Inter- and Intraoperator Variability in Measurement of On-Site CT-derived Fractional Flow Reserve Based on Structural and Fluid Analysis: A Comprehensive Analysis

Kumamaru, Kanako K.; Angel, Erin; Sommer, K.; Iyer, Vijay; Wilson, Michael F.; Agrawal, Nikhil; Bhardwaj, Aishwarya; Kattel, Sharma B; Kondziela, Sandra; Malhotra, Saurabh; Manion, Christopher; Pogorzelski, Katherine; Ramanan, Tharmathai; Sawant, Abhishek C.; Suplicki, Mary M; Waheed, Sameer; Fujimoto, Shinichiro; Sharma, Umesh C; Rybicki, Frank J.; Ionita, Ciprian N.

doi:10.1148/ryct.2019180012

Cited by 11 publications

(18 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This variation between operators can be attributed to differences in the vessel centerline placement and the segmentation of the lesion, and highlights a potential advantage to an automatic machine learning‐based algorithm. This variation between users with different levels of experience has been assessed and reported in the literature 30 . There is no such variability with the trained network.…”

Section: Discussionmentioning

confidence: 88%

See 1 more Smart Citation

Initial evaluation of a convolutional neural network used for noninvasive assessment of coronary artery disease severity from coronary computed tomography angiography data

et al. 2020

Self Cite

View full text Add to dashboard Cite

Purpose: Coronary computed tomography angiography (CTA) has one of the highest diagnostic sensitivities for detection of the significance of coronary artery disease (CAD); however, sensitivity is moderate and may result in increased catheterization rates. We performed an efficacy study to determine whether a trained machine learning algorithm that uses coronary CTA data may improve CAD diagnosis accuracy. Methods: Sixty-four-patient image datasets based on coronary CTA were retrospectively collected to generate eight views considering 45°increments around the coronary artery centerline. The dataset was randomly split into training and testing cohorts. Invasive FFR measurements were used as ground truth labels. A convolutional neural network (CNN) was trained and the model's capacity to predict severity of CAD was assessed on the testing cohort. Classification accuracy and area under the receiver operating characteristic curve (AUROC) analysis were performed. Similar CAD severity classification accuracy and AUROC analyses were performed using only percent diameter stenosis (%DS) and CT-derived FFR performed by 13 operators with various levels of expertise. Results: Classification accuracy over the test cohort was 80.9% using the trained network and 72.4% using the user-operated CT-derived FFR software. AUROC over the test cohort was 0.862 using the trained network, 0.807 using %DS, and 0.758 using the human-operated CT-derived FFR software. Conclusions: A trained neural network compared noninferiorly in-terms of classification accuracy and AUROC with human operators of a CT-derived FFR software, and in-terms of AUROC with clinical decision-making using %DS.

show abstract

Section: Discussionmentioning

confidence: 88%

“…This variation between users with different levels of experience has been assessed and reported in the literature. 30 There is no such variability with the trained network.…”

Section: Discussionmentioning

confidence: 99%

Initial evaluation of a convolutional neural network used for noninvasive assessment of coronary artery disease severity from coronary computed tomography angiography data

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…These studies featuring both on-site and off-site approaches have reported a high degree of inter-operator correlation which was consistent among operators of different expertise and training. [20,21] Moreover, the previously cited study also emphasized decreased variabilities in operators receiving face-to-face training [20]. As such in-person training may counter a potential source of variability which has been the incorrect determination of centerline [20,22].…”

Section: Discussionmentioning

confidence: 92%

“…[20,21] Moreover, the previously cited study also emphasized decreased variabilities in operators receiving face-to-face training [20]. As such in-person training may counter a potential source of variability which has been the incorrect determination of centerline [20,22].…”

Section: Discussionmentioning

confidence: 92%

Inter-Operator Reliability of an Onsite Machine Learning-Based Prototype to Estimate Ct Angiography-Derived Fractional Flow Reserve

