Search Results

The S9 Greenhills is indicated for the treatment of patients weighing more than 66 lb (30 kg) with obstructive sleep annea (OSA), central and/or mixed apneas, or periodic breathing.

It is intended for home and hospital use.

The hunidifier is intended for single patient use in the home environment and re-use in a hospital/institutional environment.

The S9 Greenhills is similar to the predicate device(s), using a blower based positive pressure system with an integrated heated humidifier and heater controller. The device platform is identical to the S9 VPAP Adapt (K113801) and contains a Micro-processor controlled blower system that generates controlled positive airway pressure between 3-25 cmH2O. The system comprises the flow generator, patient tubing, mask (patient interface) and humidifier.

The S9 Greenhills is a flow generator device designed to provide adaptive servo-ventilation therapy (ASV mode) or (ASVAuto mode) to stabilize a patient's ventilation during sleep. The device continually measures the patient's instantaneous ventilation, and calculates a target ventilation based on to the patient's recent average minute ventilation. It then adjusts the degree of pressure support to servo-control the patilation to at least equal the target ventilation. The same is true for ASVAuto mode except the EPAP is adjusted to address any obstructive apneas detected.

Therapy modes contained in the S9 Greenhills are CPAP, ASV, and ASVAuto and unchanged from the S9 VPAP Adapt (K102586). They are:

• . CPAP mode - the device delivers a continuous positive airway pressure throughout the entire therapy session;
• ASV mode the device automatically adjusts pressure support in response to the . patient's recent average minute ventilation; and
• ASVAuto mode the device automatically adjusts pressure support in response to . the patient's recent average minute ventilation and EPAP level for OSA events.

The functional characteristics of the S9 Greenhills system includes all the clinician and user friendly features of the predicate device which have been verified during usability studies in accordance with IEC 62366 Medical devices - Application of usability engineering to medical devices.

The provided 510(k) summary for the ResMed S9 Greenhills device describes its acceptance criteria and the study conducted to demonstrate its performance. However, it's important to note that this submission focuses on demonstrating substantial equivalence to a predicate device, not on proving overall clinical effectiveness from scratch. Therefore, the "study" is primarily comparative bench testing, and some of the requested information (like human reader performance with/without AI assistance, or expert consensus for ground truth on a large test set) is not applicable or not provided in this type of submission.

Here's a breakdown of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The submission does not provide a direct table of specific numerical acceptance criteria with corresponding performance metrics like a typical clinical trial report. Instead, it states that "All tests (predicate bench testing) confirmed the product met the predetermined acceptance criteria." and "Side-by-Side bench testing was performed to verify that the S9 Greenhills met the predetermined pass/fail requirements of the S9 Greenhills System Specification when compared to the predicate devices."

The key performance characteristics compared, which inherently define the acceptance criteria for substantial equivalence, are:

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Test Set: The document refers to "Side-by-Side bench testing" using "a breathing machine simulates patient breathing patterns." No specific number of simulated patient breathing patterns or cases are quantified.
• Data Provenance: The testing is described as "bench testing" and is therefore synthetic data generated by a breathing machine, not human patient data. There is no country of origin of data or retrospective/prospective designation in the clinical sense, as it is a bench study.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

This information is not applicable and not provided. The "ground truth" for this bench testing is the expected performance of a well-functioning CPAP/ASV device, primarily benchmarked against the predicate device's known performance characteristics. The tests verify engineering specifications, not clinical diagnoses that would require expert consensus.

4. Adjudication Method for the Test Set

Not applicable. This was a bench test against predetermined engineering specifications and comparison to predicate device performance. No human adjudication of clinical outcomes or images was involved.

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

Not applicable. This is a ventilation device, not a diagnostic imaging device using AI to assist human readers. Therefore, an MRMC study comparing AI-assisted vs. non-AI-assisted human reader performance is not relevant to this submission.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done

The device itself is a standalone medical device that delivers therapy based on internal algorithms and sensor inputs. The "bench testing" described evaluates the performance of these algorithms and the device's hardware independently. While not explicitly framed as an "algorithm only" study in the AI context, the bench tests verify the device's automated functioning.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

The "ground truth" for this study is primarily:

• Engineering Specifications: Predetermined pass/fail criteria from the "S9 Greenhills System Specification."
• Predicate Device Performance: The known and established performance of the S9 VPAP Adapt (K113801) is used as the benchmark for comparison, establishing "substantial equivalence." The "breathing machine simulates patient breathing patterns" to test the device's response against expected physiological responses.

8. The Sample Size for the Training Set

Not applicable. This medical device primarily uses control algorithms based on established physiological principles and engineered specifications, not machine learning algorithms that require a distinct "training set" of data in the AI sense.

9. How the Ground Truth for the Training Set was Established

Not applicable, as there is no "training set" in the context of an AI/machine learning algorithm for this device. The device's operating principles and algorithms are based on established medical science for respiratory support.

Ask a specific question about this device

The S9 VPAP Tx is indicated for the treatment and titration of patients with obstructive sleep apnea (OSA), respiratory insufficiency, central or mixed apneas, or periodic breathing. CPAP, S, ST, T and PAC modes are indicated for patients weighing more than 30lb (13 kg); all other modes are indicated for patients weighing more than 66lb (30 kg).

The S9 VPAP Tx is intended to be used in a clinical environment.

