(151 days)
qCT LN Quant is a software device used in the tracking, assessment, and quantitative characterization of detected pulmonary nodules. This automatically analyzes user-selected regions within lung CT to provide volumetric, diameter and computer analysis based on morphological characteristics in a single study, or over the time course of several thoracic studies. The system performs the measurements, allowing the preview of lung nodules in 2D and 3D reconstructed views and the respective measurements to be displayed. It is indicated for the evaluation of user detected solid pulmonary nodules.
Qure.ai's computed tomography (CT) scan software, the qCT LN Quant, is a deep-learning-based device that can process non-contract CT (NCCT) scans and assists in quantitative characterization of solid lung nodules of size ≥ 6mm on Chest CTs.
qCT LN Quant consists of a cloud module that can interacts with DICOM modality or the user's picture archiving and communication system (PACS) to receive de-identified scans and returns the results to the same destination. In addition, solid nodules are segmented by the user semi-automatically using double seed points on the nodule, followed by interactive fine-tuning of the boundaries. The segmented region is quantitatively characterized by qCT LN Quant and presented to users as an additional overlay by highlighting and labelling respectively. User-assisted segmentation generated by qCT LN Quant can be presented in two ways to the users:
a. PACS-based mode: As a new series (secondary capture) which are returned to the originating PACS system with segmentation burnt on the series. This can be done only at PACS which supports GSPS Output.
b. Web-based mode: On Qure's web application where the segmentation is overlaid on top of the original scan.
qCT LN Quant deep learning algorithm has been trained to quantify the target structures on CT scans and is coupled with pre-and post-processing functionality that allows the device to work directly with the radiology workflow. The user is presented with 2D view and 3D reconstructed view of solid nodule images labelling the quantitative characteristics based on the user-segmented structures. The output consists of information on average diameter and volumes of user identified solid nodules. The additional features include long axis diameter (mm), short axis diameter (mm), Effective diameter (mm), and Mean/Minimum/Max HU (HU) and volume change overtime with matched nodules. In addition, qCT LN Quant consists of a Brock Score - Risk Calculator that uses diameter of the nodule and clinician's input. The Lung-RADS™ calculator feature is based on ACR guideline, which can assist the physician in decision making. qCT LN Quant also provides recommendations based on Fleischner's Society guideline. Thus, qCT LN Quant offers functionality to calculate Brock and LungRADS score as part of integrated or cleared devices with capability to display such output.
qCT LN Quant is limited to analysis of imaging data and results generated are meant for information purposes only. The device is not intended for clinical diagnosis of any disease. It does not replace the role of physician or of other testing in the standard of care for lung abnormalities.
Here's a detailed breakdown of the acceptance criteria and study information for the qCT LN Quant device, based on the provided FDA 510(k) summary:
Acceptance Criteria and Device Performance
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't explicitly state "acceptance criteria" in a structured table format with specific thresholds before the study was conducted. Instead, it presents the results of the performance testing. However, we can infer the implied acceptance criteria from the reported performance, suggesting that these values were considered "good performance" and met "predefined success criteria."
| Measurement | Implied Acceptance Criteria (Likely Max. Median Absolute Normalized Error %) | Reported Device Performance (Median Absolute Normalized Error %) | 95% Confidence Interval |
|---|---|---|---|
| Short Axis Diameter | Not explicitly stated, but likely acceptable if ≤ 16.67% | 14.3 | 13.95 - 16.67 |
| Long Axis Diameter | Not explicitly stated, but likely acceptable if ≤ 12.50% | 11.1 | 9.52 - 12.50 |
| Volume | Not explicitly stated, but likely acceptable if ≤ 22.41% | 20.7 | 17.29 - 22.41 |
2. Sample Size Used for the Test Set and Data Provenance:
- Sample Size: 216 solid nodules identified from a total of 118 chest CT scans.
- Data Provenance: The document does not specify the country of origin of the data or whether it was retrospective or prospective.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications:
- Number of Experts: Three expert radiologists.
- Qualifications: The document states "three expert radiologists," but does not explicitly detail their years of experience or specific subspecialty certifications.
4. Adjudication Method for the Test Set:
- Adjudication Method: "The truthers independently read the scans and mark out the boundaries of the nodule in all slices." This implies a consensus-based approach after independent marking, but the specific adjudication rules (e.g., how disagreements were resolved, 2+1, 3+1, or simple majority) are not explicitly stated. It is a form of expert consensus, but the mechanism for reaching the final ground truth from independent readings isn't fully detailed.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:
- No MRMC study was done with AI assistance vs. without AI assistance. The study described is a standalone performance study of the device.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:
- Yes, a standalone study was done. The document explicitly states: "Performance of the qCT LN Quant device in quantitative characterization of solid nodules was assessed using the standalone study."
7. The Type of Ground Truth Used:
- Expert Consensus. The ground truth "was established by three expert radiologists" who "independently read the scans and mark out the boundaries of the nodule in all slices."
8. The Sample Size for the Training Set:
- The document does not provide the sample size for the training set. It mentions that the qCT LN Quant deep learning algorithm "has been trained to quantify the target structures on CT scans," but the size of this training dataset is not disclosed in the provided text.
9. How the Ground Truth for the Training Set Was Established:
- The document does not explicitly state how the ground truth for the training set was established. It only mentions that the algorithm was trained.
§ 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).