The COBAS Core II Immunochemistry System is an in vitro diagnostic analyzer intended for enzymeimmunoassay procedures. The instrument performs initial sample measurement and pipetting, timed addition of assay reagents, incubation and shaking of assay tubes, aspiration of used reagents and washing of coated polystyrene assay beads and photometric measurement of the chromogen / enzyme reaction.

AI/ML Overview

This document describes the Roche COBAS Core II Immunochemistry System, an in-vitro diagnostic analyzer, and its substantial equivalence to a predicate device. The information provided is primarily focused on the device's technical characteristics and a limited performance comparison, rather than a clinical study evaluating specific diagnostic accuracy or AI performance. Therefore, many of the requested elements for an acceptance criteria and study summary cannot be fully addressed from the given text.

However, based on the provided text, here's an attempt to extract and interpret the information:

Acceptance Criteria and Device Performance Summary:

The acceptance criteria are implied to be precision and correlation similar to the predicate device (COBAS Core I). The reported performance for the COBAS Core II aligns with these implied criteria.

Acceptance Criteria (Implied)	Reported Device Performance (COBAS Core II)
Precision: Within-run and run-to-run reproducibility (CVs) acceptable, comparable to COBAS Core I.	Precision Study: - Within-run: Seven times in triplicate for internal reference sera and kit control. - Run-to-run: Duplicate in 10 independent runs for the same samples. - Results: Except for one low LH serum, all CVs are "far below 10%". - Conclusion: "The precision range found for the COBAS Core II correspond to the values determined for the COBAS Core I."
Correlation: Strong correlation with the predicate device (COBAS Core I) for patient sample results.	Correlation Study: - Results: For three evaluated assays (LH, FSH, Ferritin), calculated slopes were 1.000 ± 0.014 with a correlation coefficient > 0.995.

Study Details:

Sample sizes used for the test set and the data provenance:
- Precision Study:
  - Four internal reference sera and kit control were tested.
  - Within-run: Samples tested seven times in triplicate (implies 21 measurements per sample/control).
  - Run-to-run: Samples tested in duplicate in 10 independent runs (implies 20 measurements per sample/control).
  - Data Provenance: Not specified, but likely internal lab data given they are "internal reference sera."
- Correlation Study:
  - "137 randomly selected clinical samples" were used.
  - Data Provenance: Not specified, but "clinical samples" suggests patient samples. Country of origin not mentioned. Retrospective or prospective nature not specified.
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
The concept of "ground truth" as typically used in AI studies (e.g., expert consensus on images or pathology) is not applicable here. This study is an analytical performance validation comparing an instrument's readings to those of a predicate instrument and evaluating its precision, not assessing diagnostic accuracy against an independent gold standard interpreted by experts.
Adjudication method for the test set:
Not applicable, as this is an analytical performance study, not a diagnostic accuracy study requiring expert adjudication.
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, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. This is an analytical device validation, not a study involving human readers or AI assistance.
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
Yes, in a sense. The "performance" being evaluated (precision and correlation) is the standalone analytical performance of the COBAS Core II instrument, without human intervention in the result generation process itself, beyond loading samples and controls. However, this is not an "algorithm only" study in the context of AI, but rather an instrument's analytical performance.
The type of ground truth used:
The "ground truth" in this context is the results obtained from the legally marketed predicate device (COBAS Core I) for the correlation study, and the inherent reproducibility of the measurements for the precision study.
The sample size for the training set:
Not applicable. This document describes the validation of a new instrument, not the training of an AI algorithm based on a training set. The instrument's operation is based on established immunochemistry principles and pre-programmed software, not machine learning from a training set.
How the ground truth for the training set was established:
Not applicable for the same reason as point 7.

Summary

{0}------------------------------------------------

K962231

Image /page/0/Picture/1 description: The image shows the logo for Roche Diagnostic Systems. The Roche logo is a stylized hexagon with the word "Roche" inside. The text "Roche Diagnostic Systems" is written in a bold, sans-serif font to the right of the logo. The logo and text are black against a white background.

