Search Results
Found 2 results
510(k) Data Aggregation
(215 days)
Accu-Chek Connect Diabetes Management App, Roche Diabetes Care Inc., K150910
The Advisor Pro Platform is intended for the management of diabetes by people with diabetes and their health care providers in order to report, log, track, share, monitor and review their data using the dedicated computer or mobile software. Advisor Pro Platform also enables communication between people with diabetes and their health care providers as well as among health care providers.
The Advisor Pro Platform enables the healthcare provider to use the Advisor Pro Algorithms for treatment recommendations as described below and prescribe the Advisor Pro Bolus Calculator for patient use.
Advisor Pro Algorithm is a decision-support software intended for assisting healthcare professionals in the management of their patients with diabetes who monitor their glucose levels using continuous glucose monitor (CGM) and/or Self-Monitoring Blood Glucose (SMBG) meter; and use any of the following insulin types as their therapy to manage glucose levels via subcutaneous injections or continuous sub-cutaneous insulin infusion (CSII; insulin pump) reported either manually or automatically:
- Long Acting insulins (for injections only)
- Short acting insulins:
- Rapid acting analogs (for injections and insulin pump according to manufacturer indications for use)
- Regular human insulin (for injections only)
The Advisor Pro algorithm is intended to be used for patients with:
- Type 1 diabetes over the age of 6 using an insulin pump or subcutaneous insulin injections.
- Type 2 diabetes over the age of 10 who use subcutaneous insulin injections.
Advisor Pro Algorithm is indicated for use by healthcare professionals when analyzing CGM, SMBG and/or insulin delivery data to generate recommendations for optimizing a patient's insulin treatment plan for basal therapy and/or bolus therapy and/or glucose targets; without considering the full clinical status of a particular patient. Advisor Pro Algorithm does not replace clinical judgment.
The Advisor Pro Bolus Calculator, a component of the DreaMed Diary App, is a diabetes management tool for people with type 1 diabetes above the age of 6 and type 2 diabetes above the age of 10, who use subcutaneous insulin injections therapy (not for pump use). This tool can help calculate their rapid acting analogs for insulin bolus doses based on user-entered blood glucose and/or meal information.
The initial setup of the user's treatment plans, and bolus calculator settings must be performed by a healthcare provider.
The Advisor Pro Platform is a software device that is designed to be a diabetes management platform. It includes the Advisor Pro Algorithm that provides insulin therapy adjustment recommendations to physicians to assist in the management of diabetes for patients with Type 1 diabetes using an insulin pump or multiple daily injections, a continuous glucose monitoring (CGM) system and/or self-management blood glucose meter (SMBG). The Advisor Pro Algorithm also provides insulin therapy adjustment recommendations to physicians to assist in the management of diabetes for patients with Type 2 diabetes on basal-bolus therapy via multiple daily injections (MDI), a continuous glucose monitoring (CGM) system and/or self-management blood glucose meter (SMBG).
The Advisor Pro Algorithm gathers and analyzes information inputted through the Diabetes Management Systems (DMS), which collects biological input information from various diabetes devices and data sources including the DreaMed Diary App. Diabetes device information required and used by Advisor Pro Algorithm includes glucose readings (either CGM sensor readings and/or capillary blood glucose measurements), insulin dosing logs, and meal data during daily routine care.
Following data collection and analysis, the Advisor Pro Algorithm generates results containing summary data and recommendations for adjustments to the patient's insulin therapy parameters, including basal insulin delivery rate(s), insulin to carbohydrate ratio and correction factor (insulin sensitivity) for pump patients and for MDI patients a daily injection plan including a basal plan and either a sliding scale or insulin to carbohydrate ratio and correction factor (insulin sensitivity) for bolus injections. Advisor Pro Algorithm may also advise on personalized diabetes management tips. Results are sent to the Diabetes Management System, which displays results to physicians and a report provided by the algorithm. The physician can approve, reject or change the recommendations and issue the updated treatment plan to the patient.
