Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K251636
    Date Cleared
    2025-07-28

    (60 days)

    Product Code
    Regulation Number
    880.5725
    Reference & Predicate Devices
    Predicate For
    N/A
    Why did this record match?
    Reference Devices :

    K173084

    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Spectrum IQ Infusion System with Dose IQ Safety Software is intended to be used for the controlled administration of fluids. These may include pharmaceutical drugs, blood, and blood products. The intended routes of administration consist of the following clinically accepted routes: intravenous, arterial, subcutaneous, or epidural. The Spectrum IQ Infusion System with Dose IQ Safety Software is intended to be used in conjunction with legally marketed and compatible intravenous administration sets and medications provided by the user.

    The Spectrum IQ Infusion System with Dose IQ Safety Software is suitable for a variety of patient care environments such as, but not limited to, hospitals and outpatient care areas.

    The Spectrum IQ Infusion System with Dose IQ Safety Software is intended to reduce operator interaction through guided programming, including a way to automate the programming of infusion parameters and documentation of infusion therapies. This automation is intended to reduce pump programming errors.

    The Spectrum IQ Infusion System with Dose IQ Safety Software is intended to be used by trained healthcare professionals.

    Device Description

    Spectrum IQ is a large volume pump within the Spectrum IQ infusion system used by clinicians at the patient bedside to control the delivery of medications from a bag. The pump moves fluid from the bag to the patient via specified administration sets using a peristaltic pumping action. The pump offers various programmable delivery modes to address specific patient care needs. The delivery modes available to support the patient are determined by how the pump and its associated drug library are configured.

    The pump provides delivery of fluids into a patient in a controlled manner, as identified in 21 CFR 880.5725. The system includes a software controlled, electromechanical pump used for the controlled administration of fluids including pharmaceutical drugs, blood and blood products through administration sets at clinician's selectable rates and volumes.

    The pump is intended for the controlled administration of fluids through the following clinically accepted routes of administration: intravenous, arterial, subcutaneous, and epidural. The pump is intended to be used in conjunction with legally marketed and compatible administration sets, as indicated in the device labeling, and medications provided by the user. The subject device is suitable for patient care in hospitals and outpatient health care facilities.

    Dose IQ is a standalone software application installed on a hospital-provided computing platform and used to create a drug library file. Dose IQ facilitates the generation, configuration, and management of a facility-specific drug library file for dedicated infusion pumps. The drug library file includes customers' dosing limits and additional pump configuration settings. The drug library file is intended to be distributed to all compatible infusion pumps in the hospital.

    This submission includes software design and labeling changes to address the issues leading to recalls Z-0529-2022 and Z-2104-2023.

    AI/ML Overview

    The provided FDA 510(k) clearance letter for the Spectrum IQ Infusion System with Dose IQ Safety Software (K251636) does NOT describe a study involving an AI/Machine Learning algorithm for diagnostic or prognostic purposes, which would typically involve the criteria you've outlined.

    Instead, this device is an infusion pump with safety software, intended to reduce operator interaction through guided programming to minimize "pump programming errors." The summary explicitly states: "This submission includes software design and labeling changes to address the issues leading to recalls Z-0529-2022 and Z-2104-2023." This suggests that the changes are primarily bug fixes, usability improvements, and potentially enhanced safety features rather than the introduction of a novel AI-driven diagnostic tool.

    Therefore, many of the requested details about acceptance criteria and study design (e.g., sample size for test/training sets, data provenance, number of experts, adjudication methods, MRMC studies, standalone performance, type of ground truth) are not applicable to this type of device and submission. These criteria are typically relevant for AI/ML-enabled devices that perform interpretation of medical images, signals, or patient data for diagnostic or prognostic purposes where "ground truth" and human reader performance are central to validation.

    The 510(k) summary focuses on demonstrating substantial equivalence to an existing predicate device (K230041) through a comparison of technological characteristics and non-clinical performance and safety testing.

    Here's how to interpret the provided information in the context of your request:

    Acceptance Criteria and Device Performance (Based on the provided document)

    The document doesn't present specific quantitative acceptance criteria in a table format for a "study" as you might expect for an AI/ML diagnostic device. Instead, "acceptance criteria" for an infusion pump typically refer to meeting defined performance specifications (e.g., volumetric accuracy, alarm thresholds, power consumption, environmental limits) and safety standards (e.g., IEC 60601 series).

    The document states that "Non-clinical testing met all acceptance criteria, demonstrating that the device is safe and effective for its intended use." However, it does not provide the specific quantitative criteria or the numerical reported device performance for each. It assumes these were adequately documented in the full submission to FDA.

