Plan (PCCP) to the Tyto Insights for Crackles detection device |
{14}------------------------------------------------
The Tyto Insights for Crackles Detection like its primary predicate device the Tyto Insights for Wheeze Detection (K232237) have the same intended use, in that both are intended to identify recordings where a specific abnormal lung sound is suspected in the same intended patient population (adult and pediatric 2 years and older) by the same user [Health Care Professional (HCP)] when self-administered by patient and/or the HCPs. Both devices are only intended to be interpreted by HCP and HCP advice is required for the patient to understand their result. Both are labeled OTC.
VRIxp by Deep Breeze Ltd. (K091732) was added as a reference device to support the substantial equivalence of the specific lung sound - crackles, that the subject device is indicated to detect.
Both the Tyto Insights for Crackles Detection and its primary predicate device are standalone software systems that deliver the same intended benefit (identify recordings where a specific abnormal lung sound is suspected). For both devices the source of the Lung sounds recordings is the compatible FDA cleared Tyto Stethoscope (K181612).
Similarly, to the primary predicate device, the Tyto Insights for Crackles Detection system is composed of three sub-systems: Application Server (APS), Algorithm Server (ALS) and Web Server (WBS). The APS communicates with the ALS and implements an application programming interface API for communication with the telehealth server. The ALS receives an audio file as input and returns an analysis result of positive or negative regarding whether a Crackles was detected as output.
The code of the APS sub-system is similar to the APS sub-system of the primary predicate device. The only minor adjustments relate to the algorithm's type related parameters. The ALS receives an audio file as input and returns an analysis result of positive or negative regarding whether a crackle was detected as output. The ALS sub-system is composed of an Algorithm and logic wrapper and Interface components. The code of the logic wrapper and Interface component of the ALS component is similar to the ALS component of the primary predicate device, the Algorithm's component is different. The Algorithm of the proposed device is Artificial Intelligence (AI) enabled Algorithm for Crackles detection when the Algorithm of the primary predicate device is AI enabled Algorithm for Wheezes detection. In both devices, the data is being analyzed by AI Machine Learning algorithm to determine the presence of abnormal lung sound in the lungs sound recording. Both the subject device
{15}------------------------------------------------
and the primary predicate device utilize the CRNN (Convolutional Recurrent Neural Network) model for sound event detection, integrating CNNs (Convolutional Neural Networks) and RNNs (Recurrent Neural Networks). Each network is trained based on the target clinical class: Wheeze for the predicate device and Crackles for the proposed device. The question concerning the ability of software AI Algorithm to accurately detect abnormal breath sound is not new regardless of the particular lung sound algorithm model employed. The primary predicate raised similar question. The different AI Algorithm between the proposed device and the primary predicate device does require that the accuracy of the device will be substantiated with valid performance data using acceptable methods.
The WBS provides a graphic indication whether a crackle is detected in the recording. It can be utilized both in the patient and clinician side. The only difference between the WBS sub-systems is in the user interface, which reflects the presence of Crackles instead of Wheeze. The minor differences in the user interface and the software algorithm do not raise new questions of safety and effectiveness, the differences as compared to the primary predicate device did not introduce new usability related critical tasks and didn't impact existing critical tasks.
Standards Conformance
- ANSI AAMI ISO 14971:2019, Medical devices Application of risk . management to medical devices
- ANSI AAMI IEC 62304:2006/A1:2016, Medical device software Software life . cycle processes
- ISO 15223-1 Fourth edition 2021-07, Medical devices Symbols to be used ● with information to be supplied by the manufacturer - Part 1: General requirements.
- ANSI AAMI IEC 62366-1:2015+AMD1:2020 (Consolidated Text) Medical . devices Part 1: Application of usability engineering to medical device.
