Search Results

The imageSPECTRUM V6 is an ophthalmic software system indicated for acquiring, storing, managing, and display patient, diagnostic and image data from Canon digital retinal cameras. It is also indicated for review of patient, diagnostic and image data and measurement by trained healthcare professional.

The imageSPECTRUM V6 is an ophthalmic imaging software for acquiring, storing, managing, processing and displaying patient, diagnostic and image data from Canon digital retinal cameras. The imageSPECTRUM V6 consists of three software; iS Capture. iS Review, and iS Server. The "iS Capture" has functions to communicate with Canon's retinal camera to take retinal images. The "iS Review" has functions for displaying, processing, and transferring retinal images. The "iS Server" has functions for storing and archiving retinal images.

The imageSPECTRUM V6 is an ophthalmic software system indicated for acquiring, storing, managing, processing, and display patient, diagnostic and image data from Canon digital retinal cameras. It is also indicated for review of patient, diagnostic and image data and measurement by trained healthcare professional.

This document is a 510(k) Premarket Notification from the FDA for the ImageSPECTRUM V6 ophthalmic software system. It determines the device is substantially equivalent to a previously cleared predicate device.

Here's an analysis of the acceptance criteria and study information, based on the provided text:

Important Note: The document focuses on demonstrating substantial equivalence to a predicate device, not on proving de novo clinical effectiveness or establishing performance thresholds against specific diagnostic accuracy metrics typically associated with AI/ML devices. Therefore, a table of "acceptance criteria" in the traditional sense of diagnostic performance (e.g., sensitivity, specificity) and reported performance metrics is not present in this type of 510(k) submission for a Picture Archiving and Communication System (PACS) software. The acceptance criteria here would primarily relate to functional requirements, safety, and equivalence to predecessor devices.

Acceptance Criteria and Reported Device Performance

Since this is a 510(k) for a PACS software system, the "acceptance criteria" are more about functional equivalence and safety rather than diagnostic accuracy. The document states:

"Canon concluded that the Canon imageSPECTRUM V6 is substantially equivalent to the predicate devices based on identical intended use and substantially equivalent technological characteristics and the similarities in functional design."

The "reported device performance" is implicitly that the device performs its intended functions (acquiring, storing, managing, processing, and displaying ophthalmic image data) safely and effectively, similar to the predicate device. The performance data section vaguely mentions "Software verification and validation was performed to ensure that the software device performed as intended."

Table of Acceptance Criteria and Reported Device Performance (as inferred from the document's purpose):

• Review of patient, diagnostic, and image data and measurement by trained healthcare professionals.
  Specific Functional Similarities (from Table 1):
• User Management: Supported
• Patient Management (Backup/Restoration): Supported
• Compatible Devices: Canon's retinal cameras (CR-2 Plus AF)
• Image Types: Color, FA, FAF
• Capture Function (Communication, Sequence, Auto Exposure, Auto Focus, Auto Shot Control): Supported
• View Image (Single, Comparison, Both Eyes): Supported
• Drawing Function: Supported
• Image Processing (Brightness, Contrast, Zoom, RGB Filters, Redfree, Emboss, Mosaic, Overlay): Supported (Emboss, Mosaic, Overlay are new/different but deemed to raise no safety/efficacy questions)
• Annotation (Drawing Function, Add text): Supported (New annotation tools like Macular Grid, PDT Marker, PDT Counter, AVR, Protractor added, deemed to raise no safety/efficacy questions)
• Standalone Configuration: Supported
• Server Client Configuration: Supported
• DICOM Communication (Worklist, Storage, Commitment): Supported
• STEREO Viewing: Supported
• Printing Image: Supported
• Viewing Reports by Multiple Users: Supported |
  | Safety and Efficacy Equivalence | - The device is deemed safe and effective for its stated indications for use, equivalent to the predicate device.
• No new safety or efficacy questions are raised by the differences in added annotation or image processing tools.
• Software verification and validation was performed to ensure that the software device performed as intended. |
  | Hardware/Software Compatibility | - Software requirements: Microsoft Windows 10 Pro (64-bit)
• Hardware requirements: CPU: Core-i7 2GHz or Greater, RAM: 4GB or more, Display Screen resolution: 1920x1080 (These are higher than predicate but indicate functional capability and not a performance claim for the software itself) |

Beyond this, the document provides limited details on the "study" as it would be understood for an AI/ML diagnostic device with performance metrics. This 510(k) is for a PACS system, which is typically a Class II device where the focus is on functional equivalence and risk management, rather than complex performance studies of an AI algorithm making diagnostic interpretations.

