(57 days)
DreaMed Advisor Pro is a decision-support software intended for assisting healthcare professionals in the management of patients with Type 1 diabetes who:
• use insulin pumps as their insulin delivery therapy;
• monitor their glucose levels using CGM and/or self-management blood glucose meter;
• are above the age of 6 and under 65 years old; and
• use rapid acting U-100 insulin analogs in their pump.
DreaMed Advisor Pro is indicated for use by healthcare professionals when analyzing continuous glucose monitoring (CGM), self-monitoring blood glucose (SMBG) and pump data to generate recommendations for optimizing a patient's insulin pump settings for basal rate, carbohydrate ratio (CR), and correction factor (CF); without considering the full clinical status of a particular patient. DreaMed Advisor Pro does not replace clinical judgement.
DreaMed Advisor Pro is a software device that is designed to provide insulin therapy adjustment recommendations to physicians to assist in the management of diabetes for patients with Type 1 diabetes using an insulin pump, a continuous glucose monitoring (CGM) system and self-management blood glucose meter (SMBG).
The DreaMed Advisor Pro gathers and analyzes information inputted through qualified Diabetes Management Systems (DMS), which collects biological input information from various diabetes devices. Diabetes device information required and used by DreaMed Advisor Pro 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 DreaMed Advisor Pro 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). DreaMed Advisor Pro may also advise behavioral changes. Results are sent to a qualified Diabetes Management Systems, which displays results to physicians and a report provided by DreaMed Diabetes. The physician can approve, reject or change the recommendations and issue the updated treatment plan to the patient.
The DreaMed Advisor Pro is a decision-support software for healthcare professionals managing Type 1 diabetes patients. It generates recommendations for optimizing insulin pump settings based on continuous glucose monitoring (CGM), self-monitoring blood glucose (SMBG), and pump data.
Here's an analysis of the acceptance criteria and study information provided:
1. Table of Acceptance Criteria and Reported Device Performance
The document doesn't explicitly state "acceptance criteria" in a quantitative performance table but discusses equivalence to a predicate device and clinical validity compared to expert recommendations.
| Feature / Characteristic | Acceptance Criteria (Implied / Compared to Predicate) | Reported Device Performance and Assessment |
|---|---|---|
| Agreement with Predicate Device (Retrospective Test) | - High level of agreement with DEN170043 when comparing recommendations over similar datasets. | - Retrospective tests showed a "high level of agreement" with the predicate device (DEN170043) for recommendations on similar data sets. - When sufficient data was available with simulated-SMBG, DreaMed Advisor Pro made similar recommendations to the predicate. - When insufficient data, DreaMed Advisor Pro did not recommend changes, mirroring the predicate's behavior without enough CGM data. |
| Clinical Validity vs. Experts (SMBG data alone) | - Recommendations in basal, CR, and CF plan (regarding direction of change) should be significantly as good as experts. | - An additional retrospective clinical study found that DreaMed Advisor Pro's recommendations (when based on SMBG data alone) were "significantly as good as the recommendations of expert in the basal, CR and in the CF plan with regards to the direction of change." - This suggests the Advisor Pro's recommendations are similar to those from experienced healthcare professionals. |
| Minimum Data Points for Analysis (Subject Device) | - Minimum of 12 valid days. | - Achieved: Minimum of 12 valid days required. A valid day consists of: - Minimum glucose data: At least 67% of CGM sensor readings per day (e.g., 192 samples for 5-minute interval sensor, 64 samples for 15-minute interval sensor), OR - At least 4 BG measurements a day separated by at least 160 minutes. - Minimum insulin pump data: At least 1 basal rate record and 1 bolus record. - Insulin pump settings at analysis within acceptable ranges: Basal rate (0.025-3 u/h), CR (3-70gr/u), CF (10-280gr/u), Bolus calculator targets (≤ 150 mg/dl). |
| Input Data Specifications (Accuracy) | - CGM sensors with regulatory approval demonstrating accuracy below MARD of 15%. | - Achieved: Uses CGM with regulatory approval showing accuracy below MARD of 15%. |
| Non-clinical tests (Design Validation, Human Factors, Software) | - Performs according to stated intended use. - Human Factors validation documented per FDA Guidance. - All software test results fall within pre-determined specifications and acceptance criteria. - Special controls implemented and validated. | - Achieved: Design validation testing and human factors study results confirmed performance per intended use. - Human Factors validation documented according to FDA Guidance (February 3, 2016). - Software evaluation included functional testing; all results within pre-determined specification parameters and acceptance criteria. - Special controls implemented and validated per DreaMed's software test plan. |
2. Sample Size and Data Provenance for the Test Set
- Sample Size for Clinical Study: 15 patients.