The S9 VPAP Tx is similar to the predicate devices VPAP ST-A (K113288), VPAP ST (K102513), S9 VPAP Adapt (K113801) and S8 Aspen (K091947).

The S9 VPAP Tx provides CPAP, Auto-titrating, Bilevel, VAuto and ASV modes to treat OSA and/or respiratory insufficiency, central or mixed apneas or periodic breathing. This is achieved through the use of a micro-processor controlled blower system that generates airway pressures as required to maintain an "air splint" for effective treatment of OSA and/or respiratory insufficiency.

The S9 VPAP Tx system comprises the flow generator, patient tubing, mask (patient interface) and optional H5i humidifier.

The performance and functional characteristics of the S9 VPAP Tx includes all the clinician and user friendly features of the predicate devices, VPAP ST-A (K113288), VPAP ST (K102513), S9 VPAP Adapt (K113801) and S8 Aspen (K091947).

Here's an analysis of the ResMed S9 VPAP Tx 510(k) submission based on the provided text, focusing on the absence of information regarding "acceptance criteria" and a "study that proves the device meets the acceptance criteria" in the traditional sense of an AI/ML medical device submission.

The provided document is a 510(k) summary for the ResMed S9 VPAP Tx, a continuous ventilator. It does NOT describe an AI/ML device or any study that would typically be associated with performance criteria for an algorithmic diagnostic or prognostic tool. Instead, it describes a hardware medical device and its substantial equivalence to previously cleared predicate devices.

Therefore, many of the requested points, such as "acceptance criteria and reported device performance" related to an AI/ML algorithm, sample sizes for test sets, data provenance, expert ground truth, MRMC studies, or standalone algorithm performance, are not applicable to this submission.

The submission focuses on demonstrating compliance with recognized standards for medical electrical equipment and biocompatibility, as well as the substantial equivalence of the new device to existing predicate devices based on intended use, operating principle, technology, and manufacturing process.

Summary regarding Acceptance Criteria and Study for ResMed S9 VPAP Tx (K123511):

This 510(k) submission is for a continuous ventilator (S9 VPAP Tx), which is a hardware medical device, not an AI/ML device. As such, the concept of "acceptance criteria" and "study that proves the device meets the acceptance criteria" within the context of AI/ML performance (e.g., sensitivity, specificity, AUC) is not applicable to this submission.

The document states:

• "Design and Verification activities were performed on the S9 VPAP Tx as a result of the risk analysis and design requirements. All tests confirmed the product met the predetermined acceptance criteria."
• "The S9 VPAP Tx has been tested to appropriate FDA consensus standards and other applicable requirements passing all test protocols."

This indicates that internal design verification and validation activities were conducted based on engineering specifications and compliance with relevant safety and performance standards for hardware medical devices. These are not performance metrics for an AI algorithm.

Addressing the specific points based on the provided text, noting irrelevance where appropriate for an AI/ML context:

1. A table of acceptance criteria and the reported device performance

   • Not Applicable in an AI/ML context. The document does not provide a table of performance metrics (like sensitivity, specificity) for an AI/ML component. The "acceptance criteria" here refers to the passing of various engineering and safety standards.
   • Relevant information from document: The device was tested according to:
     • IEC 60601-1-2:2007 (Electromagnetic compatibility)
     • IEC 60601-1:2005 (General requirements for safety and essential performance)
     • IEC 60601-1-8:2006 (Alarm systems)
     • ISO 10993 series (Biocompatibility)
   • Reported Device Performance: The document states "All tests confirmed the product met the predetermined acceptance criteria" and "passing all test protocols." No specific numerical performance results (e.g., for pressure delivery accuracy, flow rates) are provided in this summary, as these would typically be detailed in the full submission, not the summary.

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

   • Not Applicable. This device is a hardware ventilator. There is no "test set" of patient data for evaluating an AI/ML algorithm as described in the prompt. The testing involved bench testing of the physical device. The document states: "Clinical data for the S9 VPAP Tx is not required as the predicate devices have been subjected to clinical trial requirements or validated patient simulation models have been used during the bench testing phases." This refers to physical models or simulated patient conditions, not patient data for algorithm training/testing.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

   • Not Applicable. This is not an AI/ML device. Ground truth as typically understood for AI/ML validation (e.g., expert consensus on image interpretation) is not relevant here.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

   • Not Applicable. This is not an AI/ML device.

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

   • Not Applicable. This is not an AI/ML device and therefore no MRMC study involving human readers and AI assistance was conducted or would be relevant.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

   • Not Applicable. This device is a hardware ventilator, not an algorithm.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

   • Not Applicable in an AI/ML context. The "ground truth" for this device's testing would be established by the functional specifications and performance standards (e.g., a pressure sensor correctly measures the output pressure, the device delivers a specified flow rate). The "validated patient simulation models" mentioned refer to physical or mathematical models representing patient physiology for bench testing, not clinical ground truth derived from patients for an AI.

8. The sample size for the training set

   • Not Applicable. This device does not have a "training set" in the context of AI/ML.

9. How the ground truth for the training set was established

   • Not Applicable. This device does not have a "training set" or associated ground truth in the context of AI/ML.

In conclusion, the ResMed S9 VPAP Tx 510(k) submission is for a conventional medical device (ventilator) and does not contain information pertinent to the performance criteria or studies typically associated with AI/ML-enabled devices. The "predetermined acceptance criteria" and "test protocols" refer to engineering and safety standards, not AI algorithm performance metrics.

Ask a specific question about this device