← Product Code [MNR](/submissions/AN/subpart-c%E2%80%94monitoring-devices/MNR) · K252628 # CASSIE (K252628) _Wesper, Inc. · MNR · Apr 27, 2026 · Anesthesiology · SESE_ **Canonical URL:** https://fda.innolitics.com/submissions/AN/subpart-c%E2%80%94monitoring-devices/MNR/K252628 ## Device Facts - **Applicant:** Wesper, Inc. - **Product Code:** [MNR](/submissions/AN/subpart-c%E2%80%94monitoring-devices/MNR.md) - **Decision Date:** Apr 27, 2026 - **Decision:** SESE - **Submission Type:** Traditional - **Regulation:** 21 CFR 868.2375 - **Device Class:** Class 2 - **Review Panel:** Anesthesiology - **Attributes:** AI/ML, Software as a Medical Device ## Indications for Use Cassie is a Software as a Medical Device (SaMD) that automatically analyzes, displays, and summarizes Photoplethysmogram (PPG) data, typically collected during sleep using compatible devices. Cassie is intended for use by and by order of a healthcare professional to aid in the evaluation of sleep disorders, where it may inform or drive clinical management for adults. The Wesper Apnea Hypopnea Index (wAHI) and other outputs presented when oximeter data is available are intended to aid healthcare professionals in diagnosis and management of sleep disordered breathing. The Cassie output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. ## Device Story Cassie is an AI-based SaMD that processes physiological data from sleeping patients to assist clinicians in diagnosing sleep disorders. Input consists of PPG waveforms and optional SpO2 data from standard EDF files collected via compatible pulse oximeters (e.g., Nonin 8000 series). The system operates as a cloud-based web application accessed via API. Machine learning models analyze cardiopulmonary markers to classify sleep stages (Wake, Light, Deep, REM) and detect respiratory events. Outputs include sleep architecture metrics (TST, WASO, sleep efficiency) and respiratory indices (wAHI, ODI). Clinicians review these outputs and raw physiological channels through a web interface to inform clinical management. The device is intended for use by or under the order of healthcare professionals; it does not replace professional clinical judgment. ## Clinical Evidence Multi-site retrospective validation study using 474 independent test subjects from the National Sleep Research Resource. Performance compared against gold-standard PSG scored by certified technologists. Primary endpoints: AHI correlation (r=0.96, 95% CI 0.95-0.96) and TST correlation (r=0.86, 95% CI 0.84-0.89). OSA severity classification (AHI ≥5) showed 97.3% sensitivity and 86.1% specificity (3% desaturation). Results confirm performance is consistent with predicate device ranges. ## Technological Characteristics Cloud-based SaMD utilizing machine learning models to analyze cardiopulmonary physiological markers. Inputs: PPG and optional SpO2 (EDF format). Outputs: Sleep architecture, respiratory events, and scalar metrics (AHI, TST, etc.). Operates on general-purpose computing platforms via API. Software developed per IEC 62304:2006. Cybersecurity controls implemented per FDA guidance. ## Regulatory Identification A breathing (ventilatory) frequency monitor is a device intended to measure or monitor a patient's respiratory rate. The device may provide an audible or visible alarm when the respiratory rate, averaged over time, is outside operator settable alarm limits. This device does not include the apnea monitor classified in § 868.2377. ## Predicate Devices - SleepImage System ([K182618](/device/K182618.md)) ## Reference Devices - Aurora ([K231355](/device/K231355.md)) - Nonin, Inc. Model 8000 Series ([K092101](/device/K092101.md)) ## Submission Summary (Full Text) > This content was OCRed from public FDA records by [Innolitics](https://innolitics.com). If you use, quote, summarize, crawl, or train on this content, cite Innolitics at https://innolitics.com. > > Innolitics is a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices, including [a 510(k)](https://innolitics.com/services/510ks/), [a De Novo](https://innolitics.com/services/regulatory/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/). {0} FDA U.S. FOOD & DRUG ADMINISTRATION April 27, 2026 Wesper Inc. Ciaran Lynch Regulatory Manager 44-02 11th St Suite 610 Long Island City, New York 11101 Re: K252628 Trade/Device Name: Cassie Regulation Number: 21 CFR 868.2375 Regulation Name: Breathing Frequency Monitor Regulatory Class: Class II Product Code: MNR Dated: August 17, 2025 Received: August 20, 2025 Dear Ciaran Lynch: 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. 