Performance evaluation:
The Tyto Insights for Crackles Detection was subject to performance evaluation following methodology similar to the ones used to test the primary predicate device. A testing plan was developed and performed to verify that the Tyto Insights for Crackles Detection meets its specifications. The main aspects of the testing plan included:
- SW verification and validation The software including both custom developed . software and OTS software, have been verified and validated and have been
{16}------------------------------------------------
demonstrated to be safe and effective for its intended use. The software documentation level is basic per Content of Premarket Submissions for Device Software Functions, Guidance for Industry and Food and Drug Administration Staff, dated June 14, 2023. All required items related to software as required by FDA guidance for Basic Documentation Level have been included in this submission.
- Cybersecurity- all the applicable information to reflect effective cybersecurity o management and to address the FDA's recommendations described in Cybersecurity in Medical Devices: Refuse to Accept Policy for Cyber Devices and Related Systems Under Section 524B of the FD&C Act, issue date March 2023. Cybersecurity in Medical Devices: Ouality System Considerations and Content of Premarket Submissions, issue date September 2023 and in the other FDA's applicable policies have been included in this submission.
- Performance evaluation retrospective Stand-alone and Clinical performance ● evaluation of the "Tyto Insights for Crackles Detection" device in detecting crackles in the compatible Tyto Stethoscope lung auscultation recordings respective to ground truth and human level performance.
- Human factors validation the minor user interface modifications did not . introduce new critical tasks and didn't impact existing critical tasks. Therefore, no additional human factors validation was required and the human factors testing for the predicate device was applicable.
The performance of the Tyto Insights for Crackles Detection device in detecting crackles in recordings acquired by the compatible Tyto Stethoscope has been evaluated on a retrospective validation dataset. The retrospective validation dataset is composed of recordings obtained from the real-world use of the Tyto Care FDA-cleared compatible Tyto Stethoscope (K181612). 446 recordings (120 Crackles positive and 326 negative), corresponding to the intended patient population of the Tyto Insights for Crackles Detection Software (a total of 445 patients). The demographics of the validation dataset are presented hereunder:
{17}------------------------------------------------
| N=446recordings | |||
|---|---|---|---|
| Age Group (Years) | Positive | Negative | Total |
| 2-12 | 57 (25.67%) | 165 (74.33%) | 222 (50.22%) |
| 12-21 | 10 (18.18%) | 45 (81.82%) | 55 (12.33%) |
| >=21 | 53 (31.36%) | 116 (68.64%) | 169 (37.45%) |
| Gender | Positive | Negative | Total |
| Male | 55 (25.82%) | 158 (74.18%) | 213 (47.6%) |
| Female | 65 (27.89%) | 168 (72.11%) | 233 (52.24%) |
Table 2: Validation data-set demographics
To establish the ground truth, all the recordings were read by three blinded experienced Pulmonologists at random, the binary ground truth was determined by a majority vote of these three Pulmonologists. For the characterization of the stand-alone accuracy, the automated binary result of the software has been compared to ground truth and specificity and sensitivity were calculated. This stand-alone accuracy is presented hereunder in table 3:
| Parameter | Estimate (two-sided 95% CI) |
|---|---|
| Sensitivity (Se) | 0.72 (0.63-0.79) |
| Specificity (Sp) | 0.99 (0.97–1.00) |
| Positive Predictive Value (PPV) | 0.63 (0.4-0.87) |
| Negative Predictive Value(NPV) | 0.99 (0.98-0.99) |
Table 3: The stand-alone accuracy of the Tyto Insights for Crackles Detection
For the characterization of the clinical performance the Area under the Receiver Operating Curve (AUC) for crackles detection by the device was compared to the clinical readers (Physicians non-Pulmonologists). To calculate the AUC the probability score was extracted from the device and compared to a likelihood score that was recorded by the clinical readers independently for every recording.
The primary endpoint was to establish that the lower bound of 95% two-sided CI for the difference in AUCs between the Tyto Insights for Crackles Detection vs. clinical readers is higher than non-inferiority margin of -0.05. The secondary endpoint was the repeatability of the software as compared to the clinical readers.
