K Number
K955015
Date Cleared
1996-03-18

(138 days)

Product Code
Regulation Number
870.1425
Panel
CV
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Model 2010 Plus was designed for the busy Holter environment that places a premium on throughput analysis. This powerful computer provides the "muscle" to analyze even the most difficult recordings quickly and accurately.

Three channels of recorded patient ECG are utilized by the sophisticated arrhythmia analysis program to detect abnormalities. The system provides a number of clinical tools such as individual ECG printouts. trend data analysis, HRV time domain, and full disclosure to enable the clinician to review a patients cardiac performance.

The 2010-Plus gives the flexibility to process Holter recordings prospectively or retrospectively. Prospective interaction lets the technician supervise the analysis by viewing the ECG chronologically and fine-tuning the arrhythmia processor on-line for error-free results. Superimposition, paging, or a combination of both techniques can be used for prospective scanning. Or, the system can automatically analyze the data in approximately 12 minutes. Once the preliminary analysis is complete, the technician can use powerful retrospective tools to validate and edit the report.

Device Description

A Zymed Holter Scanner system consists of a series of interface devices to include a central monitoring computerized unit, a high Resolution Super VGA raster graphics display monitor, a minimum 500 megabites hard disk drive for two 36-hour digital tape readings, Intel Pentium System Board CPU, 8 megabites of RAM, built in math co-processor, cassette tape drive, mouse and keyboard , 3.5 " floppy disk drive, and a laser printer.

Over 1000 final reports can be stored on line, the operating system is DOS compatible, and Holter reports can be transmitted via facsimile systems.

The analysis software package includes standard, pediatric, and AFib to name a few, and the user can program individual custom styles. Other programmable software features include ECG display; scanning speed up to 240 X real time; scanning styles to include retrospective, prospective, superimposition, and paging; auto stops; highlighting; noise algorithm; color schemes, and report formats. The system 3-channel QRS detection/arrhythmia analysis, 3-channel ST segment analysis, full provides disclosure, automatic 2 of 3 channel morphology analysis, customized report software, HRV-Time Domain, and Pacemaker evaluation display.

AI/ML Overview

The provided text is a 510(k) summary for a medical device (Zymed Holter Scanner; Model 2010 Plus) from 1998. It primarily focuses on demonstrating substantial equivalence to a predicate device (Zymed Holter Scanner Model 2010) based on technological characteristics.

Therefore, much of the requested information regarding acceptance criteria and detailed study designs for device performance against specific acceptance criteria is not explicitly present in this 510(k) summary. These summaries, especially from that era, often focused on direct comparison to predicate devices rather than presenting standalone performance studies with pre-defined acceptance criteria in the way a modern regulatory submission might.

However, I can extract and infer some information based on the provided text, while also clearly stating what is not available.


Acceptance Criteria and Study for Zymed Holter Scanner; Model 2010 Plus

The 510(k) summary does not outline specific, quantified acceptance criteria for the new Zymed Holter Scanner Model 2010 Plus in terms of diagnostic performance metrics (e.g., sensitivity, specificity for arrhythmia detection) or a formal clinical study to prove these criteria. Instead, the "acceptance criteria" appear to be implicit in demonstrating substantial equivalence to the predicate device (Zymed Holter Scanner Model 2010) through a comparison of technological characteristics and a general statement about performance.

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria Category (Inferred)Acceptance Criteria (Implicit for Substantial Equivalence)Reported Device Performance
Arrhythmia Analysis CapabilityAbility to perform 3-channel QRS detection/arrhythmia analysis equivalent to the predicate device."Performance between the two systems is almost identical."
ST Segment AnalysisAbility to perform 3-channel ST segment analysis equivalent to the predicate device."Performance between the two systems is almost identical."
Maneuverability/ThroughputDesigned for "busy Holter environment that places a premium on throughput analysis." Ability to quickly and accurately analyze "even the most difficult recordings.""This powerful computer provides the 'muscle' to analyze even the most difficult recordings quickly and accurately."
System FunctionalityAll features of the predicate device (e.g., full disclosure, customized reports, pacemaker evaluation display, data acquisition specs) must be maintained or improved.All features listed as "Same" or "Yes" compared to the predicate, with HRV-Time Domain being an addition.
Safety and Electrical Standards (Implicit, but mentioned for device)Compliance with relevant electrical safety standards."The System meets UL544 and CSA601 compliance and is ETL Listed."

Note: The "acceptance criteria" are inferred based on the intention to demonstrate substantial equivalence, as explicit performance thresholds are not provided in the summary.

Study Used to Prove Device Meets Acceptance Criteria:

The key "study" described is a comparison of technological characteristics between the predicate device (Zymed Holter Scanner Model 2010) and the new device (Zymed Holter Scanner Model 2010 Plus). The conclusion drawn from this comparison is that "Performance between the two systems is almost identical and clearly supports a claim of substantial equivalence."

Specific Information based on the text:

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

  • Sample Size: Not specified. The summary does not describe a clinical test set or patient data used for a performance validation study. The comparison is based on hardware and software features.
  • Data Provenance: Not applicable, as no external data set is described for testing.

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

  • Not applicable. No ground truth establishment by experts for a test set is described.

4. Adjudication method for the test set:

  • Not applicable. No test set requiring adjudication is described.

5. 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 was not described. The device is a Holter scanner, which assists clinicians in analyzing ECG data, but the submission focuses on its equivalence to an existing scanner, not a comparative effectiveness study involving human readers with/without the new scanner. The term "AI" as understood today is also not applicable to this 1998 submission.

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

  • A standalone algorithm performance study with quantitative metrics (like sensitivity/specificity for arrhythmia detection) is not explicitly described in the summary. The statement "Performance between the two systems is almost identical" suggests an internal assessment or reliance on the predicate's established performance, but no details are given.

7. The type of ground truth used:

  • Not applicable, as no specific ground truth for an independent performance study is described. For a Holter scanner, ground truth for arrhythmias would typically be established by expert cardiologist review.

8. The sample size for the training set:

  • Not applicable. This document does not describe a machine learning model requiring a training set in the modern sense. It refers to a "sophisticated arrhythmia analysis program," but details about its development or training data are not provided.

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

  • Not applicable. (See #8)

§ 870.1425 Programmable diagnostic computer.

(a)
Identification. A programmable diagnostic computer is a device that can be programmed to compute various physiologic or blood flow parameters based on the output from one or more electrodes, transducers, or measuring devices; this device includes any associated commercially supplied programs.(b)
Classification. Class II (performance standards).