The device accepts electrical image signals and produces hard copy images signal source may be digital formatted image data from Fuji Computed Radiography image readers in the case of the FM-DP 2636CR, or unformatted image data from other imaging modalities (e.g. CT, MRI) in the case of the FM-DP 2636. The image signal source of for the FM-DP 2636 may be analog or digital.

The Fuji Medical Dry Imager uses the information in the image signals to control discrete elements in a linear thermal head which writes on the translating Fuji Medical Dry Imaging Film DI-AT, a thermal recording media. Unlike conventional laser imagers and multiformat cameras, there is no laser, cathode ray tube, or optics. Unlike conventional light-sensitive silver halide photographic media, the film requires no dark room, film processor, processing chemicals, water, drainage, or dryer ventilation, produces no chemical waste, and requires no space for chemical storage.

The major components of the imager are the film magazine, transport mechanics, imaging electronics, and thermal head.

AI/ML Overview

The provided text describes a 510(k) summary for a medical image printer, not a device that utilizes AI or machine learning. Therefore, many of the requested categories related to AI performance, ground truth, and expert evaluation are not applicable.

Here's the information that can be extracted or deduced from the provided text:

Acceptance Criteria and Device Performance

The acceptance criteria are implicitly defined by comparison to the predicate device, the Fuji Medical Laser Imager Fuji FL-IM D. The goal is to demonstrate "similar" or "judged adequate" performance for medical image hard copy devices, specifically focusing on:

Acceptance Criteria Category	Acceptance Criteria (Predicate Performance)	Reported Device Performance (Fuji Medical Dry Imager FM-DP 2636CR/FM-DP 2636)
Spatial Frequency Response	Writing density: 10 pixels/mm	Writing density: 11.8 pixels/mm or 300 dpi (better)
	Square wave response (CTF) at 2.5 cycles/mm: >0.93	Square wave response (CTF) at 2.5 cycles/mm: >0.93 (similar)
Gray Scale Resolution	12-bit (exceeds 10-bit CR data)	11-bit (exceeds 10-bit CR data, "judged adequate for the intended use")
Density Uniformity	Wiener Spectrum values (predicate)	Small-area density uniformity (Wiener Spectrum values) is "slightly greater than the predicate" (implies slightly worse, but still one to two orders of magnitude greater than radiographic film-plus-screen mottle).
	Large-area density non-uniformity (predicate)	Large-area density non-uniformity: "judged not diagnostically significant" (similar)
Safety Standard	Complies with UL 1950	Complies with UL 1950 (similar)

Study Details

The document describes a comparison study where the performance of the new device is compared against a predicate device. It is not a clinical study involving patients or diagnosticians interpreting images.

Sample size used for the test set and the data provenance: Not applicable. This is a technical performance comparison of imaging hardware, not a diagnostic test. The "test set" would be the output images generated by the devices, likely using standardized test patterns and medical image data. The provenance of this data is not specified beyond "medical image data".
Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. The "ground truth" here is objective technical performance metrics (spatial resolution, gray scale, density uniformity), not diagnostic accuracy requiring expert interpretation.
Adjudication method for the test set: Not applicable. Performance is measured objectively.
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 done. This device is an image printer, not an AI-powered diagnostic tool.
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Yes, in essence, the study evaluates the standalone technical performance of the imaging device. There is no "human-in-the-loop" component in the performance evaluation described.
The type of ground truth used: Technical specifications and measurements. For example, spatial density is measured in pixels/mm, square wave response is a quantitative value, and gray scale resolution is expressed in bits. Density uniformity is measured by Wiener Spectrum values. These are objective engineering and imaging science metrics.
The sample size for the training set: Not applicable. There is no machine learning or AI involved; thus, no training set is relevant.
How the ground truth for the training set was established: Not applicable, as there is no training set.