Han¹,

Ahmed²,

Schwemmer³

et al. 2021

Journal of the American College of Cardiology

View full text Add to dashboard Cite

Background: Advances in computed tomography (CT) and machine learning have enabled on-site non-invasive assessment of fractional ow reserve (FFR CT ).Purpose: To assess the inter-operator variability of Coronary CT Angiography-derived FFR CT using a machine learning based post-processing prototype.Materials and Methods: We included 60 symptomatic patients who underwent coronary CT angiography. FFR CT was calculated by 2 independent operators after training using a machine learning based on-site prototype. FFR CT was measured 1 cm distal to the coronary plaque or in the middle of the segments if no coronary lesions were present. Intraclass correlation coe cient (ICC) and Bland-Altman analysis were used to evaluate inter-operator variability effect in FFR CT estimates. Sensitivity analysis was done by cardiac risk factors, degree of stenosis and image quality.Results: A total of 535 coronary segments in 60 patients were assessed. The overall ICC was 0.986 per patient (95% CI: 0.977 -0.992) and 0.972 per segment (95% CI: 0.967 -0.977). The absolute mean difference in FFR CT estimates was 0.012 per patient (95% CI for limits of agreement: -0.035 -0.039) and 0.02 per segment (95% CI for limits of agreement: -0.077 -0.080). Tight limits of agreement were seen on Bland-Altman analysis. Distal segments had greater variability compared to proximal/mid segments (absolute mean difference 0.011 vs 0.025, p<0.001). Results were similar on sensitivity analysis. Conclusion:A high degree of inter-operator reproducibility can be achieved by onsite machine learning based FFR CT assessment. Future research is required to evaluate the physiological relevance and prognostic value of FFR CT .

show abstract

“…Involving the segmentation process, there may have been some errors in exclusion of hard-to-detect, soft plaque when segmenting the calcification from the vasculature, thus altering the diameter. We cross-validated this process between two users to avoid significant errors [31]. In addition, in the process of small branch and segmentation artifact elimination, we sculpted the mesh within Autodesk Meshmixer manipulating single triangular vertices, which might have created minor geometric alteration.…”

Section: Discussionmentioning

confidence: 99%

Method to simulate distal flow resistance in coronary arteries in 3D printed patient specific coronary models

et al. 2020

Self Cite

View full text Add to dashboard Cite

Background: Three-dimensional printing (3DP) offers a unique opportunity to build flexible vascular patient-specific coronary models for device testing, treatment planning, and physiological simulations. By optimizing the 3DP design to replicate the geometrical and mechanical properties of healthy and diseased arteries, we may improve the relevance of using such models to simulate the hemodynamics of coronary disease. We developed a method to build 3DP patient specific coronary phantoms, which maintain a significant part of the coronary tree, while preserving geometrical accuracy of the atherosclerotic plaques and allows for an adjustable hydraulic resistance. Methods: Coronary computed tomography angiography (CCTA) data was used within Vitrea (Vital Images, Minnetonka, MN) cardiac analysis application for automatic segmentation of the aortic root, Left Anterior Descending (LAD), Left Circumflex (LCX), Right Coronary Artery (RCA), and calcifications. Stereolithographic (STL) files of the vasculature and calcium were imported into Autodesk Meshmixer for 3D model optimization. A base with three chambers was built and interfaced with the phantom to allow fluid collection and independent distal resistance adjustment of the RCA, LAD and LCX and branching arteries. For the 3DP we used Agilus for the arterial wall, VeroClear for the base and a Vero blend for the calcifications, respectively. Each chamber outlet allowed interface with catheters of varying lengths and diameters for simulation of hydraulic resistance of both normal and hyperemic coronary flow conditions. To demonstrate the manufacturing approach appropriateness, models were tested in flow experiments. Results: Models were used successfully in flow experiments to simulate normal and hyperemic flow conditions. The inherent mean resistance of the chamber for the LAD, LCX, and RCA, were 1671, 1820, and 591 (dynes • sec/ cm 5), respectively. This was negligible when compared with estimates in humans, with the chamber resistance equating to 0.65-5.86%, 1.23-6.86%, and 0.05-1.67% of the coronary resistance for the LAD, LCX, and RCA, respectively at varying flow rates and activity states. Therefore, the chamber served as a means to simulate the compliance of the distal coronary trees and to allow facile coupling with a set of known resistance catheters to simulate various physical activity levels.

show abstract

Inter- and Intraoperator Variability in Measurement of On-Site CT-derived Fractional Flow Reserve Based on Structural and Fluid Analysis: A Comprehensive Analysis

Cited by 11 publications

References 19 publications

Initial evaluation of a convolutional neural network used for noninvasive assessment of coronary artery disease severity from coronary computed tomography angiography data

Initial evaluation of a convolutional neural network used for noninvasive assessment of coronary artery disease severity from coronary computed tomography angiography data

Inter-Operator Reliability of an Onsite Machine Learning-Based Prototype to Estimate Ct Angiography-Derived Fractional Flow Reserve

Method to simulate distal flow resistance in coronary arteries in 3D printed patient specific coronary models

Contact Info

Product

Resources

About