- Data Provenance: Retrospective clinical study. The document does not explicitly state the country of origin of the data.
3. Number of Experts and Qualifications
- Number of Experts: 17 experts.
- Qualifications of Experts: They are described as "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 specific number of years of experience or physician titles (e.g., radiologist) are provided, as this is for diabetes management.
4. Adjudication Method
- Adjudication Method: The clinical study compared recommendations to examine the "level of agreement between one expert to his colleague (total of 136 pairs) versus the level of agreement between DreaMed Advisor Pro recommendations and experts (total of 17 pairs)." This implies a comparison against individual expert opinions, rather than a formal consensus-based adjudication method like 2+1 or 3+1. Each expert's recommendation was likely treated as a reference point for comparison.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- The document describes a comparative study where the device's recommendations were compared against experts. It's a "retrospective clinical study" involving 17 experts reviewing data for 15 patients. While it involves multiple readers (experts) and multiple cases (patients), it focuses on the agreement between the AI and human experts, and experts among themselves, rather than a direct MRMC study to quantify how much human readers improve with AI assistance vs. without AI assistance. The study concluded that the Advisor Pro's recommendations were "significantly as good as the recommendations of expert," implying equivalence, but not necessarily an "improvement effect size" in a human-in-the-loop scenario.
6. Standalone Performance
- Yes, a standalone performance assessment was done. The "Additional retrospective clinical study" evaluated the DreaMed Advisor Pro's recommendations based on SMBG data alone and compared these recommendations directly to those of human experts. This demonstrates the algorithm's performance in generating recommendations without real-time human input or modification.
7. Type of Ground Truth Used
- Ground Truth: Expert consensus (or rather, expert recommendations serving as a reference) was used. The study compared the device's recommendations against the recommendations made by the 17 human experts.
8. Sample Size for the Training Set
- The document does not explicitly state the sample size used for the training set. It focuses on the validation of the device.
9. How the Ground Truth for the Training Set was Established
- The document does not describe how the ground truth for any potential training set was established. The clinical study mentioned is a validation study comparing the device's output to expert recommendations, not a description of the training data or its ground truth establishment.
§ 862.1358 Insulin therapy adjustment device.
(a)
Identification. An insulin therapy adjustment device is a device intended to incorporate biological inputs, including glucose measurement data from a continuous glucose monitor, to recommend insulin therapy adjustments as an aid in optimizing insulin therapy regimens for patients with diabetes mellitus.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include the following:
(i) A complete description of the required data inputs, including timeframe over which data inputs must be collected and number of data points required for accurate recommendations;
(ii) A complete description of the types of device outputs and insulin therapy adjustment recommendations, including how the recommendations are generated;
(iii) Robust data demonstrating the clinical validity of the device outputs and insulin therapy recommendations;
(iv) A robust assessment of all input data specifications, including accuracy requirements for continuous glucose monitors and other devices generating data inputs, to ensure accurate and reliable therapy adjustment recommendations. This assessment must include adequate clinical justification for each specification;
(v) A detailed strategy to ensure secure and reliable means of data transmission to and from the device, including data integrity checks, accuracy checks, reliability checks, and security measures;
(vi) Robust data demonstrating that users can understand and appropriately interpret recommendations generated by the device; and
(vii) An appropriate mitigation strategy to minimize the occurrence of dosing recommendation errors, and to mitigate the risk to patients of any residual dosing recommendation errors to a clinically acceptable level.
(2) The device must not be intended for use in implementing automated insulin dosing.
(3) Your 21 CFR 809.10(b) labeling must include:
(i) The identification of specific insulin formulations that have been demonstrated to be compatible with use of the device;
(ii) A detailed description of the specifications of compatible devices that provide acceptable input data (e.g., continuous glucose monitors, insulin pumps) used to provide accurate and reliable therapy adjustment recommendations;
(iii) A detailed description of all types of required data (inputs) and dosing recommendations (outputs) that are provided by the device; and
(iv) A description of device limitations, and instructions to prevent possible disruption of accurate therapy adjustment recommendations (e.g., time zone changes due to travel).