Summary

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

OCT 28 1996

K962967

510(k) Summary [as required by 21 CFR 807.92]

Date Prepared [21 CFR 807.92(a)(1)]

July 19, 1996

Submitter's Information [21 CFR 807.92(a)(1)]

Fuji Medical Systems U.S.A., Inc. Post Office Box 120035 Stamford, Connecticut 06912-0035 Telephone: +1 203 973-2664 facsimile: +1 203 973-2777 Contact: Robert A. Uzenoff

Trade Name, Common Name, Classification [21 CFR 807.92(a)(2)]

The device trade names are Fuji Medical Dry Imager FM-DP 2636CR and Fuji Medical Dry Imager FM-DP 2636.

The device common names is medical image printer.

Predicate Device [21 CFR 807.92(a)(3)]

Fuji identifies the predicate device as the Fuji Medical Laser Imager Fuji FL-IM D. FDA assigned the predicate to regulatory class II citing 21 CFR § 892.1750.

FDA's accession number for the premarket notification for the predicate device is K945475. FDA cleared the marketing of the predicate device in a letter dated March 9, 1995.

Description of the Device [21 CFR 807.92(a)(4)]

The major components of the imager are the film magazine, transport mechanics, imaging electronics, and thermal head.

Intended Use [21 CFR 807.92(a)(5)]

Page 1 of 2

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

The indications for use of the Fuji Medical Dry Imagers is the production of hard copy images from medical image data.

Technological Characteristics [21 CFR 807.92(a)(6)}

The subject device, as the predicate, produces monochrome (black-and-white) gray-scale images from medical image data. The media and the technological characteristics are different.

The predicate device uses light sensitive silver halide photographic film as the recording medium. The subject device uses thermal recording film.

The predicate device exposes the film by translating it past a directly modulated scanning laser diode source. The exposed film is subsequently photographically processed to yield a visible image. In the subject device, the media is translated by a directly-modulated discrete-element thin-film linear thermal head. The action of heat on the media produces black dye in the medium, providing the density for the visual image without the wet chemical processing of the predicate.

Performance Data [21 CFR 807.92(b)(1)]

For medical image hard copy devices, important performance characteristics bearing on safety and effectiveness are spatial frequency response, gray scale resolution, and density uniformity.

The subject and predicate devices have similar spatial frequency response characteristics. The pixel size of the subject device (84.7 u) produces a writing density of 11.8 pixels/mm or 300 dpi, compared to the writing density of the predicate, 10 pixels/mm. Square wave response (CTF) at 2.5 cycles/mm for both the subject and predicate devices is greater than 0.93.

The gray scale resolution of the subject (11-bit) and predicate (12-bit) devices both exceed the 10-bit gray scale range of Fuji Computed Radiography image data and judged adequate for the intended use.

The small-area density uniformity of the subject device, measured as Wiener Spectrum values, is slightly greater than the predicate. Both are one to two orders of magnitude greater than that of radiographic filmplus-screen mottle. Large-area density non-uniformity, as with the predicate, is judged not diagnostically significant. Both devices incorporate built-in density test patterns for use in maintaining density uniformity over time and with changes in media.

The devices comply with the UL 1950 Standard for Safety of Information Technology Equipment, Including Electrical Business Equipment.

Conclusion [21 CFR 807.92(b)(3)]

As the predicate, the subject devices have no patient contact. Nor do the subject devices control, monitor, or effect any devices directly connected to or effecting a patient. The images displayed by the subject devices are observed by medical personnel, offering ample opportunity for competent human intervention in the event of a failure.

While technologically different, the performance of the subject device in the areas of spatial and density resolution and density uniformity is similar to that of the predicate. The device, as the predicate, is provided with a test pattern generator and instructions for use to assure a consistent relationship between input signals and output film density (density and contrast).

The subject and predicate share certification of conformance to the UL 1950 Standard for Safety of Information Technology Equipment, Including Electrical Business Equipment.

We conclude that the subject devices are as safe and effective as the predicate device.

Regulation Number and Section

§ 892.2040 Medical image hardcopy device.

(a)
Identification. A medical image hardcopy device is a device that produces a visible printed record of a medical image and associated identification information. Examples include multiformat cameras and laser printers.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.