A Member of the Roche Group

AUG 30 1996

Roche Diagnostic Systems, Inc Branchburg Township 1080 U S. Highway 202 Somerwile, New Jersey 08876-3771

Direct Dial F : x

510(k) Summary

Roche COBAS Core II Immunochemistry System

In accordance with the Safe Medical Devices Act of 1990, a 510(k) summary as outlined in 21 CFR 807.92 is provided herewith.

I. Identification of 510(k) Sponsor:

Roche Diagnostic Systems, Inc. a subsidiary of Hoffmann-La Roched, Inc. Branchburg Township 1080 US Highway 202 Somerville, NJ 08876-3771

510(k) Submission dated June 7, 1996

II. Description of the Device/Statement of Intended Use:

III. Identification of the legally marketed device to which the 510(k) sponsor claims equivalence:

The COBAS Core II Immunochemistry System is substantially equivalent to the currently marketed Roche COBAS Core Immunochemistry System (K921180).

{1}------------------------------------------------

Summary of the technological characteristics of the new device in IV. comparison to those of the predicate.

Similarities to predicate

Both instruments are intended for automation of the currently marketed Roche COBAS . Core assays:

COBAS Core Ferritin (K920829)	COBAS Core TSH (K930776)
COBAS Core Prolatin (K930305)	COBAS Core T4 (K932605)
COBAS Core FSH (K930304)	COBAS Core T3 (K932608)
COBAS Core LH (K930306)	COBAS Core Free T4 (K932607)
COBAS Core IgE (K930890)	COBAS Core Free T3 (K942676)

Both instruments perform the same basic operations, which are, sample handling, ● pipetting, dilution, incubation, bead washing, measurement and data management.
Both instruments utilize the coated bead technology and photometric measurement. ●

Differences to predicate device

The differences in the above mentioned devices consist hardware and software changes in the following areas:

Area of Change	COBAS Core I	COBAS Core II
Barcode Reader	Infra Red Reader	Laser Barcode Scanner
Rack Platform	6 positions for samples andreagents	14 positions for samples andreagents
Test /Sample RacksIdentity	Binary Rack Coding	Barcode Label ID
Rack Locking Device	Racks could be unlockedduring processing	Racks are mechanically lockedwhen in use
Pipetting Syringes	Module on left side ofinstrument	Module moved to the right side ofthe instrument with nomodification to the tube length orsyringes
Operating Software	Internally developedsoftware	Internally developed softwareembedded in Microsoft WindowsNT
Microprocessor	CPU 186	PC- 486 DX33Real Time CPU 186Hard Disk - 350 MBRAM - 16 MBDiskette - 1.44 MB floppy diskdrive 3.5"

{2}------------------------------------------------

Breif discussion of the clinical and nonclinical tests relied on for a V . determination of substantial equivalence:

To demonstrate equivalence in performance characteristics, three representative assays were evaluated for precision and accuracy. The three assays are: COBAS Core LH COBAS Core FSH COBAS Core Ferritin

Precision Study

Four internal reference sera and the kit control were tested to evaluate the within-run and run-to-run reproducibility of the COBAS Core II analyzer. For within-run precision the samples were tested seven times in triplicate. For the run-to run precision the same samples were tested in duplicate in 10 independent runs. Except for one serum with a very low LH level, all CV's are far below 10%. The precision range found for the COBAS Core II correspond to the values determined for the COBAS Core I.

Correlation Study

The correlation of the COBAS Core I and COBAS Core II was evaluated using 137 randomly selected clinical samples run in duplicate on both analyzers. The results of Core I and Core II were compared using linear regression analysis. For all three assays, the calculated slopes were 1.000 ± 0.014 with a correlation coefficient > 0.995.

Regulation Number and Section

§ 862.2160 Discrete photometric chemistry analyzer for clinical use.

(a)
Identification. A discrete photometric chemistry analyzer for clinical use is a device intended to duplicate manual analytical procedures by performing automatically various steps such as pipetting, preparing filtrates, heating, and measuring color intensity. This device is intended for use in conjunction with certain materials to measure a variety of analytes. Different models of the device incorporate various instrumentation such as micro analysis apparatus, double beam, single, or dual channel photometers, and bichromatic 2-wavelength photometers. Some models of the device may include reagent-containing components that may also serve as reaction units.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 862.9.