(308 days)
IQA is a software system intended for use in importing, displaying, analyzing and managing images acquired with digital fundus cameras.
The VisionQuest Biomedical Image Quality Analyzer (IQA) is an ophthalmic software system intended to import, display, analyze, and manage retinal images from Canon CR2, Canon CR2-AF, Volk Pictor Plus and Zeiss VisuScout retinal cameras.
IQA is a software application that is used to assess the quality of retinal images acquired with supported non-mydriatic retinal cameras. All images acquired by the retinal camera passed to the IQA undergo an image quality calculation process and the results are presented to the user in five seconds or less.
Image quality is assessed based on the presence and extent of three imaging artifacts: crescents, shadows, and blurriness. An image quality output of "Adequate" or "Inadequate" is calculated automatically by IQA based on a set of pre-determined thresholds on the three imaging artifacts. The image quality output can be used by the IOA user to determine whether a retinal image should be re-acquired or not. IQA does not modify the retinal images. The images processed by the IQA can be used for further processing, manual or automatic grading, or made available to an image management system such as a PACS.
Adequate images are automatically moved by the IQA to a user-defined "Save" folder in the computer. The user can manually instruct the IOA to move inadequate images to the same Save folder, thus overriding the IQA, or to a user-defined "Discard" folder to segregate the inadequate quality images from adequate quality images in the "Save" folder. Regardless of whether an image is adequate or inadequate, IQA does not modify the image and only moves the image from the input folder to either the Save or Discard folder.
The retinal images moved to the Save folder can be made available to other steps in the clinical flow such as a PACS, manual or automatic grading, or archiving.
IQA generates an activity log that captures the images processed, the timestamp of the processing, the user's initials, a session number, the output result, the actions of the user, and the values of the image quality calculations with respect to the three imaging artifacts. This log is a comma-separated-value file (csv) that can be retrieved from the computer running the IQA and read in any of a variety of analytical software tools such as MS Excel or R.
IQA works as a part of a clinical workflow to help minimize the acquisition of retinal images of inadequate quality and improve the reading efficiency of human graders or automatic grading software. As such, IQA acts as a middle-ware between a retinal camera and an image management system, such as a PACS. IQA helps a user, normally a retinal photographer, capture images of adequate quality that can then be used for further processing. Similarly, IQA helps the user segregate images of inadequate quality so they are not used for further processing. Thus, the output results of IQA are limited to labeling a retinal image as Adequate or Inadequate as depicted in Figure 2. IQA receives as inputs retinal images in JPEG format which have been captured by a trained photographer using a supported retinal camera. The computer that runs IQA then produces an output for each image.
The images captured with the retinal camera are copied to the IQA "Fetch" folder in JPEG format. IQA then automatically analyzes the quality of the images. When the result of the analysis is that an image is adequate, IQA automatically moves the image to the user-defined Save directory and waits for the user to take the next image, start a new session or stop IQA image processing. When the result of the analysis is that the image is inadequate, IQA displays the results to the user and waits for the user's decision to move the image to either the Save or Discard folder.
IQA's automatic image quality algorithms determine if the images contain any of three different artifacts (Imaging artifact analysis block in Figure 2) namely bright artifacts (crescents), shadows, or blurriness that could degrade the ability of a human or a computer-based algorithm to determine the presence of retinal disease.
If an image does not have sufficient image quality, IQA returns an output of "Inadequate". When this is the case, IQA presents the operator with two options, either to Save or Discard the inadequate image. It is recommended that the operator try to capture another image from the patient to ensure appropriate quality. This is done by discarding the inadequate image and taking another one. Images moved to the Discard folder are segregated from further processing and remain in the Discard folder for retrieval if desired.
When an image has adequate image quality, IQA will automatically move it to the user-defined Save directory and display a thumbnail in the IQA user interface. Then the operator can continue the imaging process. Images moved to the Save folder are available for further processing according to the site's clinical flow and standard operating procedures.
IQA can process as many images as are acquired regardless of the field of view or number of previously acquired images. It is up to the operator to determine whether sufficient images of adequate quality have been acquired from a patient.
Here's a summary of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Acceptance Criteria and Device Performance
| Criteria | Reported Device Performance (IQA) |
|---|---|
| Adequate Images: Successfully identifies "adequate" images. (Acceptance Criterion: 90% success rate) | 181 out of 182 images identified as "adequate". Success Rate: 99.45% |
| Inadequate Images: Successfully identifies "inadequate" images. (Acceptance Criterion: 90% success rate) | 166 out of 166 images identified as "inadequate". Success Rate: 100% |
Study Details
1. Sample Size for the Test Set and Data Provenance:
- Total Images: 348 images
- 182 "adequate" images
- 166 "inadequate" images
- Breakdown by Camera:
- Canon CR2: 109 images (55 adequate, 54 inadequate)
- Canon CR2-AF: 113 images (60 adequate, 53 inadequate)
- Volk Pictor Plus/Zeiss VisuScout: 126 images (67 adequate, 59 inadequate)
- Data Provenance: The document does not explicitly state the country of origin or if the data was retrospective or prospective. It only mentions that the images were "from the supported cameras."
2. Number of Experts and Qualifications for Ground Truth:
- Number of Experts: One expert.
- Qualifications: Referred to as "an expert with specific artifacts" who "known to have 'adequate' or 'inadequate' quality." No further specific qualifications (e.g., medical specialty, years of experience) are provided.
3. Adjudication Method for the Test Set:
- The document does not describe an adjudication method for the test set. Ground truth was established by a single expert.
4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No MRMC comparative effectiveness study was mentioned or performed in the provided text.
5. Standalone Performance (Algorithm Only):
- Yes, a standalone performance study was conducted. The IQA device's ability to classify images as "adequate" or "inadequate" was tested independently against expert-established ground truth. The results (99.45% for adequate, 100% for inadequate) are based solely on the algorithm's performance.
6. Type of Ground Truth Used:
- Expert Consensus (single expert): The images were "known to have 'adequate' or 'inadequate' quality as rated by an expert with specific artifacts."
7. Sample Size for the Training Set:
- The document does not provide the sample size for the training set. It only details the test set.
8. How Ground Truth for the Training Set Was Established:
- The document does not provide information on how the ground truth for the training set was established.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(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).