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" U.S. Food & Drug Administration 10903 New Hampshire Avenue Silver Spring, MD 20993 www.fda.gov {1} K252628 - Ciaran Lynch Page 2 (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 Management System Regulation (QMSR) (21 CFR Part 820), which includes, but is not limited to, ISO 13485 clause 7.3 (Design controls), ISO 13485 clause 8.3 (Nonconforming product), ISO 13485 clause 8.5.2 (Corrective action), and ISO 13485 clause 8.5.3 (Preventative action). Please note that regardless of whether a change requires premarket review, the QMSR requires device manufacturers to review and approve changes to device design and production (ISO 13485 clause 7.3 and ISO 13485 clause 7.5) and document changes and approvals in the Medical Device File (ISO 13485 clause 4.2.3). 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 (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-reporting-combination-products); good manufacturing practice requirements as set forth in the Quality Management System Regulation (QMSR) (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 Rule"). 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-devices/device-advice-comprehensive-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-devices/medical-device-safety/medical-device-reporting-mdr-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/medical-devices/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-devices/device-advice-comprehensive-regulatory-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). {2} K252628 - Ciaran Lynch Page 3 Sincerely, Binoy J. Mathews -S Digitally signed by Binoy J. Mathews -S Date: 2026.04.27 16:44:26 -04'00" For Rachana Visaria Assistant Director DHT1C: Division of Anesthesia, Respiratory, and Sleep 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} DEPARTMENT OF HEALTH AND HUMAN SERVICES Food and Drug Administration Indications for Use Form Approved: OMB No. 0910-0120 Expiration Date: 07/31/2026 See PRA Statement below. 510(k) Number (if known) K252628 Device Name Cassie Indications for Use (Describe) Cassie is a Software as a Medical Device (SaMD) that automatically analyzes, displays, and summarizes Photoplethysmogram (PPG) data, typically collected during sleep using compatible devices. Cassie is intended for use by and by order of a healthcare professional to aid in the evaluation of sleep disorders, where it may inform or drive clinical management for adults. The Wesper Apnea Hypopnea Index (wAHI) and other outputs presented when oximeter data is available are intended to aid healthcare professionals in diagnosis and management of sleep disordered breathing. The Cassie output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. 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." FORM FDA 3881 (8/23) PEC Publishing Services (301) 443-6740 {4} WESPER 510(k) Summary # Submitter information Company Name: Wesper Inc. Company Address: 234 Fifth Avenue, New York, NY 10001 Company Contact: Ciaran Lynch, RA Manager Contact Phone: (805) 490-5223 Contact Email: Ciaran@wesper.co Date Prepared: April 26, 2026 # Device Identification Name of Device: Cassie Classification Name: Breathing frequency monitor Regulation Number: 21 CFR 868.2375 Product Code: MNR Device Classification: Class II Classification Panel: Anesthesiology # Predicate Device Device Trade Name: SleepImage System – K182618 Device Manufacturer: SleepImage Regulation Number: 21 CFR 868.2375 Product Code: MNR Device Classification: Class II # Reference Device Device Trade Name: Aurora – K231355 Device Manufacturer: EnsoData Regulation Number: 21 CFR 868.2375 Product Code: MNR Device Classification: Class II General Summary for Cassie {5} General Summary for Cassie Page 2 # WESPER ## Device Description Cassie is an Assistive Artificial Intelligence system that processes physiological data from a sleeping patient to produce clinical metrics which aid sleep professionals in diagnosing and managing sleep disorders. Cassie operates as a standalone software process requiring internet connectivity and AWS authorized credentials. Cassie is designed to run in a remote secure location, where it can be accessed by authorized users through an API. Cassie is designed to process Physiological data from a standard EDF file. Device outputs are generated in the form of a standard JSON file. Data collected from a sleeping patient include: 1. Photoplethysmogram (PPG) - Optical waveform 2. SpO2 - Continuous reading of blood oxygen saturation from a peripheral site. This is an optional input. At the output, it produces the following high-level classes of clinical metrics: 1. Sleep Architecture - an epoch-by-epoch classification of the patient's sleep stage (Wake, Light, Deep or REM) 2. Respiratory Events - instances throughout the recording time where a respiratory disturbance was registered. This output will only apply if SpO2 was provided at the input. 