For MDI patients using rapid acting analogs for insulin bolus dose, the healthcare provider may prescribe the Advisor Pro Bolus Calculator which is integrated in the DreaMed Diary App to aid in calculating their bolus injections.
The provided text describes the DreaMed Advisor Pro Platform, a software device intended for diabetes management. It includes two main components: the Advisor Pro Algorithm (a decision-support tool for healthcare professionals) and the Advisor Pro Bolus Calculator (for patient use).
Here's an analysis of the acceptance criteria and the study proving device performance based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly present a table of quantitative acceptance criteria with corresponding performance metrics like sensitivity, specificity, accuracy, or specific thresholds for agreement. Instead, it focuses on demonstrating that the device's recommendations are "as good as" or "similar to" those of experts.
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Device performs according to stated intended use ("functional testing") | "All test results fell within the pre-determined specification parameters and acceptance criteria." |
| User interface adequately validated for use | "The human factors, design, and labeling information provided in the submission confirm that the user interface has been adequately validated for use per the labeling." (Human factors testing was conducted with intended user populations of patients and healthcare providers) |
| Algorithm recommendations agree with expert recommendations | "The study results show that the recommendations of the Advisor Pro Algorithm Pro for MDI patients Type 1 and Type 2 were significantly as good as the recommendations of expert in the basal and bolus plan with regards to the overall level of agreement in the direction of change. Therefore, the Advisor Pro Algorithm recommendations could be considered similar to those given by healthcare professional who work at leading centers with a wealth of experience in the field of diabetes and who are especially familiar with diabetes technology devices." |
| No new safety or effectiveness concerns | "DreaMed Diabetes believes that the changes as described in this 510(k) submission do not present additional safety or effectiveness concerns for the Advisor Pro Platform including the Advisor Pro Algorithm as well as the Advisor Pro Bolus Calculator and is substantially equivalent to the predicates cited." |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size: The clinical study involved a data set of 17 Type 1 patients and 15 Type 2 patients.
- Data Provenance: The document does not specify the country of origin for the data. It indicates the study was "retrospective clinical study."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
- Number of Experts: A total of 36 experts were involved.
- Qualifications of Experts: The text states these were "healthcare professional who work at leading centers with a wealth of experience in the field of diabetes and who are especially familiar with diabetes technology devices." Specific professional titles (e.g., endocrinologist, specialized diabetes nurse), years of experience, or board certifications are not provided.
4. Adjudication Method for the Test Set
The document states, "Recommendations were compared to examine the level of agreement between one expert to his colleague versus the level of agreement between Advisor Pro Algorithm recommendations and experts." This implies some form of comparison or consensus evaluation, but a specific adjudication method (e.g., 2+1, 3+1) is not explicitly detailed. It seems to be a comparison of agreement levels rather than a formal adjudication process to establish a single ground truth from multiple expert reads for each case.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size
While the study involved multiple experts evaluating cases and comparing their recommendations to the algorithm and to each other, the description doesn't explicitly label it as a formal MRMC comparative effectiveness study in the sense of directly measuring human reader improvement with AI assistance vs. without.
The study aimed to show that the AI's recommendations were "as good as" expert recommendations. It compared "the level of agreement between one expert to his colleague versus the level of agreement between Advisor Pro Algorithm recommendations and experts."
Therefore, it was not a direct MRMC study to quantify how much human readers improve with AI assistance. It was designed to show the equivalence/similarity of the algorithm's recommendations to those of experts. No effect size for human reader improvement with AI assistance is reported because that was not the stated purpose of this particular comparison.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, the study described is a standalone performance evaluation of the Advisor Pro Algorithm's recommendations compared to expert recommendations. The algorithm generated recommendations independently, which were then compared to expert decisions.
