(88 days)
The Imation™ SE-196 laser imager is intended use as a high quality hard copy device for output from digital imaging source modalities for use in medical imaging diagnosis and referral. Electronic image information signals are managed in the SE-196 and transformed optically to expose Imation imaging media. The system is intended for use with a variety of digital modalities including CT, MR and CR for diagnostic use by medical radiologists and communications to referring physicians and their patients.
The Imation SE-196 Laser Imager provides high quality hard copy film output from digital imaging source modalities for use in medical imaging diagnosis and referral. Electronic image information signals are managed in the SE-196 and transformed optically to expose Imation imaging media. The system is intended for use with a variety of digital modalities including CT, MR and CR for diagnostic use by medical radiologists and communications to referring physicians and their patients.
Here's an analysis of the provided text regarding the acceptance criteria and study for the Imation™ SE-196 Laser Imager, formatted to address your specific questions.
It's important to note that the provided document is a 510(k) summary from 1997 for a laser imager, not a modern AI/ML medical device. Therefore, many of the requested categories (e.g., sample size for test set, number of experts for ground truth, MRMC study, training set details) are not applicable or not present in this type of submission for this particular device. The "Performance Data" section discusses traditional imaging characteristics rather than AI algorithm performance.
Acceptance Criteria and Study for Imation™ SE-196 Laser Imager (K972163)
The device described is a laser imager, which is a physical device for printing medical images. Its performance criteria are related to image quality reproduction, not diagnostic accuracy of an AI algorithm.
1. Table of Acceptance Criteria
| Acceptance Criterion | Reported Device Performance |
|---|---|
| Spatial Frequency Response (DPI) | - Subject Device (Imation™ SE-196): 325 dpi (pixel size) |
- Predicate Devices: 325 dpi (pixel size)
Similar to predicate devices. |
| Gray Scale Resolution | - Subject Device (Imation™ SE-196): 12 bit - Predicate Devices: 8 or 12 bit
Equivalent to or better than predicate devices (only 12 bit, which is the higher end of predicate). Image properties are "same or better". |
| Density Uniformity | - Subject Device (Imation™ SE-196): Same or better than predicate devices. - Mechanism: Built-in density test patterns and AIQC (Automatic Image Quality Control) that maintains density uniformity over time and changes in film media.
Image properties are "same or better". |
| Safety (Voluntary Standards Compliance) | - UL544 - IEC601-1
- IEC 825
- Imation™ SE-196 Engineering Specification (Part B)
- Predicate Devices: Designed to UL1950.
Subject device designed to UL1950 (same as predicate) and other applicable standards. |
| Reliability, Qualification, Validation | - Successfully concluded field test. - Successfully concluded internal tests for qualification, validation, and reliability.
No specific numerical performance metric is given for these, but successful conclusion is stated as the criteria for final release. |
| AIQC (Automated Image Quality Control) | - Matches printing power with film characteristics to provide consistently high image quality. - Assure consistency between input signals and output density.
This is a feature that contributes to image quality (density uniformity) and is incorporated in both subject and predicate devices. |
Study Details:
The document describes "field tests and internal tests for qualification, validation and reliability." However, it does not provide detailed information about these studies in the context of AI/ML device evaluations. This is a traditional medical device submission, focusing on equivalence to predicate devices based on technological characteristics and functional performance related to image output quality.
2. Sample size used for the test set and the data provenance:
- Not applicable / Not specified for an AI algorithm. This document describes a laser imager, a hardware device for printing images. The "performance data" refers to characteristics of the printed output, not a diagnostic algorithm's performance on a dataset of patient images.
- The "field test" and "internal tests" would likely involve evaluating printed films, but the specific methodologies, sample sizes (of films or images), or data provenance (country, retrospective/prospective) are not detailed here.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable. The device is an imager, not a diagnostic AI tool. Ground truth in the context of an AI algorithm's diagnostic performance is not relevant here. The evaluation involves technical image properties, possibly assessed by engineering or quality control personnel against specifications.
- Medical personnel review the images displayed by the device, and their "competent human intervention" is mentioned as a safety mechanism, but not as part of establishing a ground truth for a diagnostic algorithm.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not applicable. This concept of adjudication is specific to evaluating diagnostic AI algorithms where expert consensus is needed to establish ground truth for ambiguous cases. It is not relevant to a laser imager's performance evaluation as 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. This is not an AI-assisted diagnostic device. Therefore, no MRMC study comparing human readers with and without AI assistance was performed or would be relevant.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not applicable. This device is a laser imager. It does not contain a standalone AI algorithm for diagnostic interpretation. Its "AIQC" is an Automated Image Quality Control system, managing printing power and film characteristics, not a diagnostic AI.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
- For the reported performance characteristics (spatial frequency response, gray scale resolution, density uniformity), the "ground truth" would be established through technical specifications, calibrated measurement tools, and industry standards for image quality. For example, dpi is measured directly, and gray scale is a technical specification of the digital-to-analog converter and laser modulation. Density uniformity would be measured using a densitometer against a calibrated test pattern. This is not medical ground truth like pathology or expert consensus.
8. The sample size for the training set:
- Not applicable. The device is a laser imager, not an AI/ML algorithm that requires a training set of medical images. The "AIQC" system is likely rule-based or uses internal calibration data, not a "training set" in the context of deep learning.
9. How the ground truth for the training set was established:
- Not applicable. As there is no AI/ML training set in the modern sense, the concept of establishing ground truth for it does not apply.
§ 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.