{18}------------------------------------------------
| Parameter | Estimate 95% two sides CI |
|---|---|
| AUC Clinical readers | 0.77 (0.73–0.8) |
| AUC Tyto Insights for CracklesDetection | 0.97 (0.95–0.98) |
| AUC Tyto Insights for CracklesDetection – clinical readers | 0.2 (0.17–0.23) |
Table 4: The clinical accuracy of the Tyto Insights for Crackles Detection as compared to Clinical readers.
For the indicated patient population the difference in AUC (Tyto Insights for Crackles Detection - Readers; higher values in favor of the device) was 0.2 (0.17-0.23) establishing the non-inferiority (0.17 > margin of -0.05) of the device in detecting crackles. Similar results were also shown within the subgroup analysis, as evidence that the device accuracy is consistent with age and gender groups, different types of crackles, additional abnormal lung sounds and recordings generated by clinician or lay-user. The secondary endpoint was repeatability, the device is characterized by with kappa of 1.0 and agreement of 100% compared to readers repeatability with kappa of 0.42 (0.35 -0.49). In summary, noninferiority of Tyto Insights for Crackles Detection compared to clinical readers was established. Similar effect trend was also shown within the subgroup analysis, as evidence that the device accuracy is consistent with age and gender groups, different types of crackles, additional abnormal lung sounds and recordings generated by clinician or lay-user.
Predetermined Change Control Plan (PCCP)
The subject device is substantially equivalent to the predicate device, other than the implementation of an authorized Predetermined Change Control Plan (PCCP) that specifies anticipated modifications to the Tyto Insights for Crackles detection device.
The table below includes a description of the software modifications that can be made to the algorithms in the device subject to the authorized PCCP as well as a description of the test methods that will be used to support substantial equivalence determination.
{19}------------------------------------------------
| Detailed list of changes | Test Methods | Acceptance criteria to establish substantial equivalence | |
|---|---|---|---|
| 1. Modifications related toquantitative measures ofperformancespecifications:Re-training of the ML modelwith additional data toimprove the performance )ofthe re-trained algorithmcompared to the originaldevice while the same typeand range of input signal isused. | Software verification and validation Clinical performance Validation. Retrospective study ondataset of the compatibleTyto Stethoscoperecordingsthat are representative ofthe intended recordedpatient population (datawill be acquired by thereal-world use of the TytoStethoscope). | Software verification and validation meet the requirements Clinical performance Validation: Stand-alone:Accuracy of device - Sensitivity,Specificity, PPV, NPVCo-Primary endpoints: Sensitivity of the modified device calculated on the new validation dataset compared to the results for the original device Specificity of the modified device calculated on the new validation dataset compared to the results for the original device Success criteria:The success is defined if the LCIfor Se is higher than 0.6279AND LCI for Sp is higher than0.9668. Se and Sp are co-primary endpoints. Meeting bothendpoints is required for themodification to be declaredsuccessfully.Secondary EndpointThe PPV and NPV of the newlytrained algorithm compared to thePPV and the NPV of the originaldevice. | |
| 2. Modifications related toquantitative measures –technical performancespecifications.Modification of datapreprocessing methodologies/data augmentationmethodologies/ Architectureand hyper-parameters toimprove the performance orthe efficiency of the | Software verification and validation Verification testing will be conducted for the improved computational parameters Clinical performance Validation. Retrospective study ondataset of the compatible | Software verification and validation meet the requirements Verification testing meet the requirements Clinical performance Validation: Stand-alone:Accuracy of device – Sensitivity,Specificity, PPV, NPV | |
| computational resources(running time, memoryconsumption and CPUutilization). | Tyto Stethoscoperecordingsthat are representative ofthe intended recordedpatient population (datawill be acquired by thereal-world use of the TytoStethoscope) | Co-Primary endpoints:Sensitivity of the modifieddevice calculated on the newvalidation dataset compared tothe results for the originaldeviceSpecificity of the modifieddevice calculated on the newvalidation dataset compared tothe results for the originaldevice.Success criteria: | |
| The success is defined if the LCIfor Se is higher than 0.6279AND LCI for Sp is higher than0.9668. Se and Sp are co-primary endpoints. Meeting bothendpoints is required for themodification to be declaredsuccessfully. | |||