    The "performance" described relates to the functional aspects of the pump and software:

    CharacteristicAcceptance Criteria (Implicit from "Subject Device" values)Reported Device Performance (Same as Subject Device values)
    Spectrum IQ Infusion System
    Pumping MechanismLinear peristaltic designLinear peristaltic design
    Modes of DeliveryContinuous, IntermittentContinuous, Intermittent
    Routes of AdministrationIV, Arterial, Subcutaneous, EpiduralIV, Arterial, Subcutaneous, Epidural
    User Interface DisplayColor LCDColor LCD
    AC Power (Input)100-240 VAC, 50 / 60 Hz / 300 mA Max100-240 VAC, 50 / 60 Hz / 300 mA Max
    AC Power (Output)9 VDC/1000 mA, short circuit protected9 VDC/1000 mA, short circuit protected
    Operating Temperature15.6 to 32.2°C (60 to 90°F), 20 to 90% RH non-condensing15.6 to 32.2°C (60 to 90°F), 20 to 90% RH non-condensing
    Atmospheric Pressure66kPa to 102kPa66kPa to 102kPa
    Storage Temperature-10 to +35°C (14 to 95°F), 10 to 90% RH non-condensing-10 to +35°C (14 to 95°F), 10 to 90% RH non-condensing
    Single Fault Condition BolusMax 0.56 mLMax 0.56 mL
    Air-In-Line Detection (> 2.5 cm)Detects >2.5 cm (approx 140 μL) bubblesDetects >2.5 cm (approx 140 μL) bubbles
    Air-In-Line Accumulated Air (Room Temp)Detects >1 mL over 15 min, excluding <10μLDetects >1 mL over 15 min, excluding <10μL
    Air-In-Line Accumulated Air (15.5°C)Detects >1.5 mL over 15 min, excluding <10μLDetects >1.5 mL over 15 min, excluding <10μL
    Low Battery Alarm≤15 minutes of battery power remaining≤15 minutes of battery power remaining
    Flow Rate Range0.5 to 999 mL/hr0.5 to 999 mL/hr
    Low-Flow Continuity (No-Flow Period)Max 90 seconds at 0.5 mL/hrMax 90 seconds at 0.5 mL/hr
    Low-Flow Continuity (Bolus Volume)Not exceed 15 μL over 5 sec at 0.5-1.0 mL/hrNot exceed 15 μL over 5 sec at 0.5-1.0 mL/hr
    Volumetric Accuracy (Non-DEHP, 0.5-1.9 mL/hr)±0.1 mL/hr over ≤ 4 hrs±0.1 mL/hr over ≤ 4 hrs
    Volumetric Accuracy (Non-DEHP, 2.0-999 mL/hr)±5% over ≤ 4 hrs±5% over ≤ 4 hrs
    Dose IQ Safety Software
    Drug Library Capacity (Drugs)5000 drugs5000 drugs
    Drug Library Capacity (Care Areas)32 care areas32 care areas
    Drug Library Capacity (Clinical Advisories)400 clinical advisories400 clinical advisories
    Drug Library Capacity (Modifiers)500 modifiers500 modifiers
    Rate/Dose Limits Fall within Pump Range0.5 to 999 mL/hr0.5 to 999 mL/hr

    Regarding the other specific questions:

    1. Sample sizes used for the test set and data provenance:
      This information is not provided in a way that aligns with AI/ML model validation. The "non-clinical testing" would refer to engineering tests (e.g., bench testing for flow accuracy, alarm functionality, battery life) rather than a dataset for an AI model. Therefore, specific "test set" sample sizes or data provenance (e.g., country, retrospective/prospective) are not detailed because it's not a data-driven AI/ML diagnostic product.

    2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
      Not applicable. This device does not use human experts to establish "ground truth" for a diagnostic "test set" in the way an AI/ML imaging device would. Ground truth for an infusion pump's performance is typically established by engineering measurement standards and physical verification methods, not expert consensus on medical data.

    3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
      Not applicable for this type of device.

    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:
      Not applicable. This device is an infusion pump with safety software, not an AI-assisted diagnostic or interpretation tool that augments human readers. Its purpose is to "reduce operator interaction through guided programming... to reduce pump programming errors," which implies improved safety and efficiency, but not through a "reader" study.

    5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
      Not applicable. The "Dose IQ Safety Software" is part of the "Spectrum IQ Infusion System," which is a physical medical device (an infusion pump). It's inherently a human-in-the-loop device where clinicians interact with it for programming and administration. Its software functionality is integral to the pump's operation.

    6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
      Ground truth for an infusion pump's performance is based on engineering specifications, physical measurements (e.g., volume delivered over time, pressure readings, temperature, electrical parameters), and adherence to recognized industry standards (e.g., IEC 60601-1-8 for alarms). It's not based on expert consensus on clinical findings, pathology, or patient outcomes for diagnostic purposes.

    7. The sample size for the training set:
      Not applicable. This is not an AI/ML device trained on a "dataset." The software is designed and validated against functional requirements and safety standards, not "trained" on a dataset in the machine learning sense. The "training set" for software development would typically be the iterative development and testing cycles against specifications.

    8. How the ground truth for the training set was established:
      Not applicable, as there isn't a "training set" in the AI/ML context. Ground truth in this engineering context is established by design specifications derived from medical necessity, regulatory requirements, and established physical principles.

    In conclusion, the provided FDA clearance letter details the regulatory approval of an infusion pump. While it includes "Dose IQ Safety Software," this software is focused on guided programming and error reduction (likely via a drug library and limits), not on AI/ML-driven diagnostics where the detailed validation criteria you've asked for would apply. The "study" referenced is non-clinical testing against performance and safety requirements, not a clinical trial or AI model validation study.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1