3. Summary Information - a distillation of the above vectorized data into physiological channels and scalar metrics, including AHI, TST, WASO, etc. ![img-0.jpeg](img-0.jpeg) {6} General Summary for Cassie Page 3 # WESPER Cassie accepts the inputs, feeds them into the model, and the model's predictions set the output's labels. Cassie is currently compatible with Nonin, Inc. Model 8000 Series (K092101). # Indications for Use Cassie is a Software as a Medical Device (SaMD) that automatically analyzes, displays, and summarizes Photoplethysmogram (PPG) data, typically collected during sleep using compatible devices. Cassie is intended for use by and by order of a healthcare professional to aid in the evaluation of sleep disorders, where it may inform or drive clinical management for adults. The Wesper Apnea Hypopnea Index (wAHI) and other outputs presented when oximeter data is available are intended to aid healthcare professionals in diagnosis and management of sleep disordered breathing. The Cassie output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. # Comparison to Predicate Device ## Indication for Use comparison The subject and predicate device have the same intended use: specifically, for analyzing, summarizing and displaying data recorded from a sleeping patient. Both the subject and predicate device rely on a PPG (Photoplethysmogram) input to calculate sleep-related parameters, such as Total Sleep Time, Sleep Efficiency and others, and have the option to also receive an SpO2 channel and produce sleep-disordered breathing information. Similarly, both the subject and predicate device produce physiological channels and a standard output, which can be reviewed and modified by healthcare professionals using commonly available, industry-standard tools. The output from both the subject and predicate device is not intended to be interpreted or clinical action taken without the consultation of a qualified healthcare professional. The subject device is intended for use in adults only, which is a subset of the population indicated for the predicate device (adults, adolescents, and children). {7} General Summary for Cassie Page 4 # WESPER The minor differences between the predicate and subject devices do not alter the intended use of the device, nor do they raise different questions of safety and effectiveness of the subject device relative to the predicate. ## Technological Characteristics Analysis of PPG to aid healthcare professionals in the diagnosis or management of sleep disorders is the technological foundation for both devices. In addition, the two devices share the following technological principles: a. Input in the form of a continuous PPG and, if available, SpO2 from a sleeping patient b. Outputs in the form of Sleep Architecture information, such as sleep staging, total sleep time, etc., and/or Sleep-Disordered Breathing information, such as individually detected respiratory events, and wAHI (Wesper Apnea-Hypopnea Index) c. Cloud-based web applications, using trained machine learning models running on general-purpose computing platforms d. Analysis of cardiopulmonary physiological markers e. Graphical and numerical outputs (including physiological channels) for clinical review and verification, leveraging industry-accepted methods Specifically, the two devices employ similar methods, techniques, technological implementations, and physiological drivers to realize all the above. The following technological differences exist between the subject and predicate device: ## Model internals Both devices primarily utilize cardiopulmonary information to infer physiological states related to sleep architecture and sleep-disordered breathing. While the subject device implements machine learning techniques, compared to automatic analysis of the predicate, both