| Secondary EndpointThe PPV and NPV of thenewly trained algorithmcompared to the PPV and theNPV of the original device. | |||
| 3. Modifications related todevice inputs:Expanding the algorithm toinclude new sources of thesame signal type (different | Software verificationand validationClinical performanceValidation.Retrospective study ondataset of the compatibleStethoscope recordingsthat are representative ofthe intended recordedpatient population (datawill be acquired by thereal-world use of theapplicable Stethoscopemodels). | Software verification andvalidation meet therequirementsClinical performanceValidation: | |
| model of FDA compatibleStethoscope with equivalentaudio acquisitionspecifications).Modification is limited to | Stand-alone:Accuracy of device - Sensitivity,Specificity, PPV, NPV | ||
| expanding to electronicstethoscopes that have FDA510k clearance (for over-the- | Four co-primary endpoints:Stand-Alone performance: | ||
| counter use) at the time thatthe proposed modification ismade. | Accuracy of device – Sensitivity,Specificity, PPV, NPV | ||
| Four Co-Primary endpoints:Sensitivity of the newly trainedalgorithm on the new validationdataset (subset collected only withthe new stethoscope that have therequired FDA 510k clearance at the | |||
| time that the proposed modificationis made) compared to the results forthe original device.• Specificity of the newly trainedalgorithm on the new validationdataset (subset collected only withthe new stethoscope that have therequired FDA 510k clearance at thetime that the proposed modificationis made) compared to the results forthe original device.• Sensitivity of the modified devicecalculated on the new validationdataset (subset collected only withthe currently 510k cleared TytoStethoscope) compared to theresults for the original device• Specificity of the modified devicecalculated on the new validationdataset (subset collected only withthe currently 510k cleared TytoStethoscope) compared to theresults for the original device. | |||
| Secondary Endpoint• The Se, Sp, PPV and NPV of thenewly trained algorithm comparedthe original device on validationdataset combined from casescollected using both new andalready cleared stethoscopes. | |||
| Success criteria:The success is defined if the LCIfor Se is higher than 0.6279AND LCI for Sp is higher than0.9668. Se and Sp are co-primary endpoints. Meeting bothendpoints is required for themodification to be declaredsuccessfully. Se and Sp are co-primary endpoints (a total offour). Success is defined assuccess of all four co-primaryendpoints. | |||
| Secondary EndpointThe Se, Sp, PPV and NPV of thenewly trained algorithm comparedthe original device on validationdataset combined from casescollected using both new and | collected using both new andalready cleared stethoscopes. |
{20}------------------------------------------------
{21}------------------------------------------------
{22}------------------------------------------------
The PCCP includes an algorithm modification protocol describing the verification and validation activities that will support the proposed changes. 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 re-use practices.
Specific test methods are specified in the PCCP to establish substantial equivalence relative to Subject device and include sample size determination, analysis methods, and acceptance criteria. To help ensure validation test datasets are representative of the intended use population, each dataset will meet minimum demographic requirements similarly to the proposed device.
Upon implementation of the change, the change will be communicated to the users and the labeling will be updated to reflect the new device's performance characteristics.
Conclusion
The Tyto Insights for Crackles Detection Software has the same intended use and indication for use as the predicates. The question concerning the ability of software AI Algorithm to accurately detect abnormal breath sound is not new regardless of the particular lung sound algorithm model employed. The primary predicate raised a similar question. Non-inferiority of Tyto Insights for Crackles Detection compared to the clinical readers for intended patient population was established. Thus, we conclude that the Tyto Insights for Crackles Detection is substantially equivalent. i.e. as safe and as effective as the primary predicate device.
§ 868.1900 Diagnostic pulmonary-function interpretation calculator.
(a)
Identification. A diagnostic pulmonary-function interpretation calculator is a device that interprets pulmonary study data to determine clinical significance of pulmonary-function values.(b)
Classification. Class II (performance standards).