(60 days)
The Howtek Fulcrum Family consists of a transparency film digitizer and paper CCD scanner that is used to digitize radiographs or X-ray film as well as paper reports or doctor's orders. When the Fulcrum is used to digitize radiographic films, the digital image is intended for use in primary, secondary and over reading applications.
The Fulcrum Family does not include applications specific software (Picture Archiving and Communications (PAC) system, Teleradiology, or Computer Aided Detection (CAD) software). The manufacturer of the application software will determine specific indications for use. These third-party software packages or complete PAC systems are approved separately.
The device is a transparency film digitizer and paper, charge coupled device (CCD) digitizer that can be used to digitize radiographs or paper reports. The digitizer reads digital gray scale information for each pixel deposits the raw data in the computer. The user interface for the digitizer resides in the applications software. There is no activate button on the digitizer. The activate button is provided trough the digitizer user interface which is part of the third party software package.
Overall the Fulcrum uses a drive, which transport the paper or film over the camera as it digitizes the image. Each scan line is captured by the chargecoupled device (CCD). The scan line of data is normalized and calibrated to align the response from each individual pixel. The image data signal is converted from analog to digital format by and A to D converter. The Fulcrum creates a digital image in either 16-bit, 12-bit or 8-bit raw format. The information is transmitted to the computer via USB II.
The provided text is related to the 510(k) premarket notification for the "Howtek Fulcrum Family" film digitizer. However, it does not contain the specific details required to complete all sections of your request regarding acceptance criteria and a detailed study description.
Here's a breakdown of what can be extracted and what is missing:
What can be extracted/inferred:
- Device Type: Film Digitizer
- Purpose: Digitize radiographs/X-ray film and paper reports.
- Intended Use for Radiographs: Primary, secondary, and over-reading applications.
- Comparison: To a predicate device, the Howtek MultiRAD.
- Claimed Performance: The Fulcrum is a "superior product" with "the same or better range and resolution" than the MultiRAD.
- Study Existence: "The performance testing results for the Fulcrum demonstrated that the device meets its intended use specifications..."
What is missing entirely or in detail:
- Detailed Acceptance Criteria Table: The document states that performance testing demonstrated that the device meets its intended use specifications, but it does not list these specifications (e.g., specific thresholds for resolution, density range, gray scale resolution, signal-to-noise ratio, etc.) nor provide the quantitative "reported device performance." It only uses qualitative comparative terms like "better" or "same or better."
- Sample Size for Test Set and Data Provenance: No information provided.
- Number of Experts and Qualifications: Not mentioned.
- Adjudication Method: Not mentioned.
- MRMC Comparative Effectiveness Study: Not mentioned, and highly unlikely for a film digitizer from this era. This type of study is more common for diagnostic AI algorithms directly impacting human reader performance. This device is a component.
- Standalone Performance: While the document implies standalone performance testing of the digitizer's characteristics (resolution, density range), it doesn't present it as a formal "standalone study" in the context of an AI algorithm's diagnostic accuracy. It's a performance evaluation of the hardware's output.
- Type of Ground Truth: Not specified for any performance testing. For a digitizer, ground truth would likely be based on objective physical measurements of the film and its digital representation, rather than medical outcomes or pathology.
- Sample Size for Training Set: Not applicable as this is a hardware device, not a machine learning model.
- How Ground Truth for Training Set was Established: Not applicable.
Based on the provided text, here is what can be filled out, with explicit notes about missing information:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Conceptual) | Reported Device Performance |
|---|---|
| Specific quantitative thresholds for density range, gray scale resolution, spatial resolving power, scan rates. | "The Fulcrum family has the same or better range and resolution as the MultiRAD family." |
| Meets intended use specifications. | "The performance testing results for the Fulcrum demonstrated that the device meets its intended use specifications and therefore meets the requirements necessary for its intended use..." |
| Equivalent or superior performance to predicate device (Howtek MultiRAD). | "The Fulcrum is a superior product." "The Fulcrum is more than capable of providing equivalent performance than the MultiRAD." |
| Specific quantitative thresholds for various image quality metrics after digitization. | No specific quantitative metrics reported in the document. |
2. Sample sized used for the test set and the data provenance
- Sample Size: Not specified.
- Data Provenance: Not specified. (The document describes general "performance testing" but does not detail the dataset used for such testing.)
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Not applicable/Not specified. The ground truth for evaluating a digitizer would likely involve objective physical measurements rather than expert interpretations of digitized images, as the device's role is to accurately convert analog to digital.
4. Adjudication method for the test set
- Not applicable/Not specified.
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 comparative effectiveness study was not conducted or mentioned. This type of study is relevant for AI algorithms that assist human interpretation, not for a standalone film digitizer. The device itself "does not include applications specific software (Picture Archiving and Communications (PAC) system, Teleradiology, or Computer Aided Detection (CAD) software)."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Yes, performance testing of the digitizer's characteristics (scan rates, density range, gray scale resolution, spatial resolving power) was implicitly done in "standalone" mode, as it's a hardware device. However, the exact methodology and quantitative results of this "performance testing" are not detailed in the provided text. The document states, "The performance testing results for the Fulcrum demonstrated that the device meets its intended use specifications."
7. The type of ground truth used
- Inferred: For a film digitizer, the "ground truth" for performance testing would typically involve objective measurements of the original physical film characteristics (e.g., known optical density ranges, resolution patterns) and comparing them to the fidelity and accuracy of the digitized output. The document itself does not explicitly state the ground truth methodology.
8. The sample size for the training set
- Not applicable. This device is a hardware film digitizer, not an AI/machine learning algorithm requiring a "training set."
9. How the ground truth for the training set was established
- Not applicable.
§ 892.2030 Medical image digitizer.
(a)
Identification. A medical image digitizer is a device intended to convert an analog medical image into a digital format. Examples include Iystems employing video frame grabbers, and scanners which use lasers or charge-coupled devices.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std.). 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.