systems process the same underlying physiological data to achieve equivalent clinical endpoints and performance. {8} General Summary for Cassie Page 5 # WESPER ## Performance comparison Key performance metrics against gold standard Polysomnography (PSG) were compared to the predicate demonstrating similar performance (See table 1 and Performance Data section for more details). For the key performance endpoints (AHI, TST & Sleep staging) correlation to PSG was similar to the predicate and similar 510(k) cleared devices. These comparisons show that any technological differences do not negatively impact safety or effectiveness. ## Summary Based on the analysis performed above, the technological differences between the subject and predicate devices do not raise different questions of safety or effectiveness. ## Substantial Equivalence Table Following is a summary comparison of the two devices in tabular format: {9} WESPER Table 1: Substantial Equivalence Comparison of the Subject, Predicate and Reference Device | | Cassie (Subject) – K252628 | Sleeplmage (Predicate) – K182618 | Aurora (Reference) – K231355 | Similarities and Differences | | --- | --- | --- | --- | --- | | Classification Regulation | 21 CFR 868.2375 | 21 CFR 868.2375 | 21 CFR 868.2375 | Same. | | Product Code | MNR | MNR | MNR | Same. | | Indications for Use | Cassie is a Software as a Medical Device (SaMD) that automatically analyzes, displays, and summarizes Photoplethysmogram (PPG) data, typically collected during sleep using compatible devices. Cassie is intended for use by and by order of a healthcare professional to aid in the evaluation of sleep disorders, where it may inform or drive clinical management for adults. The Wesper Apnea Hypopnea Index (wAHI) and other outputs presented when oximeter data is available are intended to aid healthcare professionals in diagnosis | The Sleeplmage System is Software as a Medical Device (SaMD) that establishes sleep quality. The Sleeplmage System analyzes, displays and summarizes Electrocardiogram (ECG) or Plethysmogram (PLETH) data, typically collected during sleep, that is intended for use by or on the order of a Healthcare Professional to aid in the evaluation of sleep disorders, where it may inform or drive clinical management for children, adolescents and adults. The Sleeplmage Apnea Hypopnea Index (sAHI), presented when oximeter data is available, is | Aurora is a Software as a Medical Device (SaMD) that establishes sleep quality. Aurora automatically analyzes, displays, and summarizes Photoplethysmogram (PPG) data collected during sleep. Aurora is intended for use by and by order of a healthcare professional to aid in the diagnosis of sleep disorders including sleep apnea in adults. The Aurora output, including automatically detected respiratory events and parameters, may be displayed and edited by a qualified healthcare professional. | Similar. The subject device's outputs and inputs are a subset of the predicate device's outputs and inputs. | General Summary for Cassie {10} WESPER | | and management of sleep disordered breathing. The Cassie output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. | intended to aid healthcare professionals in diagnosis and management of sleep disordered breathing. The SleepImage System output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. | The Aurora output is not intended to be interpreted or clinical action taken without consultation of a qualified healthcare professional. Aurora is not intended for use with polysomnography devices. | | | --- | --- | --- | --- | --- | | Patient Population | Adults | Children, Adolescents and Adults | Adults | The subject device's patient population is a subset of the predicate's patient population. | | Environment of Use | N/A (Software) | N/A (Software) | N/A (Software) | Same. | | Method of Access | General-purpose computing platform with internet connection | General-purpose computing platform with internet connection | General-purpose computing platform with internet connection | Same. | | Principle of Operation | Cloud-based web application as a Medical Device (SaMD) | Cloud-based web application as a Medical Device (SaMD) | Cloud-based web application as a Medical Device (SaMD) | Same. | | Input Source | PPG optionally oximetry | PPG optionally oximetry | EDF signal files from compatible FDA-cleared medical purpose pulse oximeters PPG devices | Similar. | General Summary for