Dataset Information

Angiography-Based Machine Learning for Predicting Fractional Flow Reserve in Intermediate Coronary Artery Lesions.

ABSTRACT: Background An angiography-based supervised machine learning ( ML ) algorithm was developed to classify lesions as having fractional flow reserve ≤0.80 versus >0.80. Methods and Results With a 4:1 ratio, 1501 patients with 1501 intermediate lesions were randomized into training versus test sets. Between the ostium and 10 mm distal to the target lesion, a series of angiographic lumen diameter measurements along the centerline was plotted. The 24 computed angiographic features based on the diameter plot and 4 clinical features (age, sex, body surface area, and involve segment) were used for ML by XGBoost. The model was independently trained and tested by 2000 bootstrap iterations. External validation with 79 patients was conducted. Including all 28 features, the ML model with 5-fold cross-validation in the 1204 training samples predicted fractional flow reserve ≤0.80 with overall diagnostic accuracy of 78±4% (averaged area under the curve: 0.84±0.03). The 12 high-ranking features selected by scatter search were involved segment; body surface area; distal lumen diameter; minimal lumen diameter; length of a lumen diameter <2.0 mm, <1.5 mm, and <1.25 mm; mean lumen diameter within the worst segment; sex; diameter stenosis; distal 5-mm reference lumen diameter; and length of diameter stenosis >70%. Using those 12 features, the ML predicted fractional flow reserve ≤0.80 in the test set with sensitivity of 84%, specificity of 80%, and overall accuracy of 82% (area under the curve: 0.87). The averaged diagnostic accuracy in bootstrap replicates was 81±1% (averaged area under the curve: 0.87±0.01). External validation showed accuracy of 85% (area under the curve: 0.87). Conclusions Angiography-based ML showed good diagnostic performance in identifying ischemia-producing lesions and reduced the need for pressure wires.

SUBMITTER: Cho H

PROVIDER: S-EPMC6405668 | biostudies-other | 2019 Feb

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Angiography-Based Machine Learning for Predicting Fractional Flow Reserve in Intermediate Coronary Artery Lesions.

Cho Hyungjoo H Lee June-Goo JG Kang Soo-Jin SJ Kim Won-Jang WJ Choi So-Yeon SY Ko Jiyuon J Min Hyun-Seok HS Choi Gun-Ho GH Kang Do-Yoon DY Lee Pil Hyung PH Ahn Jung-Min JM Park Duk-Woo DW Lee Seung-Whan SW Kim Young-Hak YH Lee Cheol Whan CW Park Seong-Wook SW Park Seung-Jung SJ

Journal of the American Heart Association 20190201 4

Background An angiography-based supervised machine learning ( ML ) algorithm was developed to classify lesions as having fractional flow reserve ≤0.80 versus >0.80. Methods and Results With a 4:1 ratio, 1501 patients with 1501 intermediate lesions were randomized into training versus test sets. Between the ostium and 10 mm distal to the target lesion, a series of angiographic lumen diameter measurements along the centerline was plotted. The 24 computed angiographic features based on the diameter ...[more]

PMID: 30764731

Similar Datasets

Project description:Purpose To compare the diagnostic accuracy of different computed tomographic (CT) fractional flow reserve (FFR) algorithms for vessels with intermediate stenosis. Materials and Methods This cross-sectional HIPAA-compliant and human research committee-approved study applied a four-step CT FFR algorithm in 61 patients (mean age, 69 years ± 10; age range, 29-89 years) with a lesion of intermediate-diameter stenosis (25%-69%) at CT angiography who underwent FFR measurement within 90 days. The per-lesion diagnostic performance of CT FFR was tested for three different approaches to estimate blood flow distribution for CT FFR calculation. The first two, the Murray law and the Huo-Kassab rule, used coronary anatomy; the third used contrast material opacification gradients. CT FFR algorithms and CT angiography percentage diameter stenosis (DS) measurements were compared by using the area under the receiver operating characteristic curve (AUC) to detect FFRs of 0.8 or lower. Results Twenty-five lesions (41%) had FFRs of 0.8 or lower. The AUC of CT FFR determination by using contrast material gradients (AUC = 0.953) was significantly higher than that of the Huo-Kassab (AUC = 0.882, P = .043) and Murray law models (AUC = 0.871, P = .033). All three AUCs were higher than that for 50% or greater DS at CT angiography (AUC = 0.596, P < .001). Correlation of CT FFR with FFR was highest for gradients (Spearman ρ = 0.80), followed by the Huo-Kassab rule (ρ = 0.68) and Murray law (ρ = 0.67) models. All CT FFR algorithms had small biases, ranging from -0.015 (Murray) to -0.049 (Huo-Kassab). Limits of agreement were narrowest for gradients (-0.182, 0.147), followed by the Huo-Kassab rule (-0.246, 0.149) and the Murray law (-0.285, 0.256) models. Conclusion Clinicians can perform CT FFR by using a four-step approach on site to accurately detect hemodynamically significant intermediate-stenosis lesions. Estimating blood flow distribution by using coronary contrast opacification variations may improve CT FFR accuracy. © RSNA, 2017 Online supplemental material is available for this article.