7. The Type of Ground Truth Used
The ground truth for the clinical study appears to be expert consensus or expert recommendations for insulin therapy adjustments. The algorithm's recommendations were compared against the recommendations of these experts to establish "level of agreement."
8. The Sample Size for the Training Set
The document does not provide information regarding the sample size used for the training set for the Advisor Pro Algorithm. The clinical study described involved a test set.
9. How the Ground Truth for the Training Set Was Established
The document does not provide information on how the ground truth for the training set was established. The clinical study described focuses on the evaluation of the algorithm's performance on a test set, not on its training data.
Ask a specific question about this device
(305 days)
K150910 (Accu-chek Connect Diabetes Management Application)
The WellDoc BlueStar® System is indicated for use by healthcare providers (HCPs) and their patients - aged 18 years and older - who have type 1 or type 2 diabetes. The BlueStar System is intended to provide secure capture, storage, and transmission of blood glucose data as well as information to aid in diabetes self-management. The BlueStar® System analyzes and reports blood glucose test results and supports medication adherence. In addition, the BlueStar System provides coaching messages (motivational, behavioral, and educational) based on real-time blood glucose values and trends. It includes software intended for use on mobile phones or personal computers in the home or in professional healthcare settings. The software also allows for entry of other diabetes-related healthcare information and provides educational information.
The BlueStar® System is not intended to replace the care provided by a licensed healthcare professional, including prescriptions, diagnosis, or treatment.
The WellDoc BlueStar® Rx System is indicated for use by healthcare providers (HCPs) and their patients - aged 18 years and older - who have type 1 or type 2 diabetes. The BlueStar® Rx System is intended to provide secure capture, storage, and transmission of blood glucose data as well as information to aid in diabetes self-management. The BlueStar® Rx System analyzes and reports blood glucose test results and supports medication adherence. In addition, the BlueStar® Rx System provides coaching messages (motivational, behavioral, and educational) based on real-time blood glucose values and trends. It includes software intended for use on mobile phones or personal computers in the home or in professional healthcare settings. The software also allows for entry of other diabetes-related healthcare information and provides educational information. BlueStar® Rx includes an insulin dose calculator to allow patients to use their prescribed regimen to calculate a dose of insulin for a given amount of carbohydrates and/or blood glucose value.
The BlueStar® Rx System is not intended to replace the care provided by a licensed healthcare professional, including prescriptions, diagnosis, or treatment.
WellDoc® BlueStar® is a stand-alone software system intended to be used by healthcare providers (HCPs) and their patients - aged 18 years and older - who have Type 1 or Type 2 Diabetes. The system is intended to assist Type 1 and Type 2 Diabetes patients to self-manage their disease, as cleared under K162532. Patients receive guidance on diabetes self-management and are encouraged to reach out to their healthcare team when needed. There are two versions of BlueStar® - BlueStar® and BlueStar® Rx - which differ in the availability of the insulin dose calculator, which is restricted to a prescription from a licensed HCP with prescribing authority.
BlueStar® and BlueStar® Rx are implemented through an enterprise such as a health plan or large physician group in tandem with a patient's healthcare team and are comprised of the following applications:
- Enterprise Director Portal
- HCP Service
- Patient Mobile Application
- Patient Web Portal
The Enterprise Director application is used for administrative purposes. The HCP Service houses the Medication Reconciliation feature, which allows for the process of identifying the most accurate list of all medications that the patient is taking, including name, dosage and frequency, by comparing the medical record to an external list of medications obtained from a patient, hospital, or other provider (cleared under K141273).
The Patient Web Portal and the Patient Mobile application have a similar feature set. Data (including blood glucose values, medications, carbohydrates, and activity) entered into these applications are stored in the database and can be retrieved for display in either application. Both applications require the initial web- or mobile-based registration before the patient can access them. On the patient applications (Mobile and Web), BlueStar® and BlueStar® Rx function as an (1) information repository (logbook and Personal Health Record) and (2) diabetes education resource (learning library and health tips) and provide (3) motivational, behavioral, and educational coaching based on real-time blood glucose values and trends, (4) a secure communication system (Message Center), (5) medication information (dose and schedule), and (6) workflow and decision support for healthcare providers.
