(52 days)
Cleerly Labs is a web-based software application that is intended to be used by trained medical professionals as an interactive tool for viewing and analyzing cardiac computed tomography (CT) data for determining the presence and extent of coronary plaques (i.e. atherosclerosis in patients who underwent Coronary Computed Tomography Angiography (CCTA) for evaluation of CAD or suspected CAD. This software post processes CT images obtained using any Computed Tomography (CT) scanner. The software provides tools for the measurement and visualization of coronary arteries.
The software is not intended to replace the skill and judgment of a qualified medical practitioner and should only be used by people who have been appropriately trained in the software's functions, capabilities and limitations. Users should be aware that certain views make use of interpolated data. This is created by the software based on the original data set. Interpolated data may give the appearance of healthy tissue in situations where pathology that is near or smaller than the scanning resolution may be present.
Cleerly Labs is a post-processing web-based software application that enables trained medical professionals to analyze 2D/3D coronary images acquired from Coronary Computed Tomography (CCTA) scans. The software is a post-processing tool that aids in determining treatment paths for patients suspected to have coronary artery disease (CAD).
Cleerly Labs utilizes machine learning and simple rule-based mathematical calculation components which are performed on the backend of the software applies deep learning methodology to identify high quality images, segment and label coronary arteries, and segment lumen and vessel walls. 2D and 3D images are presented to the user for review and manual editing. This segmentation is designed to improve efficiency for the user, and help shorten tedious, time-consuming manual tasks.
The user is then able to edit the suggested segmentation as well as adjust plaque thresholds, and demarcate stenosis, stents, and chronic total occlusions (CTOs) as well as select dominance and indicate coronary anomalies. Plaque, stenosis, and vessel measurements are output based the combination of user-editable segmentation and user-placed stent, and CTO markers. These outputs are mathematical calculations and are not machine-learning based.
Cleerly Labs provides a visualization of the Cleerly Labs analysis in the CORONARY Report. The CORONARY Report uses data previously acquired from the Cleerly Labs image analysis to generate a visually interactive and comprehensive report that details the atherosclerosis and stenosis findings of the patient. This report is not intended to be the final report (i.e., physician report) used in patient diagnosis and treatment. Cleerly Labs provides the ability to send the text report page of the CORONARY Report to the user's PACS system.
Cleerly Labs software does not perform any functions that could not be accomplished by a trained user with manual tracing method or other commercially available software. Rather, it represents a more robust semiautomatic software intended to enhance the performance of time-intensive, potentially error-prone, manual tasks, thereby improving efficiency for medical professionals in the assessment of coronary artery disease (CAD).
The provided text details the 510(k) summary for Cleerly Labs v2.0, focusing heavily on its substantial equivalence to the predicate device, Cleerly Labs (K190868). The performance data presented for Cleerly Labs v2.0 is referenced as being reproduced from the testing done for the predicate device (K190868), as no new clinical testing was conducted for v2.0. Therefore, the information provided below pertains to the study that proved Cleerly Labs (K190868) met its acceptance criteria, as the performance claims for v2.0 are based on this prior validation.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are implied by the "Pearson Correlation Coefficient" and "Bland-Altman Agreement" values for specific volume measurements. The device performance (reported for the predicate device, K190868, and reproduced for K202280) is presented in Table 6. While explicit "acceptance criteria" numerical thresholds are not stated (e.g., "must achieve a Pearson Correlation >= 0.8"), the reported values represent the performance achieved to demonstrate substantial equivalence to expert reader results.
| Output | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| Lumen Volume | High Pearson Correlation | 0.91 |
| High Bland-Altman Agreement | 96% | |
| Vessel Volume | High Pearson Correlation | 0.93 |
| High Bland-Altman Agreement | 97% | |
| Total Plaque Volume | High Pearson Correlation | 0.85 |
| High Bland-Altman Agreement | 95% | |
| Calcified Plaque Volume | High Pearson Correlation | 0.94 |
| High Bland-Altman Agreement | 95% | |
| Non-Calcified Plaque Volume | High Pearson Correlation | 0.74 |
| High Bland-Altman Agreement | 95% | |
| Low-Density-Non-Calcified Plaque Volume | High Pearson Correlation | 0.53 |
| High Bland-Altman Agreement | 97% |
Note: The document states, "Additionally, the performance of the software was previously compared to ground truth results produced by expert readers (K190868). Pearson Correlation Coefficients and Bland-Altman Agreements between Cleerly Labs and expert reader results are reproduced in Table 6." This indicates that these performance metrics were the basis for acceptance for the predicate device.
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size: The document does not specify the sample size used for the test set in either the original K190868 submission or this K202280 submission.
- Data Provenance: The document does not specify the country of origin of the data, nor does it explicitly state whether the data was retrospective or prospective. It only mentions that the data was from "patients who underwent Coronary Computed Tomography Angiography (CCTA)."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
- Number of Experts: The document refers to "expert readers" (plural) but does not specify the exact number of experts used to establish the ground truth.
- Qualifications of Experts: The document does not provide the specific qualifications of these experts (e.g., years of experience, specific certifications). It only refers to them as "expert readers."
4. Adjudication Method for the Test Set
- The document does not describe any specific adjudication method (e.g., 2+1, 3+1) used for establishing the ground truth from the expert readers. It only states that ground truth results were "produced by expert readers."
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No MRMC Comparative Effectiveness Study was done for Cleerly Labs v2.0. The document explicitly states: "No clinical testing was conducted to demonstrate safety or effectiveness as the device's non-clinical testing was sufficient to support the intended use of the device."
- The performance data provided (Table 6) refers to the comparison of the device's measurements against expert reader results in the predicate device's validation (K190868). This is a comparison between the device and human expert ground truth, not a study evaluating human readers' improvement with AI assistance.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Performance
- The data in Table 6, showing "Pearson Correlation Coefficients and Bland-Altman Agreements between Cleerly Labs and expert reader results," represents the standalone performance of the Cleerly Labs algorithm (or at least the initial segmented values before user editing).
- The description of the software functionality notes: "Cleerly Labs utilizes machine learning... applies deep learning methodology to identify high quality images, segment and label coronary arteries, and segment lumen and vessel walls. 2D and 3D images are presented to the user for review and manual editing." The performance metrics in Table 6 likely refer to the agreement of these automated segmentations with expert results. However, the subsequent statement also says: "Plaque, stenosis, and vessel measurements are output based the combination of user-editable segmentation and user-placed stent, and CTO markers. These outputs are mathematical calculations and are not machine-learning based." This suggests the final outputs are influenced by user edits. The document doesn't explicitly clarify if the correlation values are based on the initial automated output or the final user-edited output compared to the expert ground truth. Given the context of showing an automated system's performance, it's more likely referring to the algorithm's capability.
7. Type of Ground Truth Used
- The ground truth used was expert consensus (or at least expert reader results). The document states that the software's performance was "compared to ground truth results produced by expert readers."
8. Sample Size for the Training Set
- The document does not specify the sample size used for the training set for the machine learning components of Cleerly Labs.
9. How the Ground Truth for the Training Set Was Established
- The document does not describe how the ground truth for the training set was established. It only mentions that the machine learning components underwent validation.
§ 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).