(84 days)
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 and increase accuracy of infusion documentation.
The Spectrum IQ Infusion System with Dose IQ Safety Software is intended to be used by trained healthcare professionals.
The proposed device, Spectrum IO Infusion System with Dose IO Safety Software is a large volume infusion system that provides safe and effective 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 infusion of pharmaceutical drugs, blood and blood products through intravenous administration sets at user selectable rates and volumes, and a software application that allows the generation, configuration and management of a downloadable drug library into the pump.
This document is a 510(k) summary for the Baxter SIGMA Spectrum Infusion Pump with Master Drug Library (K230041). It references a previous 510(k) notification (K222048) for the Spectrum IQ Infusion System with Dose IQ Safety Software as the predicate device.
The provided text focuses on the K230041 submission, which specifically addresses an update to the algorithm and labeling related to the upstream occlusion alarm of the pump. It states that performance testing for the software was completed for this update.
However, the document does not provide detailed information about acceptance criteria or specific study results for this particular update. It primarily asserts substantial equivalence to the predicate device and mentions that the scope of the notification is limited to the occlusion alarm update.
Therefore, many of the requested details about acceptance criteria, study design, sample sizes, ground truth, and expert involvement for the occlusion alarm algorithm validation are not present in the provided text. The information below is limited to what can be inferred or directly stated from the provided document, acknowledging the significant gaps.
Acceptance Criteria and Device Performance
The document does not explicitly state specific quantitative acceptance criteria or detailed reported device performance for the updated upstream occlusion alarm algorithm. It only mentions that "Performance testing for the software was completed" for this update. Without the actual test plan or report, these details cannot be provided.
| Acceptance Criteria (Inferred from general device purpose) | Reported Device Performance (Not explicitly detailed for occlusion alarm update) |
|---|---|
| Ability to detect upstream occlusion events | Performance testing was completed. (Specific metrics not provided) |
| Alarm activation within acceptable timeframes | Not detailed. |
| Avoidance of false positive alarms | Not detailed. |
| Overall safe and effective delivery of fluids, as identified in 21 CFR 880.5725 | The device is deemed substantially equivalent, implying it meets overall safety and effectiveness. |
| Reduction of pump programming errors and increased accuracy of infusion documentation (from Indications for Use) | Addressed by the Dose IQ Safety Software; not specifically evaluated for occlusion alarm update in this document. |
Study Information (Specific to the upstream occlusion alarm algorithm update)
-
Sample size used for the test set and the data provenance:
- Not specified. The document states "Performance testing for the software was completed" but does not give details on the size or nature of the test set used for the occlusion alarm algorithm.
- Data provenance: Not specified. It's likely internal testing data, but the origin (e.g., country, retrospective/prospective) is not mentioned.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not specified. Given it's a software/algorithm update for an occlusion alarm, ground truth would likely be established through controlled experimental setups (e.g., intentionally inducing occlusions) rather than expert review of clinical cases.
-
Adjudication method for the test set:
- Not applicable/Not specified. Adjudication methods like "2+1" are typically for subjective assessments (e.g., image interpretation). For a technical alarm system like an occlusion detector, the "ground truth" is objectively verifiable (e.g., physical presence of an occlusion).
-
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. An MRMC study is not relevant here. This device is an infusion pump with a safety algorithm, not an AI diagnostic tool requiring human interpretation. The "AI" (algorithm) in this context is automating a safety function, not assisting human "readers" with diagnostic tasks.
-
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- Likely yes by implication, but not explicitly stated. The "performance testing for the software" would typically involve evaluating the algorithm's ability to detect occlusions in a controlled environment as a standalone component of the pump's safety system. The entire pump system, which includes the algorithm, operates primarily in a standalone manner to detect occlusions and trigger alarms, though a human operator would then respond to the alarm.
-
The type of ground truth used:
- Not specified, but likely objective, empirical truth. For an occlusion alarm, ground truth would be established by physically creating true occlusion events (and non-occlusion conditions) in a laboratory or simulated environment and measuring the algorithm's response. This is an objective "engineering" ground truth, not pathology, expert consensus, or outcomes data in the traditional sense.
-
The sample size for the training set:
- Not specified. There's no mention of a training set or machine learning model in the context of this specific occlusion alarm algorithm update. It might be a rule-based or signal-processing algorithm rather than a machine learning algorithm requiring a distinct training set. If it is a machine learning algorithm, the training set size is not provided.
-
How the ground truth for the training set was established:
- Not specified. Due to the lack of information on a training set or its nature.
§ 880.5725 Infusion pump.
(a)
Identification. An infusion pump is a device used in a health care facility to pump fluids into a patient in a controlled manner. The device may use a piston pump, a roller pump, or a peristaltic pump and may be powered electrically or mechanically. The device may also operate using a constant force to propel the fluid through a narrow tube which determines the flow rate. The device may include means to detect a fault condition, such as air in, or blockage of, the infusion line and to activate an alarm.(b)
Classification. Class II (performance standards).