Search Results

Deltec Cozmo® Insulin Pump: The Deltec Cozmo® Insulin Pump is a syringe infusion pump designed for Continuous Subcutaneous Insulin Infusion (CSII) for the control of diabetes.
CoZmonitor™ Glucose Monitor: The Deltec CoZmonitor™ Blood Glucose Monitor is available as an accessory to be used solely with the Deltec Cozmo® Insulin Pump. Because the monitor is powered by FreeStyle™, blood sampling can be performed using the finger, forearm, upper arm, thigh, calf or hand. All testing is performed outside of the body (in other words, in-vitro diagnostic use only). Use only with FreeStyle™ Test Strip and FreeStyle™ Control Solution, as other products can give inaccurate results.
Deltec Cozmo® Insulin Pump with CoZmonitor™ Glucose Monitor: The Deltec CoZmonitor™ Blood Glucose Monitoring System is designed to be used only with the Deltec Cozmo® Insulin Pump to help you and your healthcare provider monitor your diabetes management program. When used properly, the monitor provides a quantitative measurement of glucose (or sugar) in whole blood. Because the monitor is powered by FreeStyle™, blood sampling can be performed using the finger, forearm, upper arm, thigh, calf or hand. All testing is performed outside of the body (in other words, in-vitro diagnostic use only). Use only with FreeStyle™ Test Strip and FreeStyle™ Control Solution, as other products can give inaccurate results.

A CoZmonitor™ Glucose Monitor is a modified FreeStyle Tracker™ meter module designed to be coupled with the Deltec Cozmo® Insulin Pump. The Deltec Cozmo® Insulin Infusion Pump with CoZmonitor™ Glucose Monitor is lightweight and easy to use. Users follow their usual practice of blood sampling. The FreeStyle™ test strip is inserted into the CoZmonitor™ Glucose Monitor, which is attached to the Deltec Cozmo® Insulin Pump. The CoZmonitor™ Glucose Monitor will automatically transfer the blood glucose information to the Deltec Cozmo® Pump through an infrared port. If the user's blood glucose level is higher than normal, he or she can go directly to the bolus menu of the pump to correct it without having to manually re-enter the blood glucose value, ensuring more consistent insulin delivery. In addition, blood glucose results are combined with insulin delivery data and stored in the Deltec Cozmo® Pump's history log to provide a complete picture of the user's diabetes management, allowing them to be more proactive in their insulin decisions. The system's central data storage capabilities also allow users to go through fewer steps and therapy calculations – reducing the risk of data or transcription errors.

The provided document is a 510(k) Summary for the Deltec Cozmo™ Insulin Infusion Pump with CoZmonitor™ Glucose Monitor. This type of submission focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than presenting new clinical study data with specific acceptance criteria and performance metrics for the entire system based on strict statistical endpoints.

Based on the document, here's a breakdown of the requested information, focusing on what is explicitly stated and noting what is not applicable or not provided in this regulatory submission format:

1. Table of Acceptance Criteria and Reported Device Performance

The submission does not present formal acceptance criteria (e.g., precision, accuracy ranges) or a detailed statistical performance table for the entire integrated system (Deltec Cozmo® Insulin Pump with CoZmonitor™ Glucose Monitor) in terms of glucose measurement accuracy as one might expect from a de novo IVD submission. Instead, it relies on the predicate device's established performance and focuses on the usability and functional aspects of the integration.

The usability study provides direct performance observations related to the successful completion of tasks:

Note on Glucose Measurement Accuracy: The document states, "Because the monitor is powered by FreeStyle™, blood sampling can be performed using the finger, forearm, upper arm, thigh, calf or hand. All testing is performed outside of the body (in other words, in-vitro diagnostic use only). Use only with FreeStyle™ Test Strip and FreeStyle™ Control Solution, as other products can give inaccurate results." This implies that the glucose measurement accuracy itself relies on the established performance of the FreeStyle™ system, which is the "predicate" for the glucose monitoring component, and not newly established in this submission. The "correct blood glucose readings" in the usability study likely refer to whether the device correctly displayed the expected values for the control solutions, indicating proper function, rather than an independent validation of its analytical accuracy against a gold standard in a clinical setting.

2. Sample Size used for the test set and the data provenance

• Test Set Sample Size: 57 participants (23 females, 32 males, 2 did not provide age).
• Data Provenance: The study appears to be prospective, as participants were "asked to obtain their blood glucose reading" and "provided three control samples." The study was conducted at the location of the manufacturer or a contracted research facility, as it involved "Deltec Cozmo Insulin Pump users" as participants. The country of origin is implicitly the US, given the FDA 510(k) submission.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

• Glucose Monitor Performance Ground Truth: For the "obtain correct blood glucose readings using the low, middle, and high control samples," the ground truth was the known nominal values of the control solutions. This does not involve human experts establishing ground truth in the usual sense for clinical diagnoses.
• Usability Ground Truth: "Correct" performance for usability tasks (e.g., successfully obtaining a reading, attaching the monitor) was likely determined by study protocol observers. The document does not specify the number or qualifications of these observers.

4. Adjudication method for the test set

• For "correct blood glucose readings" with control samples: Adjudication was against the known, expected values of the control solutions. Any reading outside of the expected range would be considered incorrect.
• For usability tasks: The document describes direct observation of task completion. No formal adjudication method (like 2+1 or 3+1 consensus) among multiple expert reviewers is mentioned for the usability study.

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

• No MRMC study was conducted. This device is an insulin pump with an integrated glucose monitor, not an AI-powered diagnostic imaging device that would typically involve human readers interpreting cases. Therefore, this section is not applicable.

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

• Yes, in spirit. The performance of the glucose monitoring component relies on the validated FreeStyle™ technology, which functions in a standalone manner to measure glucose. The integration piece (transferring data to the pump, displaying messages, guiding users to bolus options) is software-driven. The "functional testing" and "software validation" mentioned would cover the standalone performance of these digital components. The usability study itself assesses the user's interaction with the integrated system.

7. The type of ground truth used

• Usability: Direct observation of task completion by trained personnel and participant self-reporting of ease-of-use.
• Glucose Readings (for control samples): Manufacturer-stated known values for low, middle, and high glucose control solutions.

8. The sample size for the training set

• Not applicable / Not provided. This 510(k) submission is for a medical device (insulin pump with an integrated glucose monitor) and does not describe the development or training of an AI algorithm in the contemporary sense. The "software validation" refers to traditional software engineering verification and validation, not machine learning model training. The glucose monitoring technology ("powered by FreeStyle™") would have undergone its own extensive development and validation prior to this integration.

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

• Not applicable. As no AI training set is described, there's no ground truth establishment for a training set in this context.

Ask a specific question about this device