Search Results

The CADD-Sentry Pro"" Medication Safety Software – Administrator allows you to establish a therapy-based protocol library that will be used by the CADD-Sentry Pro Medication Safety Software - Point of Care to program the CADD-Prizm® PCS II Ambulatory Infusion Pump (software revision H or higher).

The CADD-Sentry Pro" Medication Safety Software – Point of Care allows use of a personal computer to send CADD-Sentry Pro "Medication Safety Software --Administrator established therapy-based protocols to the CADD-Prizm® PCS II Ambulatory Infusion Pump (software revision H or higher).

The Smiths Medical MD, Inc. CADD-Sentry Pro™ Medication Safety Software, a software program that operates on commercially available personal computers or similar hardware platforms such as tablets, is designed for pump programming of the CADD-Prizm" PCS II Ambulatory Infusion Pump (software revision H or higher) through a therapy-based protocol database defined by the user. The CADD-Sentry Pro™ Medication Safety Software consists of an Administrator and a Point-of-Care (POC) software module that employs serial communications to send and receive pump information. Both modules are compatible with barcode scanners (or similar input devices) through various PC connections. Barcode format is determined by the user; but is limited to 20 alphanumerical characters. The CADD-Sentry Pro™ Medication Safety Software does not allow duplicative Drug, Protocol or User identification entries.

The CADD-Sentry Pro™ Medication Safety Software - Administrator module allows the user to create and save therapy-based protocols, including pump settings. The Administrator module allows therapy-based protocols to be stored and edited within user-defined protocol libraries. The Administrator user determines POC user access and editing capabilities of these libraries. Other Administrator module features include barcode printing, reports, and sending and receiving pump identification.

The CADD-Sentry Pro™ Medication Safety Software - Point-of-Care module allows the user to send and receive pump settings via serial communication sourced from protocol libraries established using the Administrator module. Additional features include storing and printing pump programs and reports, verifying pump settings to established protocols and viewing history logs in the easier view of the PC monitor.

The provided text describes the CADD-Sentry Pro™ Medication Safety Software. However, it does not contain specific acceptance criteria, a detailed study proving performance against those criteria, or the typical elements of a performance study for AI/machine learning medical devices.

Instead, the document focuses on:

• Device Description: What the software does (Administrator module for creating protocols, Point-of-Care module for sending protocols to a pump).
• Intended Use: How the software is meant to be used.
• Device Comparison: Stating equivalence to a predicate device (itself, indicating a modification or re-submission).
• Summary of Studies: A very brief mention of "Functional Testing" related to software validation and "verification of software controlled programming functions," and a statement that "Human clinical studies were deemed not necessary."
• Conclusion: The software is "safe, effective and performs to established specifications."

Given the information, I can only provide a general response based on the limited details regarding testing.

Acceptance Criteria and Device Performance

The document does not explicitly state numerical acceptance criteria (e.g., sensitivity, specificity, accuracy targets) or measurable device performance metrics in the typical sense of a diagnostic or an AI-driven device. The "acceptance criteria" appear to be met through successful completion of "Functional Testing" and "verification of software controlled programming functions."

Study Details

1. Sample size used for the test set and data provenance:

   • Test Set Sample Size: Not specified. The document only mentions "Test plans associated with software validation, verification of software controlled programming functions, and software related to proper software and pump operation were performed." This implies internal software testing rather than a dataset of patient cases.
   • Data Provenance: Not applicable, as no external data set (e.g., patient data, images, clinical records) was used in the described testing suitable for AI/ML performance evaluation. The testing would have involved the software interacting with the infusion pump.

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

   • Not applicable. The "ground truth" in this context would likely be the expected correct behavior of the software and pump based on engineering specifications and design, rather than expert interpretation of clinical data.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • Not applicable. This type of adjudication method is used for clinical data where human expert disagreement needs resolution. The testing described is functional software testing.

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. The document explicitly states, "Human clinical studies were deemed not necessary to evaluate the safety or effectiveness of the CADD-Sentry Pro™ Medication Safety Software." This software is a pump communication and programming system, not an AI diagnostic or decision support tool for human readers/clinicians, so an MRMC study is not relevant here.

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

   • The "functional testing" performed would be considered standalone in the sense that it evaluated the software's ability to perform its programmed functions (sending/receiving data, managing protocols) without a human interpreting outputs for diagnostic purposes. However, it's not "standalone" in the typical AI sense of providing a clinical assessment without human oversight. The software's function is to facilitate human programming of a pump.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The "ground truth" for this device's functional testing would be the pre-defined functional specifications and expected behavior of the software and the CADD-Prizm pump as determined by its design and engineering requirements.

7. The sample size for the training set:

   • Not applicable. This is not an AI/ML device that requires a training set of data. The software's logic is explicitly programmed.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no training set for this type of software.

Ask a specific question about this device