Cassie {11} General Summary for Cassie Page 8 # WESPER | Signals Analyzed | PPG, optionally oximetry | ECG or PPG; optionally oximetry. Accelerometer | PPG and oximetry | Similar. The subject device's analyzed signals are a subset of the predicate device's analyzed signals. | | --- | --- | --- | --- | --- | | Algorithm Used | Machine Learning analysis of cardiopulmonary physiological markers | Automatic analysis of cardiopulmonary physiological markers | Machine Learning analysis of cardiopulmonary physiological markers | Similar. The algorithms are substantially equivalent. While some technological differences exist, clinical performance testing and safety profiling confirm that these differences do not raise new or different questions of safety and effectiveness relative to the predicate. | | Interoperable Data Format | EDF | EDF, ASCII | EDF | Similar. | {12} WESPER | Output/Diagnostic Parameters | Total Sleep Time (TST) Sleep efficiency (SE) Sleep Latency (SL) Wake after sleep onset (WASO) Sleep Stages (Wake, Light, Deep REM) Respiratory Disturbance Index (RDI) Oxygen Desaturation Index (ODI) Apnea/Hypopnea Index (AHI) Obstructive Apnea/Hypopnea Index (oAHI) Central Apnea/Hypopnea Index (cAHI) Raw data physiological channels for review | Total Sleep Time (TST) Sleep efficiency (SE) Sleep Latency (SL) Wake after sleep onset (WASO) Sleep Stages (Wake, Light, Deep REM) Respiratory Disturbance Index (RDI) Oxygen Desaturation Index (ODI) Apnea/Hypopnea Index (AHI) Obstructive Apnea/Hypopnea Index (oAHI) Central Apnea/Hypopnea Index (cAHI) Sleep Quality index (SQI) Sleep Fragmentation Index (SFI) Periodicity Raw data physiological channels for review | Total Sleep Time (TST) Sleep efficiency (SE) Sleep Latency (SL) Wake after sleep onset (WASO) Sleep Stages (Wake, Light, Deep REM) Oxygen Desaturation Index (ODI) Apnea/Hypopnea Index (AHI) Raw data physiological channels for review | Similar. The subject device's output/diagnostic parameters are a subset of the predicate device's output/diagnostic parameters. | | --- | --- | --- | --- | --- | | Performance | Key performance endpoints (such as AHI, TST, sleep staging etc.) were compared to the gold standard (PSG) using established clinical methodologies. | Key performance endpoints (such as AHI, TST, sleep staging etc.) were compared to the gold standard (PSG) using established clinical methodologies. | Key performance endpoints (such as AHI, TST, sleep staging etc.) were compared to the gold standard (PSG) using established clinical methodologies. | Similar. The subject device has a similar performance when compared to the gold standard and is in line with the predicate device and similar 510(k) cleared devices. | General Summary for Cassie {13} WESPER | AHI [Events/hr] | Correlation to PSG | Correlation to PSG | Correlation to PSG | Similar. The subject device has a similar performance when compared to the gold standard and is in line with the reference device. | | --- | --- | --- | --- | --- | | Performance (compared to gold standard) | 0.96 (95% CI 0.95-0.96) Slope β1: 1.094 (1.0233, 1.1646) Intercept β0: -0.2119 (-1.168, 0.7442) Mean Difference (MD): 1.55 (1.02, 2.08) Upper Limit (ULOA): 13.02 (12.12, 13.93) Lower Limit (LLOA): -9.92 (-10.82, -9.02) | ρ = 0.89 | Not publicly available Slope β1: 0.936 (0.853, 1.033) Intercept β0: 0.023 (-1.185, 1.122) Mean Difference (MD): 1.000 (0.630, 1.367) Upper Limit (ULOA): 14.575 (13.779, 15.363) Lower Limit (LLOA): -12.574 (-13.371, -11.786) | | | AHI 1A cutoff > 5 | Sensitivity: 97.3% Specificity: 86.1% | Not publicly available | Sensitivity: 92.6% Specificity: 71.6% | Similar. The subject device has a similar performance when compared to the gold standard and is in line with the reference device. | | AHI 1B cutoff > 5 | Sensitivity: 97.8% Specificity: 89.9% | Not publicly available | Sensitivity: 89.4% Specificity: 76.8% | Similar. The subject device has a similar performance when compared to the gold standard and is in line with the reference device. | | TST [Minutes] Performance (compared to gold standard) | Correlation to PSG r = 0.86 Slope β1: 1.0449 (0.9964, 1.0935) Intercept β0: -9.5919 (-28.3213, 9.1375) | Not publicly available | Correlation to PSG Not publicly available Slope β1: 1.159 (1.035, 1.318) Intercept β0: -41.7 (-94.56, -0.3) | Similar. The subject device has a similar performance when compared to the gold standard and is in line with the reference device. | General