Here are the details from the document regarding the "study" closest to what was requested:

1. Sample size used for the test set and the data provenance:

   • The document mentions "Software verification and validation was performed to ensure that the software device performed as intended." However, it does not specify any sample size for a test set of medical images or patient data.
   • Data Provenance: Not specified. It's likely that internal corporate data (from Canon, Japan) was used for software validation, but no specifics are provided regarding country of origin or whether it was retrospective or prospective.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Not applicable / Not specified. This type of information is generally required for AI/ML diagnostic devices where expert consensus is needed to establish ground truth for a labeled dataset. For a PACS system, the "ground truth" is that the software correctly acquires, stores, manages, and displays the image data as designed, and that its tools function accurately (e.g., measurement tools calculate correctly). The document implies internal software testing and validation.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • Not applicable / Not specified. Adjudication methods are typically used in clinical studies involving human readers and ground truth establishment for diagnostic tasks, which is not the primary focus of this PACS software.

4. 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 a PACS system, not an AI-based diagnostic tool designed to assist human readers. Its function is to manage and display images.

5. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

   • Not applicable / Not explicitly stated for diagnostic performance. The "algorithm" here is the PACS software itself. Its performance is its ability to handle and display images, and its various tools. There isn't a complex diagnostic "algorithm" that would perform in a standalone mode. The functions listed (image processing, annotation, etc.) are features of the software for use by a "trained healthcare professional."

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • Not explicitly stated in terms of medical ground truth. As a PACS system, the "ground truth" for its validation would be the verification that the software correctly implements its specified functions (e.g., images are stored correctly, retrieved accurately, measurements are mathematically precise, display quality is as expected). This would typically involve internal software testing against functional specifications.

7. The sample size for the training set:

   • N/A. This device is a PACS software, not an AI/ML model that is 'trained' on a dataset in the conventional sense. The "training" for this software would be its development and programming based on established software engineering principles.

8. How the ground truth for the training set was established:

   • N/A. See point 7.

Ask a specific question about this device

Canon non-mydriatic retinal camera CR series (CR-2 plus AF and CR-2 AF) is intended to be used for taking digital images of the retina of the human eye without mydriatic. The CR-2 plus AF has the following photography modes: color, red-free, cobalt digital and fundus autofluorescence (FAF). The CR-2 AF has the following photography modes: color, red-free, cobalt digital and additional infrared photography for anterior segment.

The Canon non-mydriatic retinal camera CR series is composed of two retinal cameras and its control software and is used for taking digital images of a human retina without mydriatic.
Two retinal cameras: The CR-2 Plus AF and CR-2 AF

• Both models take digital photographic retinal images of patient's eyes across an angle of view of 45 degrees, and have autofocus and automatic shooting of retinal image, and automatic switching from anterior segment image to retinal image.
• The CR-2 Plus AF camera is identical to the camera cleared under K123208. The photography mode includes color, red-free, cobalt and fundus autofluorescence (FAF).
• The CR-2 AF camera is modified version of the predicate CR-2 Plus AF (K123208). Most of the specifications are identical to the predicate; however, the photography mode does not include fundus autofluorescence (FAF) photography.
• Both the CR-2 AF and the CR-2 Plus AF uses infrared light to observe the anterior segment, but CR-2 AF can save such an image.
  Two Software: Retinal image control software (RICS) and Non-myd RC Capture Utility software (CU) programs
• RICS software program supports browsing, processing, and storage function of the images. It also supports the output of the images to the DICOM storage server, export in DICOM or JPEG format and output to a printer.
• CU software program is a simplified version of RICS, without most of the graphic user interface and the database. It allows user to capture and transfer the retinal images from the retinal camera to the PC.
  Both RICS and CU software programs are compatible with both CR-2 AF and CR-2 Plus AF cameras.

