Search Results

The t.connect Diabetes Management Application is intended for use by individuals with diabetes mellitus who use Tandem Diabetes Care insulin pumps, their care givers, and their healthcare providers in home and clinical settings. The t:connect Application supports diabetes management through the display and analysis of information downloaded from Tandem Diabetes Care insulin pumps and specified blood glucose meters.

The Tandem t:connect is a web based software system which allows pump users, care professionals to view data retrieved from a Tandem Ambulatory Insulin Pump and Blood Glucose meters. t:connect is comprised of a Data Uploader, a Web Application, a Secure Web and Database Server, and an Admin Web Application.

The t:connect Data Uploader is a client based software application installed on the pump user's personal computer. Using a USB cable connecting the pump to a personal computer, the application uploads pump data to a secure online database server through a secure, encrypted, internet connection. The t.connect Data Uploader also uploads data from a limited set of Blood Glucose meters.

The t:connect Web Application is a reporting application that provides users with the ability to view and print (to PDF) both tabular and graphical reports representing the data uploaded to the online database.

The tconnect Web and Database Server is an application that runs on the secure server. Data displayed on the t:connect Web Application is retrieved from the secure Database "intermediary" providing an extra layer of security and restricted access to the Database Server.

For both the Data Uploader and the Web Application, users are authenticated against a "or both the extir oploads to insure maximum security and secure access to only their data.

The Admin Web Application is an application used to manage accounts. It also allows holders of certain types of accounts to Logon As User to the t.connect Web Application as other users.

t:connect does not have the ability to modify any therapy provided by or parameters internal to the devices it communicates with. t:connect only reads data from these devices.

The major functions of the device include:

• Retrieve History Log data from the pump and transmit to the database server .
• Retrieve History Log data from various BG meters and transmit to the database server .
• Retrieve Diagnostic information from the pump and transmit to the database server .
• Provide a means to view the pump's complete history log
• Retrieve Pump Settings from the pump for easy viewing by the user
• Display Therapy Data textually and graphically .
• Manage User Accounts for retrieving and viewing pump data .
• Generate Notifications for the user
• Print reports to PDF .
• Export data for use in generating customer reports using other software applications
• Provision of user help .
• Administrator functionality related to creation and managing user accounts

Here's a breakdown of the acceptance criteria and study information for the Tandem t:connect Data Management System, based on the provided 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

The provided document primarily focuses on demonstrating substantial equivalence to predicate devices and verifying that design inputs have been met through software testing. It doesn't explicitly state quantitative performance acceptance criteria in terms of clinical metrics (e.g., accuracy, sensitivity, specificity) for a device that makes a diagnosis or has a specific clinical outcome. Instead, the "performance" here refers to the software's functional correctness and usability.

However, we can infer some "acceptance criteria" related to functionality and usability from the description and outline the reported performance based on the provided text.

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

The document describes several types of testing rather than a single "test set" for performance evaluation:

• Software Verification and Validation Testing: This broad category of testing "demonstrates the t:connect Data Management System meets the established design inputs and its intended use." No specific sample size for data used in this testing is given.
• Human Factors Validation Test:
  • Sample Size: 30 participants.
  • Data Provenance: Prospective (participants interacted with the system in a controlled setting). The country of origin is not explicitly stated, but given the company location (San Diego, California), it is likely the US.

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

For the human factors study:

• The "ground truth" here isn't a complex medical diagnosis, but rather the usability and safety of the device's interface and workflow.
• Experts involved in the study:
  • Moderator: Seated in an observation room (no specific qualifications mentioned, but implied to be trained in human factors observation).
  • Certified Diabetes Educator: Provided 30 minutes of training to participants. This individual would be considered an expert in diabetes care and educating patients, which is relevant to the usability of a diabetes management system.
• The 30 participants themselves, divided into 15 healthcare professionals and 15 type 1 diabetics, represent the target users whose feedback and performance define the "ground truth" of usability for this device. Their specific qualifications (e.g., years of experience for healthcare professionals) are not detailed beyond "healthcare professionals."

4. Adjudication Method for the Test Set

• Human Factors Validation Test: The description mentions "The participant's interaction with the system was observed by the moderator... and were added new room so and well as system interaction and which afterou partionmative study did not reveal any safety related issues." This suggests a qualitative observation and assessment approach. There is no mention of a formal adjudicated consensus process (like 2+1 or 3+1) for evaluating interaction, as this is typically used for diagnostic image interpretation. The "ground truth" for human factors is generally established through direct observation of user performance and user feedback, potentially with the moderator or a team synthesizing findings.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done

• No, an MRMC comparative effectiveness study was not done.
  • The provided text describes software verification/validation, human factors testing, and a comparison to predicate devices for substantial equivalence. It does not mention a study evaluating human readers' performance with and without AI assistance for tasks that would require such an analysis (e.g., diagnostic accuracy).
  • The t:connect system is a data management and display tool, not a diagnostic AI system that "assists" human readers in interpreting clinical data to improve their diagnostic accuracy.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• This concept does not directly apply to the t:connect system in the context of this 510(k).
  • The t:connect is a data management application designed to be used by humans (patients, caregivers, healthcare providers) to view and analyze downloaded data. It does not perform autonomous diagnostic or decision-making functions that would typically be evaluated in a "standalone" fashion for an AI algorithm.
  • The software verification and validation tests cover the "algorithm only" aspect in terms of its ability to correctly process, store, and display data according to its design specifications.

7. The Type of Ground Truth Used

• For functional correctness (software V&V): The ground truth is the design input requirements and the intended use of the software. The tests verified that the software performed as designed and met these specifications.
• For usability (human factors): The ground truth is established by user performance and feedback during controlled task scenarios. The absence of safety-related issues in formative studies and successful completion of tasks by representative users served as evidence of acceptable usability.

8. The Sample Size for the Training Set

• The document does not report a "training set" in the context of machine learning or AI algorithm development.
• The t:connect system, as described, is a data management application with defined functionalities rather than a machine learning model that requires a large dataset for training. Its functionality is based on programmed logic and rules, not learned patterns from a training set.

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

• As no "training set" for a machine learning model is mentioned, this question is not applicable to the information provided for the t:connect Data Management System.

Ask a specific question about this device