(224 days)
The CellChek 20 rc is a software program intended to analyze ophthalmic images captured by the Konan Specular Microscope XVII for examination of corneal endothelium.
Konan Medical has developed the CellChek 20 rc to provide photographic data taken exclusively by the Konan Specular Microscope XVII, CellChek 20, which was cleared by FDA under 510(k) number K191558 on Mar 26, 2020, to research and learning centers for the advancement of ophthalmic sciences and practice. The CellChek 20 rc was developed based on the software program of CellChek 20.
CellChek 20 rc is a software program to analyze ophthalmic images for examination of corneal endothelium. This has the cell counting analysis program, and allows for analysis of the images of the cell distribution of the eye.
The software program is installed on a general-use computer to analyze corneal endothelial images photographed exclusively by the Konan Specular Microscope XVII, CellChek 20. The analysis function is to calculate mainly the cell density, the coefficient of variation of cell area, and the percent hexagonality. In the manual methods, cornea endothelial cells and cell boundaries are actually identified by users. In the automatic methods, this software detects cells and cell boundaries, however, users can modify the detection results. During operating, the users interact with the software by visually placing dots in the center of each of cells and/or by tracing cell boundaries displayed on a computer screen, or use the automatic algorithm.
The provided document primarily focuses on the FDA 510(k) clearance process for the Konan Medical CellChek 20 rc software, establishing its substantial equivalence to a predicate device. It does not contain detailed information about specific acceptance criteria or an explicit study proving performance against those criteria.
However, it does mention that "CellChek 20 rc was developed according to the harmonized standard for software, IEC 62304, and FDA requirements for software and cybersecurity for the 510(k) clearance." It also states that "The following testing was performed on the CellChek 20 rc which was the same software function standard as those for CellChek 20: CellChek 20 rc device was subjected to software testing in accordance with IEC62304."
Based on the information available, here's what can be extracted and what is NOT available:
1. A table of acceptance criteria and the reported device performance
No explicit table of acceptance criteria with corresponding performance metrics is provided in the document. The document focuses on demonstrating substantial equivalence to a predicate device and adherence to general software safety standards.
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
This information is not provided in the document.
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)
This information is not provided in the document. The document mentions that users can identify cells and boundaries manually or modify automatic detections, implying human interaction with the software's analysis. However, it doesn't specify how ground truth for testing was established.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
This information is not provided in the document.
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
There is no mention of an MRMC comparative effectiveness study in the document. The software allows for both manual and automatic methods for cell analysis, with user modification, but no study is described to quantify human performance improvement with AI assistance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The document mentions both "manual methods" where "cornea endothelial cells and cell boundaries are actually identified by users" and "automatic methods" where "this software detects cells and cell boundaries, however, users can modify the detection results." This implies the algorithm can operate somewhat standalone, but the user is always in a position to review and adjust. However, no formal standalone performance study results, such as sensitivity, specificity, or accuracy, are presented for the algorithm itself.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
This information is not provided in the document.
8. The sample size for the training set
This information is not provided in the document. The document states that "CellChek 20 rc was developed based on the software program of CellChek 20," suggesting that any machine learning components (if present and requiring training data) would have been integrated or refined from the existing CellChek 20 platform.
9. How the ground truth for the training set was established
This information is not provided in the document.
Summary of available information regarding acceptance criteria and study:
The document primarily focuses on regulatory clearance through the 510(k) pathway, emphasizing "substantial equivalence" to a predicate device and adherence to general software development and safety standards (IEC 62304). It does not detail specific performance studies with quantitative acceptance criteria, ground truth establishment, or human reader performance metrics that are typical for demonstrating the effectiveness of an AI-driven medical device. The "study" mentioned is general "software testing in accordance with IEC62304," which is a standard for medical device software life cycle processes, focusing on safety and quality, rather than clinical performance metrics.
§ 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).