(105 days)
The BD FlowSmart™ set is intended for the subcutaneous infusion of medication, including insulin, from an external infusion pump.
The BD FlowSmart™ set is a subcutaneous administration set intended to interface with commercially available infusion pumps with suitable connections. The infusion set features a flexible perforated catheter perpendicular to an adhesive patch and detachable tubing. The tubing is connected on one end to the medication reservoir of an external infusion pump and on the other end to the patient, attached to the skin by an adhesive base. It is a single-use sterile device.
This document is a 510(k) summary for the BD FlowSmart™ Set, an infusion set. It does not describe an AI/ML powered medical device, therefore, the requested information regarding acceptance criteria, study details, and AI-specific metrics cannot be extracted.
The document primarily focuses on demonstrating "substantial equivalence" of the BD FlowSmart™ Set to existing predicate devices (Medtronic's MiniMed Quickset and Ypsomed's Orbit Infusion Set). This process typically involves comparing the new device's technological characteristics, intended use, and performance data (often non-clinical testing) to those already cleared devices, rather than establishing acceptance criteria for a novel AI algorithm.
Here's what can be extracted, albeit not directly addressing an AI device's acceptance criteria:
-
Acceptance Criteria and Reported Device Performance (General, not AI-specific): The document states that "Results of these tests were found to be acceptable and demonstrated that the BD FlowSmart™ set met requirements for its intended use and is as safe and as effective as its predicate devices." However, it does not provide a table of specific quantitative acceptance criteria or detailed performance metrics. The testing performed includes broad categories:
- Biocompatibility: ISO 109993: Biological evaluation of medical devices
- Functional Performance: ISO 10555-1: Sterile single use intravascular catheters, ISO ID26: Medical Electrical Equipment. Part 2: Particular requirements for the safety of infusion pumps, ISO 1135-1: Sterilization of healthcare products
- Animal Testing: United States Department of Agriculture (USDA) standards, US Animal Welfare Act
- Human Factor Evaluations: IEC 62366:2007
The "reported device performance" is a general statement of meeting requirements and being "as safe and as effective," without specific numerical values.
-
Sample Size and Data Provenance for Test Set: This information is not provided for individual tests. The document refers to "non-clinical testing," "animal testing," and "human factor evaluations," but does not specify the sample sizes within these tests or the data provenance (e.g., country of origin, retrospective/prospective). It's likely that for a device like an infusion set, "test set" wouldn't mean a clinical data set in the same way it would for an AI algorithm.
-
Number of Experts and Qualifications for Ground Truth: Not applicable. The "ground truth" for an infusion set would be adherence to scientific and engineering standards (e.g., material properties, flow rates, sterility) rather than expert interpretation of a clinical dataset.
-
Adjudication Method: Not applicable. This concept is typically relevant to clinical studies involving human interpretation or disagreements, which is not the primary focus here.
-
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study: Not applicable. This type of study assesses the impact of AI on human reader performance, which is not relevant for an infusion set.
-
Standalone (Algorithm Only) Performance: Not applicable. The device is a physical infusion set, not an AI algorithm.
-
Type of Ground Truth Used: The "ground truth" for this device would be established by validated scientific and engineering methods, regulatory standards (like ISO standards), and animal models for biocompatibility and function. It's not based on expert consensus, pathology, or outcomes data in the context of an AI system.
-
Sample Size for Training Set: Not applicable. The device is not an AI algorithm that requires a training set.
-
How Ground Truth for Training Set was Established: Not applicable.
In summary, the provided document details the submission for a medical device (infusion set) under the 510(k) pathway, focusing on demonstrating substantial equivalence to predicate devices through non-clinical testing. It does not involve an AI/ML component, and therefore, the requested information pertinent to AI device evaluation criteria is not present.
§ 880.5440 Intravascular administration set.
(a)
Identification. An intravascular administration set is a device used to administer fluids from a container to a patient's vascular system through a needle or catheter inserted into a vein. The device may include the needle or catheter, tubing, a flow regulator, a drip chamber, an infusion line filter, an I.V. set stopcock, fluid delivery tubing, connectors between parts of the set, a side tube with a cap to serve as an injection site, and a hollow spike to penetrate and connect the tubing to an I.V. bag or other infusion fluid container.(b)
Classification. Class II (special controls). The special control for pharmacy compounding systems within this classification is the FDA guidance document entitled “Class II Special Controls Guidance Document: Pharmacy Compounding Systems; Final Guidance for Industry and FDA Reviewers.” Pharmacy compounding systems classified within the intravascular administration set are exempt from the premarket notification procedures in subpart E of this part and subject to the limitations in § 880.9.