Summary for Cassie Page 10 {14} WESPER | | Mean Difference (MD): 6.61 (2.66, 10.55) **Upper Limit (ULOA):** 92.48 (85.73, 99.22) **Lower Limit (LLOA):** - 79.26 (-86.01, -72.52) | | Mean Difference (MD): - 5.58 (-7.92, -3.54) **Upper Limit (ULOA):** 68.7 (63.6, 72.96) **Lower Limit (LLOA):** -79.8 (-84.84, -75.54) | | | --- | --- | --- | --- | --- | | **Data review** | The clinician can view raw data for interpretation, adjust epochs, and write clinical notes. Clinicians may review and utilize the results to make recommendations for further testing, referral, and/or therapy. | The clinician can view raw data for interpretation, adjust epochs, and write clinical notes. Clinicians may review and utilize the results to make recommendations for further testing, referral, and/or therapy. | The clinician can view raw data for interpretation, adjust epochs, and customize sleep reports for the patient. Clinicians may review and utilize the results to make recommendations for further testing, referral, and/or therapy. | Same. | | **Intended Users** | Used by or on the order of physicians, trained technicians, or other healthcare professionals to evaluate sleep disorders where it may inform or drive clinical management taking into consideration other factors that normally are considered for clinical management of sleep disorders. | Used by or on the order of physicians, trained technicians, or other healthcare professionals to evaluate sleep disorders where it may inform or drive clinical management taking into consideration other factors that normally are considered for clinical management of sleep disorders. | Used by or on the order of physicians, trained technicians, or other healthcare professionals to evaluate sleep disorders where it may inform or drive clinical management taking into consideration other factors that normally are considered for clinical management of sleep disorders. | Same. | General Summary for Cassie Page 11 {15} WESPER # Performance Data ## Software Testing: The Cassie system is documentation level (Basic). Non-clinical testing involved validating the software requirements and ensuring proper cloud service configuration. Software development and non-clinical testing have been performed for the Cassie system in accordance with IEC 62304:2006 Medical Device Software – Software Life-Cycle Processes. Software verification and validation testing were conducted, and documentation was provided based on Guidance Documents for Industry and FDA Staff, "Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions," issued June 27, 2025, "Content of Premarket Submissions for Device Software Functions" Issued June 14, 2023, and "Software as a Medical Device (SaMD): Clinical Evaluation" (IMDRF/SaMD WG/N41FINAL:2017). Taken together, the non-clinical software validation and cybersecurity assessments concluded that the Cassie system performs as intended, does not introduce new questions of safety or effectiveness, and is as safe and effective for the intended use. ## Clinical Validation: A clinical validation study was conducted to assess the accuracy of the Cassie photoplethysmography (PPG)-based home sleep testing system relative to polysomnography (PSG) and in the context of the performance characteristics of the legally marketed predicate device, the SleepImage System (K182618). The study utilized PSG recordings as the reference standard and was collected under standardized, AASM-compliant conditions with centralized scoring by certified technologists. **Study Design:** Multi-site retrospective validation study using PSG sleep study datasets available from the National Sleep Research Resource for a similar indication for use, using PSG to assess sleep architecture, breathing patterns, and associated physiological signals. **Study dataset:** 474 independent test subjects (25% holdout from 1,898 total studies) were taken from datasets available by the National Sleep Research Resource, supported by the General Summary for Cassie {16} General Summary for Cassie Page 13 # WESPER National Heart, Lung, and Blood Institute (NHLBI). The validation protocol, inclusion criteria, and statistical analysis plan were pre-specified, and analyses were conducted blinded to the reference PSG. The dataset provides PSG recordings from diverse patient populations, enabling a comparison between Wesper's PPG technology and the reference PSG device across a wide range of demographics and sleep apnea