In BlueStar® Rx, the patient web portal and mobile application also provide an insulin calculator to allow patients to use their prescribed regimen to calculate a dose of insulin for a given amount of carbohydrates and/or blood glucose value. This submission introduces Insulin on Board to be considered in the calculation. The insulin dose calculator is indicated for prescription-use only.
Furthermore, as cleared under K162225, BlueStar® and BlueStar® Rx have the ability to connect to the One Touch Verio Flex Blood Glucose Meter (K150214) via Bluetooth. This will allow users to send data from their meter to the BlueStar® Rx app, which will provide coaching messages (motivational, behavioral, and educational) based on the real-time blood glucose values and trends. The BlueStar Server will also have the ability to transmit data to the OneTouch Reveal Server. With this application, 3-hour delayed continuous glucose monitoring device data can be accessed via API and uploaded for data visualization purposes only. These modifications do not change the fundamental scientific technology of the device.
This document describes the WellDoc BlueStar and BlueStar Rx System, a software system for diabetes self-management. This 510(k) summary (K190013) references K162532 as the predicate device.
1. Table of Acceptance Criteria and Reported Device Performance
The provided document does not contain a table of acceptance criteria with specific performance metrics (e.g., accuracy percentages, error rates). Instead, it states that the device's performance was evaluated through various non-clinical tests to demonstrate substantial equivalence to its predicate.
| Test Category | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| Software | Compliance with FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices (May 11, 2005) for a "Major Level of Concern," and "Guidance for Industry, FDA Reviewers and Compliance on Off-the-Shelf Software Use in Medical Devices." Implied acceptance of proper functionality and reliability of the software. | "Software verification and validation per the FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices (May 11, 2005) for a Major Level of Concern FDA Guidance “Guidance for Industry, FDA Reviewers and Compliance on Off-the-Shelf Software Use in Medical Devices”" (No specific performance metrics are given, but compliance with these guidances indicates successful V&V.) |
| Cybersecurity | Evaluation per FDA Guidance Content of Premarket Submissions for Management of Cybersecurity in Medical Devices (October 2, 2014), specifically addressing Identify, Protect, Detect, Response, and Recover areas. Implied acceptance of robust cybersecurity measures. | "Cybersecurity was evaluated per the FDA Guidance Content of Premarket Submissions for Management of Cybersecurity in Medical Devices Guidance for Industry and Food and Drug Administration Staff, (October 2, 2014). Specifically, addressing the following areas: Identify and Protect, Detect, Response and Recover" (No specific performance metrics are given, but evaluation against this guidance implies successful implementation of cybersecurity.) |
| Human Factors | Adequate validation of the user interface for use per the labeling for both patients and healthcare providers, ensuring safe and effective interaction for the intended user populations. | "Human factors testing was conducted with the intended user populations of patients and healthcare providers. The human factors, design, and labeling information provided in the submission confirm that the user interface has been adequately validated for use per the labeling." (This indicates successful human factors testing that supports the user interface's validity for its intended use.) |
| Insulin on Board | No new questions of safety or effectiveness are raised by the addition of Insulin on Board (IOB) to the insulin dose calculator function. The linear decay model used must be "well-understood and accepted." | "Both the subject device and reference device use a well-understood and accepted linear decay model to model the Insulin on Board decay. There are no notable differences with regards to this technological characteristic to a previously-cleared device." (This confirms the successful application of the IOB feature without raising new safety/effectiveness concerns, aligning with the implied criteria.) |
| CGM Data Access | No new questions with regards to software or cybersecurity due to the ability to receive 3-hour delayed data from continuous glucose monitoring devices via API for data visualization only. Compliance with special controls stated in 21 CFR 862.1350 (related to CGM secondary display regulation). | "The subject device does not use Bluetooth to connect to a Continuous Glucose Meter directly. There are no new questions with regards to software or cybersecurity. De Novo DEN140038 classified CGM secondary display regulation. BlueStar has complied per the special controls stated in 21 CFR 862.1350." (This indicates successful integration of CGM data access without introducing new software or cybersecurity risks, and compliance with relevant regulations.) |