The provided document, a 510(k) Premarket Notification for the Canon non-mydriatic retinal camera CR series, primarily focuses on demonstrating substantial equivalence to a predicate device rather than presenting a detailed clinical study with acceptance criteria for an AI/algorithm-driven diagnostic device.

Therefore, the document does not contain the specific information required to answer many of the questions, particularly those related to acceptance criteria, sample sizes for test sets in an AI context, expert ground truth establishment, MRMC studies, or standalone algorithm performance.

The document discusses the technical specifications and modifications of the CR-2 AF camera compared to its predicate (CR-2 Plus AF) and confirms compliance with various electrical safety, EMC, performance, and software validation standards. It emphasizes that the changes do not introduce new risks or affect the basic design principle, usage, effectiveness, and safety.

Given the information in the provided document, here's what can be extracted and what cannot:

1. A table of acceptance criteria and the reported device performance

The document does not specify quantitative clinical acceptance criteria or performance metrics (e.g., sensitivity, specificity, AUC) for a diagnostic output. Its performance evaluation focuses on technical compliance and equivalence.

2. Sample sizes used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

The document mentions "Performance testing" and "Software Verification and Validation" but does not detail a specific test set with sample sizes, data provenance, or whether it was retrospective or prospective in a clinical performance context. The testing appears to be primarily for engineering performance and software functionality.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)

Not applicable, as no clinical test set for diagnostic accuracy requiring expert ground truth is described.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set

Not applicable.

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 MRMC study or AI assistance is mentioned in the context of this device. The device is a retinal camera, not an AI-powered diagnostic algorithm.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Not applicable.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

Not applicable. The "ground truth" for the device's performance appears to be its ability to capture images to a certain standard and its compliance with safety regulations, not diagnostic accuracy against a clinical ground truth. A "qualitative assessment of the function of the anterior segment infrared image" was performed, but details are absent.

8. The sample size for the training set

Not applicable, as this is a hardware device with control software, not a machine learning model that requires a training set of images for diagnostic prediction.

9. How the ground truth for the training set was established

Not applicable.

Summary of Device and Performance Information from the Document:

The document describes the Canon non-mydriatic retinal camera CR series (Model CR-2 AF and CR-2 Plus AF).

• Intended Use: To be used for taking digital images of the retina of the human eye without mydriatic.
• Key Differences (CR-2 AF vs. Predicate CR-2 Plus AF):
  • Photography Modes: CR-2 AF includes color, red-free, cobalt, and infrared for anterior segment. It does not have fundus autofluorescence (FAF) photography, which the predicate and CR-2 Plus AF model do.
  • Light Source: CR-2 AF uses White LED (lower output) instead of Xenon tube (required for FAF).
  • Software: Introduction of "Non-myd RC Capture Utility software (CU)" as a simplified version of RICS, without GUI and database, for capturing and transferring images.
  • Physical: CR-2 AF is slightly lighter and smaller due to the light source change.
• Performance Data:
  • Non-clinical tests: Electrical safety, Electromagnetic Compatibility, Performance testing, and Software Verification and Validation.
  • Compliance:
    • ANSI/AAMI ES 60601-1
    • IEC 60601-1-2
    • IEC 60601-1-6
    • ISO 15004-1
    • ISO 10940
    • ANSI Z80.36 (eye safety limits of Group 1 instrument for light hazard)
    • ISO10993 series (biocompatibility of patient contact parts, concluded no new concerns)
  • Software Validation: Validated according to the FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices (May 2005).
  • Qualitative Assessment: A qualitative assessment of the function of the anterior segment infrared image for CR-2 AF camera was performed.

The conclusion is that the proposed CR-2 AF is substantially equivalent to the predicate CR-2 Plus AF (K123208) because the differences do not affect the basic design principle, usage, effectiveness, and safety, and therefore do not raise any new questions regarding safety and effectiveness.

Ask a specific question about this device

KOWA nonmyd 8 is indicated for true color, infrared and fundus autofluorescent (FAF) imaging of a human retina without the use of a mydriatic agent.