severities. Reference Standard: Polysomnography (PSG) scored per AASM guidelines by certified technologists. Clinical Performance: - Apnea-Hypopnea Index (AHI): Tests were performed to compare the agreement of the automated Apnea-Hypopnea Index (AHI) output from the subject device against full Polysomnography (PSG). The evaluation demonstrated a Pearson correlation of 0.96 (95% CI: 0.95-0.96) and Bland-Altman limits of agreement, -9.92 to 13.02 events/hr, relative to the reference standard. - Total Sleep Time (TST): Tests were performed to evaluate the Total Sleep Time (TST) Pearson correlation 0.86 (95% CI: 0.84-0.89) and Bland-Altman limits of agreement (LOA), -79.26 to 92.48 min, relative to the reference standard.. - OSA Severity Classification: Sensitivity and specificity for detection of obstructive sleep apnea at AHI thresholds ≥5 & ≥15 events/hour under 3% and 4% desaturation criteria, using polysomnography as the reference standard. The classification accuracy of Cassie is presented in the table below: {17} WESPER | Apnea Hypopnea Index (wAHI vs PSG) | | | | | | --- | --- | --- | --- | --- | | AHI Cutoff | Desaturation | Sample Size (n) | Sensitivity | Specificity | | AHI thresholds ≥5 | 3% | 474 | 97.3% (95.0%, 98.5%) | 86.1% (78.3%, 91.4%) | | | 4% | 474 | 97.8% (95.4%, 99.0%) | 89.9% (84.8%, 93.3%) | | AHI thresholds ≥15 | 3% | 474 | 94.8% (90.9%, 97.1%) | 94.7% (91.2%, 96.8%) | | | 4% | 474 | 96.5% (92.0%, 98.5%) | 94.9% (92.0%, 96.8%) | Additional performance details (secondary endpoints): Additional endpoints listed below include Sleep Staging, Wake After Sleep Onset (WASO), Oxygen Desaturation Index (ODI), and Sleep Fragmentation Index (SFI). The performance remained consistent across age, sex, race, and BMI subgroups and none of the endpoint results exceeded predicate performance ranges or raised new questions of safety or effectiveness. | Sleep Staging | | | | | --- | --- | --- | --- | | Category | Sample Size (n) | Sensitivity | Specificity | | Wake | 109,763 | 75.2% (74.9%, 75.4%) | 93.9% (93.8%, 93.9%) | | Light Non-REM | 244,239 | 66.4% (66.2%, 66.6%) | 79.9% (79.7%, 80.1%) | | Deep Non-REM | 34,432 | 73.6% (73.1%, 74.0%) | 86.5% (86.4%, 86.6%) | | REM | 63,165 | 73.9% (73.6%, 74.2%) | 95.9% (95.8%, 95.9%) | | Sleep Profile and Oxygen Saturation | | | | | | | --- | --- | --- | --- | --- | --- | | Category | Deming Regression | | Bland-Altman | | | | | Slope β1 | Intercept β0 | Mean Difference (MD) | Upper Limit (ULOA) | Lower Limit (LLOA) | | wAHI (3%) [events/hour] | 1.094 (1.0233, 1.1646) | -0.2119 (-1.168, 0.7442) | 1.55 (1.02, 2.08) | 13.02 (12.12, 13.93) | -9.92 (-10.82, -9.02) | | wAHI (4%) [events/hour] | 1.0814 (1.0171, 1.1457) | 0.4539 (-0.089, 0.997) | 1.51 (1.09, 1.92) | 10.55 (9.84, 11.26) | -7.54 (-8.25, -6.83) | General Summary for Cassie {18} © WESPER | Total Sleep Time [minutes] | 1.0449 (0.9964, 1.0935) | -9.5919 (-28.321, 9.137) | 6.61 (2.66, 10.55) | 92.48 (85.73, 99.22) | -79.26 (-86.01, -72.52) | | --- | --- | --- | --- | --- | --- | | Sleep Fragmentation Index [events/hour] | 0.7935 (0.5870, 1.0000) | 2.6487 (1.389, 3.9086) | 1.21 (-0.85, 1.57) | 9.07 (8.45, 9.68) | -6.65 (-7.26, -6.03) | | Wake After Sleep Onset [minutes] | 0.9077 (0.777, 1.038) | -4.332 (-14.669, 6.003) | -13.24 (-17.39, -9.10) | 76.95 (69.86, 84.03) | -103.43 (-110.52, -96.35) | | Oxygen Desaturation Index [events/hours] | 1.074 (0.971, 1.178) | -2.2034 (-4.568, 0.161) | -0.03 (-0.79, -0.72) | 16.39 (15.10, 17.68) | -16.45 (-17.74, -15.16) | # Conclusion Based on the results of the nonclinical and clinical testing, and the comparison to the predicate device, Cassie is substantially equivalent to the predicate device SleepImage (K182618). Cassie is as safe, as effective, and performs as well as the legally marketed predicate device for its intended use. General Summary for Cassie --- **Source:** [https://fda.innolitics.com/submissions/AN/subpart-c%E2%80%94monitoring-devices/MNR/K252628](https://fda.innolitics.com/submissions/AN/subpart-c%E2%80%94monitoring-devices/MNR/K252628) **Published by [Innolitics](https://innolitics.com)** — a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices. If you're preparing [a 510(k)](https://innolitics.com/services/510ks/), [a De Novo](https://innolitics.com/services/regulatory/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/), [get in touch](https://innolitics.com/contact). **Cite:** Innolitics at https://innolitics.com