| Overall Equivalence | The differences between the subject and predicate devices (expanded patient population, IOB, CGM data access) do not raise different questions of safety or effectiveness, and performance data demonstrates substantially equivalent performance to the predicate. | "The differences between the predicate and the subject devices do not raise any new or different questions of safety or effectiveness. The WellDoc® BlueStar® and BlueStar® Rx System is substantially equivalent to WellDoc® BlueStar® and BlueStar® Rx System cleared under K162532 with respect to the indications for use, target populations, treatment method, and technological characteristics." "Performance data provided in the submission, including human factors, design, and labeling information, demonstrate substantially equivalent performance to the predicate." (This is the overarching conclusion of the submission, confirming the device meets the substantial equivalence criteria.) |
2. Sample Sizes used for the Test Set and Data Provenance
The document explicitly states: "Clinical Tests: Not Applicable."
Therefore, there is no test set or information regarding sample sizes, country of origin, or retrospective/prospective nature of data for clinical validation of device performance. The performance claim is based on non-clinical testing and substantial equivalence to a predicate device.
For the Human Factors testing, it mentions: "Human factors testing was conducted with the intended user populations of patients and healthcare providers." However, specific sample sizes for these user populations are not provided. The data provenance is not stated (e.g., country of origin, prospective or retrospective).
3. Number of Experts used to Establish the Ground Truth for the Test Set and the Qualifications of those Experts
As there were no clinical tests and no explicit "test set" in the context of clinical performance evaluation (only non-clinical V&V and human factors), there is no information provided about experts establishing ground truth for a test set.
For the human factors study, "healthcare providers" were part of the intended user population, implying experts were involved as participants, but not in establishing ground truth in the traditional sense of diagnostic accuracy. Their qualifications beyond being "healthcare providers" are not detailed.
4. Adjudication Method for the Test Set
Since there was no clinical "test set" requiring ground truth establishment, no adjudication method (e.g., 2+1, 3+1, none) is mentioned.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
No, an MRMC comparative effectiveness study was not done. The document states "Clinical Tests: Not Applicable."
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done
The document primarily focuses on the software system's functionality and its ability to process data, provide coaching messages, and calculate insulin doses. The "Software verification and validation" would assess the algorithm's standalone performance in terms of its intended functions. However, there isn't a separate, explicit study titled "standalone performance study" with specific metrics beyond the general software V&V. The human factors testing then evaluates it with human interaction. The insulin dose calculator performs its calculations algorithmically, and its "performance" is implicitly evaluated through software V&V and comparison to a well-understood linear decay model.
7. The type of ground truth used
For the software validation, the "ground truth" would be the expected output of the algorithms and software logic based on design specifications and established clinical models (e.g., for the insulin dose calculator, the "ground truth" is a correct calculation based on inputted parameters and a 'well-understood and accepted linear decay model'). For cybersecurity, the ground truth is adherence to security best practices and robustness against known vulnerabilities. For human factors, the ground truth is the usability and safety of the interface as determined by user interaction and expert review. No pathology or outcomes data is mentioned as ground truth.
8. The sample size for the training set
The document does not mention a training set or its sample size. This is a software medical device, and the evaluation is primarily focused on verification and validation of its deterministic functions, safety, cybersecurity, and human factors, rather than a machine learning model that would require a distinct training set.
9. How the ground truth for the training set was established
Since no training set is mentioned, the method for establishing its ground truth is not applicable.
Ask a specific question about this device
Page 1 of 1