Evaluation of an AI-based, automatic coronary artery calcium scoring software

Abstract

Objectives To evaluate an artificial intelligence (AI)-based, automatic coronary artery calcium (CAC) scoring software, using a semi-automatic software as a reference. Methods This observational study included 315 consecutive, non-contrast-enhanced calcium scoring computed tomography (CSCT) scans. A semi-automatic and an automatic software obtained the Agatston score (AS), the volume score (VS), the mass score (MS), and the number of calcified coronary lesions. Semi-automatic and automatic analysis time were registered, including a manual double-check of the automatic results. Statistical analyses were Spearman's rank correlation coefficient (⍴), intra-class correlation (ICC), Bland Altman plots, weighted kappa analysis (κ), and Wilcoxon signed-rank test. Results The correlation and agreement for the AS, VS, and MS were ⍴ = 0.935, 0.932, 0.934 (p < 0.001), and ICC = 0.996, 0.996, 0.991, respectively (p < 0.001). The correlation and agreement for the number of calcified lesions were ⍴ = 0.903 and ICC = 0.977 (p < 0.001), respectively. The Bland Altman mean difference and 1.96 SD upper and lower limits of agreements for the AS, VS, and MS were − 8.2 (− 115.1 to 98.2), − 7.4 (− 93.9 to 79.1), and − 3.8 (− 33.6 to 25.9), respectively. Agreement in risk category assignment was 89.5% and κ = 0.919 (p < 0.001). The median time for the semi-automatic and automatic method was 59 s (IQR 35-100) and 36 s (IQR 29-49), respectively (p < 0.001). Conclusions There was an excellent correlation and agreement between the automatic software and the semi-automatic software for three CAC scores and the number of calcified lesions. Risk category classification was accurate but showing an overestimation bias tendency. Also, the automatic method was less time-demanding. Key Points • Coronary artery calcium (CAC) scoring is an excellent candidate for artificial intelligence (AI) development in a clinical setting. • An AI-based, automatic software obtained CAC scores with excellent correlation and agreement compared with a conventional method but was less time-consuming.

Deep learning analysis provides accurate COVID-19 diagnosis on chest computed tomography

“…RT-PCR to diagnose COVID-19 has its own limitations: the test is not universally available, turnaround times can be lengthy, and reported sensitivities vary [ 20 ]. The combination of chest CT and ML has the advantage of obtaining an immediate result – practically after image reconstruction and while the patient is still on the examination bed – and independent of the presence of the radiologist [ [21] , [22] , [23] , [24] ]. This scenario could potentially play a role in either minor peripheral hospitals or places with radiological personnel shortage due to various reasons including a high load of COVID-19 cases and could also act as a support for inexperienced radiologists during night-duty.…”

Study on the relationship between Cardio-cerebrovascular events and the level of Atherosclerosis evaluated by Heart-brain Integrated CTA

“…The commonly used clinical calcification score CACS refers to Agatston score [7][8][9] , its disadvantage is due to the partial volume effect, the small calcification plaque score is easy to change during reexamination, the improved APQ technique avoids this change. Keelan 10 et al pointed out that calcification score can be used as an independent predictor of future coronary events, and is superior to the degree of stenosis and traditional risk factors.…”

The Future of Concurrent Automated Coronary Artery Calcium Scoring on Screening Low-Dose Computed Tomography

“…CAC scoring has been traditionally measured by imaging specialists with a semi-automatic software, starting with the manual demarcation of the coronary artery calcifications and subsequent computer calculation of the Agatston score. With greater volumes of studies being performed and the potential of CAC scoring being added to other imaging protocols, there is a growing need for a more efficient and automated method [41]. The primary problem with the development of an automated system arises from the fact that calcifications are not solely present in the coronary arteries but also in surrounding cardiac tissues and cardiac valves, making automated detection complicated [42].…”

“…Yang et al proposed an algorithm on producing an automatic method for calcium scoring by using data from GCTs to create segmentation maps, which were then applied to the NCT datasets ( Figure 3). Reproduced with permission from Sandstedt et al [41], copyright © 2019, European Radiology http://creativecommons.org/licenses/by/4.0/…”

“…A recent study by Sandstedt et al described completely automated software that utilized machine learning trained on annotated routine GCTs. The algorithm was then used for CAC scoring using 315 GCTs and validated against semi-automatic scoring on the same scans [41]. They were able to show that the algorithm had excellent correlation and agreement with the semi-automatic manual method but took less time (Figure 4).…”