The KOWA nonmyd 8 is an instrument that enables retinal examination and retinal image capturing using infrared rays without requiring the patients to take any mydriatics. The KOWA nonmyd 8 is used with KOWA VK-2s software (510(k) number is K190056) which is designated filing software installed in a computer and enables to save, register or print the captured images. The KOWA nonmyd 8 has Color mode and FAF mode for photographing. Those are the functions to capture a normal color image and the Fundus Autofluorescence (FAF) image.

The KOWA nonmyd 8 is an ophthalmic camera used for true color, infrared, and fundus autofluorescent (FAF) imaging of the human retina without pupil dilation.

Here's an analysis of the provided information regarding its acceptance criteria and supporting study:

1. Table of Acceptance Criteria and Reported Device Performance:

   Acceptance Criterion (Implicit)	Reported Device Performance
   Electrical Safety and Electromagnetic Compatibility (EMC): Compliance with relevant standards.	Complies with IEC60601-1:2005 and IEC60601-1-2:2014.
   Biocompatibility: No new biocompatibility concerns for patient contact parts (forehead rest, chin rest).	Forehead rest is identical to KOWA nonmyd 7 (K053026); chin rest is identical to KOWA DR-1α (K190573). Nature and duration of body contact are identical to reference devices. Determined no new biocompatibility concern.
   Software Verification and Validation: Compliance with FDA guidance for software in medical devices.	Software has been validated according to FDA guidance "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices."
   Optical Radiation Safety: Classification regarding light hazard and compliance with optical safety standards.	Classified in Group 1 of continuous wave instrument according to ANSI Z80.36-2016. Complies with ISO 10940:2009. Risk of radiation is low level, even with higher output power for photographing compared to the predicate device.
   FAF Image Quality: FAF image quality similar to or better than a legally marketed reference device (Canon CR-2 Plus AF).	75% of images taken by KOWA nonmyd 8 were superior or equivalent to the Canon CR-2 Plus AF (K123208). The remaining 25% were inferior but diagnosable. The clinical testing demonstrates that FAF image quality of the proposed device is similar to that of the Canon CR-2 Plus AF (K123208) and no new concern was raised.

2. Sample size used for the test set and data provenance:

   • Sample Size: 8 subjects.
   • Data Provenance: Not explicitly stated, but the study was conducted to compare the KOWA nonmyd 8 with the Canon CR-2 Plus AF, which implies prospective data collection for this purpose. The country of origin is not mentioned.

3. Number of experts used to establish the ground truth for the test set and their qualifications:

   • Number of Experts: One doctor (referred to as "a doctor").
   • Qualifications: "a doctor" is mentioned, indicating a medical professional capable of evaluating diagnostic image quality. Specific qualifications (e.g., years of experience, subspecialty) are not provided.

4. Adjudication method for the test set:

   • The method described is that "a doctor determined that the FAF image is not good quality for diagnosis" to prompt re-takes, and then the FAF images from the KOWA nonmyd 8 were "evaluated by comparing image quality with the FAF image from the Canon CR-2 Plus AF (K123208)." This suggests a single reader evaluation rather than a multi-reader adjudication method (like 2+1 or 3+1).

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 MRMC comparative effectiveness study was done in the context of human readers improving with or without AI assistance. The study described is a comparison of image quality between two devices (KOWA nonmyd 8 and an existing Canon device). There is no mention of AI.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • This question is not applicable as the KOWA nonmyd 8 is an ophthalmic camera, not an AI algorithm. The performance evaluation focuses on the image quality produced by the camera.

7. The type of ground truth used:

   • Expert Consensus: The "ground truth" for FAF image quality comparison was established by a single doctor's determination of diagnostic quality and subsequent comparison of images between the two devices. This leans towards expert opinion/evaluation. There is no mention of pathology or outcomes data being used as ground truth for image quality.

8. The sample size for the training set:

   • Not applicable. The document describes a comparison study of a new photographic device (KOWA nonmyd 8) against an existing one, not the training of an algorithm via a training set.

9. How the ground truth for the training set was established:

   • Not applicable, as there was no training set for an algorithm.

Ask a specific question about this device