Search Results

The HealthCCSng device is intended for use as a non-invasive post-processing software to evaluate calcified plaques in the coronary arteries, which present a risk for coronary artery disease.

The HealthCCSng device analyzes routine non-gated, non-contrast CT studies that include the entire heart of adult patients of age 30-85.

The device generates an exact calcium soore and a four coronary artery calcium detection category output representing the estimated quantity of calcium detected together with preview axial images of the detected calcium meant for informational purposes only.

The device output will be available to the radiologist as part of their standard workflow. A list of studies that received a successful algorithm analysis result will also be available for further clinician assessment (such as by a cardiologist, general practitioner etc.).

The HealthCCSng results are not intended to be used on a stand-alone basis for risk attribution, clinical decision-making or otherwise preclude clinical assessment of CT studies.

HealthCCSng product is a software device that automatically estimates the coronary artery calcium category from non-cardiac-gated adult CT scans. The product is aimed to leverage the high utilization of CT scans in the medical care environment (both inpatient and outpatient), including lung cancer screening programs, in order to automatically detect calcification in the coronary arteries of patients in an opportunistic manner.

The HealthCCSng product analyzes cases using an artificial intelligence algorithm for the automated detection and estimation of coronary calcium and outputs a result for review by the clinician. The device works in parallel to and in conjunction with the standard of care workflow. The final diagnosis is made by the clinician after reviewing the scan independently of the software. The device is intended for use by the clinicians as a non-diagnostic analysis software in conjunction with additional patient information and professional judgment.

HealthCCSng receives a non-gated, non-contrast CT study from the storage application, Nanox AI's Imaging Analytics Platform (IMA)/ other platforms. For each CT study received, the software shall validate at least one compliant series in which the entire heart is present and performs an analysis. For each complaint study, the software shall output:

1. Estimated Coronary Calcium Detection, based on the measurement of calcium deposits in the coronary arteries.
2. A corresponding Estimated Coronary Calcium Detection Category, based on the Estimated Coronary Calcium measurements.

The software output will include the following calcium categories:

= 400 | High

For patients in which calcium was detected, the user will be presented with representative key images - all the slices containing the measured coronary calcifications (130 HU and above). On these images, the calcified areas will be annotated.

The following modules compose the HealthCCSng software:

1. Data input and validation: DICOM validation receives imaging study from hosting application and the validation feature assessed the input data (i.e. age, modality, view, etc.) to ensure compatibility for processing by the algorithm.
2. HealthCCSng algorithm: Once a study has been validated, the algorithm analyzes the CT for analysis and quantification.
3. IMA Integration feature: The results of a successful study analysis are provided to the hosting application.
4. Error codes feature: In the case of a study failure during data validation or the analysis by the algorithm, an error is provided to the system.

The product operates in the following manner:

1. A CT scan is sent to the Image Analytics Platform (IMA).
2. The Image Analytics Platform (IMA) forwards the scan to the HealthCCSng algorithm for analysis.
3. The scan is analyzed, with the following possible results sent back to IMA:
   a. Non-compliant: the scan is not compliant with the input criteria of the product and is therefore not analyzed.
   b. Success -
4. Success A Zero calcium, Low, Medium or High Coronary Calcium Detection has been identified.
5. Success No Category Cause: metal artifact(s) suggestive of known cardiac disease is suspected.
   c. Failure: the scan has been analyzed. An error prevented the device from completing the analysis and therefore there is no output available.

In all cases of success. HealthCCSng will provide key images of visualization of the detected coronary calcium (except for zero category), the category name and the exact score.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Additional Performance Metric:

• Pearson's Correlation Coefficient (GT vs. HealthCCSng scores): 0.959 (95% CI: [0.9499, 0.9655], p<0.0001) – This indicates a strong positive correlation between the device's exact calcium score and the ground truth score.

Study Details

1. Sample Size for the Test Set and Data Provenance:

• Sample Size: 436 cases (algorithm returned results on 427 cases, leading to a yield of 97.94%).
• Data Provenance: Retrospective, anonymized CT scans from 4 healthcare institutions: Intermountain Healthcare (US), Clalit Health System (OUS / Outside US), Northwell Health (US), and USARAD (US). The sample was "truthed and enriched" to ensure sufficient cases within all coronary artery calcium categories.

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:

• Number of Experts: 3 US board-certified radiologists.
• Qualifications: Experienced in identifying coronary calcium on non-gated CT studies. Specific years of experience are not mentioned.

3. Adjudication Method for the Test Set:

• Method: Majority agreement of two out of three (2+1) radiologists established the ground truth for the Coronary Artery Calcium Category.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

• The document does not indicate that an MRMC comparative effectiveness study was performed to show how human readers improve with AI vs. without AI assistance. The study described is a standalone performance study comparing the device to a ground truth established by experts.
• It does state: "In addition, the device performed as well as radiologists did, however the user should note that Coronary Calcium Scoring on non-gated CT scans, is not as accurate as on gated CT scans." This suggests a general comparison, but not an MRMC study demonstrating improvement with AI assistance.

5. Standalone Performance (Algorithm Only without Human-in-the-Loop):

• Yes, a standalone performance study was conducted. The study evaluated the HealthCCSng's performance "compared to the established ground truth."

6. Type of Ground Truth Used:

• Expert Consensus: The ground truth for the Coronary Artery Calcium Category was determined by the "majority agreement of two out of three US board certified radiologists."
• For exact calcium score: The document mentions "the corresponding ground truth (GT) score" but does not explicitly detail how this exact GT score was derived beyond the radiologists determining categories. Given the context, it implicitly refers to the quantitative measurement radiologists would arrive at or validate.

7. Sample Size for the Training Set:

• The document does not specify the sample size used for the training set. It only details the validation (test) dataset.

8. 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 describes the validation dataset's ground truth establishment. Given it's a deep-learning-based algorithm, it's highly probable that the training set's ground truth was also established by expert annotation or consensus, similar to the validation set.

Ask a specific question about this device