Search Results
Found 1 results
510(k) Data Aggregation
(189 days)
Personalized Therapy Comfort Settings ("PTCS") is an optional software accessory indicated for use by healthcare professionals and patients, providing recommended non-prescription comfort settings options to support patients when using their compatible Resmed PAP therapy device. The outputs provided by PTCS are optional recommendations and are not required to use the therapy device. It is intended for use in a home or clinical environment.
The subject device, Personalized Therapy Comfort Settings (PTCS), is an AI-enabled software accessory intended for use with compatible ResMed positive airway pressure (PAP) therapy devices. PTCS provides optional, non-prescription comfort setting (e.g., ramp, EPR and AutoSet Response) recommendations within the cleared configuration limits of the connected device. It is designed to support patients in personalizing comfort settings during onboarding and early therapy adaptation.
PTCS functions as a stand-alone, machine learning (ML)-based algorithm that generates individualized comfort setting recommendations based on patient-specific input data. The software outputs a limited set of predefined configuration options that can be reviewed by the patient or healthcare provider. PTCS does not automatically adjust device settings, modify prescribed therapy parameters, create new settings, or alter configuration limits. The recommendations are optional and may be accepted, disregarded, or further adjusted according to personal preference or clinical guidance. PTCS is integrated into compatible ResMed products through an application programming interface (API) and has no direct user interface or end-user access.
The provided FDA 510(k) clearance letter and summary for the Personalized Therapy Comfort Settings (PTCS) device contains only limited information regarding specific acceptance criteria and the detailed study that proves the device meets these criteria. Much of the information requested in your prompt is not explicitly stated in the provided text.
Based on the available text, here's what can be extracted and what remains unknown:
Acceptance Criteria and Reported Device Performance
The document does not explicitly provide a table of acceptance criteria with specific numerical targets. It broadly states that "PTCS met all acceptance criteria and passed testing requirements," and that "Analytical validation demonstrated robust sensitivity, specificity, and generalizability." It also mentions "measurable improvements in patient therapy engagement, including increased nightly usage and additional usage days, while maintaining comparable residual AHI and mask leak."
Without specific numerical acceptance criteria, the table below provides what can be inferred.
| Acceptance Criteria (Inferred from text) | Reported Device Performance |
|---|---|
| Reliable, safe, and effective AI algorithm performance for intended use. | "PTCS met all acceptance criteria and passed testing requirements." "Analytical validation demonstrated robust sensitivity, specificity, and generalizability, thereby establishing safety and effectiveness without prospective clinical study." |
| Consistent and reliable model performance across large, representative datasets. | "Analytical validation confirmed consistent and reliable model performance across large, representative datasets, with consistent treatment-effect results and no identified safety concerns." |
| Measurable improvements in patient therapy engagement. | "Retrospective comparisons of PTCS recommended settings to default configurations showed measurable improvements in patient therapy engagement, including increased nightly usage and additional usage days." |
| Comparable residual AHI (Apnea-Hypopnea Index). | "maintaining comparable residual AHI" |
| Comparable mask leak. | "maintaining comparable residual AHI and mask leak." |
| User-comprehensible decision support without impacting therapy delivery or patient safety. | "Usability evaluations with both clinical and non-clinical users confirmed clear comprehension of PTCS recommendations, appropriate application of judgment, and no critical tasks or use-related risks." |
Study Details
1. Sample sized used for the test set and the data provenance:
* Sample Size: Not specified. The document only mentions "large, representative datasets."
* Data Provenance: "Retrospective, de-identified clinical data representative of the intended population." The country of origin is not specified.
2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
* Not specified. The document does not detail how ground truth for the test set was established.
3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
* Not specified.
4. 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:
* No MRMC comparative effectiveness study involving human readers is mentioned. The study referenced "Usability evaluations with both clinical and non-clinical users confirmed clear comprehension of PTCS recommendations, appropriate application of judgment, and no critical tasks or use-related risks," but this is not a comparative effectiveness study measuring human reader improvement with AI assistance.
5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
* Yes, a standalone performance evaluation was done. The document states: "PTCS functions as a stand-alone, machine learning (ML)-based algorithm that generates individualized comfort setting recommendations based on patient-specific input data." And "Analytical validation demonstrated robust sensitivity, specificity, and generalizability, thereby establishing safety and effectiveness without prospective clinical study." This refers to the algorithm's performance independent of human input during the core recommendation generation.
6. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
* Not explicitly stated for the "ground truth" of the comfort settings. However, for evaluating the impact of the recommendations, "outcomes data" were used, specifically "patient therapy engagement, including increased nightly usage and additional usage days, while maintaining comparable residual AHI and mask leak."
7. The sample size for the training set:
* Not specified. The document mentions "large, representative datasets" for model performance validation but does not differentiate between training and test sets in terms of size.
8. How the ground truth for the training set was established:
* Not specified. The document only mentions that the model is "ML-based" and that "Analytical validation demonstrated robust sensitivity, specificity, and generalizability." It does not provide details on how the training data was curated or how its ground truth was established.
Ask a specific question about this device
Page 1 of 1