Project description:Backgroundintravascular ultrasound (IVUS) and fractional flow reserve (FFR) have both been shown to be superior to angiography in optimizing percutaneous coronary intervention (PCI). However, there is still a lack of comparative studies between PCI optimization using physiology and intravascular imaging head-to-head. The aim of this study was to compare the effectiveness of FFR and IVUS PCI optimization strategies on the functional PCI result (assessed with FFR) immediately post-PCI and at 9-12 months after the treatment of long coronary lesions.MethodsThis was a single-center study comparing post-PCI FFR between two different PCI optimization strategies (FFR and IVUS). The study included 154 patients who had hemodynamically significant long lesions, necessitating a stent length of 30 mm or more. The procedural outcomes were functional PCI result immediately post-PCI and at 9-12 months after treatment. Clinical outcomes included target vessel failure (TVF) and functional target vessel restenosis rate during follow-up.ResultsBaseline clinical characteristics and FFR (0.65 [0.55-0.71]) did not differ significantly between the two groups and the left anterior descending artery was treated in 82% of cases. The FFR optimization strategy resulted in a significantly shorter stented segment (49 mm vs. 63 mm, p = 0.001) compared to the IVUS optimization strategy. Although the rates of optimal functional PCI result (FFR > 0.9) did not significantly differ between the FFR and IVUS optimization strategies, a proportion of patients in the FFR group (12%) experienced poor post-PCI functional outcome with FFR values ≤ 0.8, which was not observed in the IVUS group. At the 9-12 month follow-up, 20% of patients in the FFR group had target-vessel-related myocardial ischemia, compared to 6% in the IVUS group. The rates of TVF and functional target vessel restenosis during follow-up were also numerically higher in the FFR optimization group.ConclusionsThe use of FFR PCI optimization strategy in the treatment of long coronary artery lesions is associated with a higher incidence of poor functional PCI result and larger myocardial ischemia burden at follow-up compared to the IVUS optimization strategy. However, this discrepancy did not translate into a statistically significant difference in clinical outcomes. This study highlights the importance of using IVUS to optimize long lesions functional PCI outcomes.

Project description:Data on the clinical utility of coronary computed tomography angiography-derived fractional flow reserve (FFRCT) are sparse. In patients with intermediate (40-70%) coronary stenosis determined by coronary computed tomography angiography, we investigated the association of replacing standard myocardial perfusion imaging with FFRCT testing with downstream utilization of invasive coronary angiography (ICA) and the diagnostic yield of ICA (rate of no obstructive disease, and rate of revascularization). This was a single-center observational study of symptomatic patients with suspected coronary artery disease referred to coronary computed tomography angiography between 2013 and 2015. Patients were divided into 3 historical groups based on the adjunctive functional testing approach: myocardial perfusion imaging (n=1332) or FFRCT "implementation" (n=800) or "clinical use" (n=1391). Propensity score matching was used to estimate the average period effect on outcomes. Patients in the FFRCT clinical use group versus the myocardial perfusion imaging group were older and had higher pretest probability of obstructive disease. After adjusting for baseline risk characteristics, there was a reduction in downstream ICA utilization (absolute risk difference: -4.2; 95% CI, -6.9 to -1.6; P=0.002). In patients referred to ICA, findings of no obstructive coronary artery disease decreased (-12.8%; 95% CI, -22.2 to -3.4; P=0.008) and rate of coronary revascularization increased (14.1%; 95% CI, 3.3-24.9; P=0.01), as did availability of functional information for guidance of revascularization (27.8%; 95% CI, 11.3-44.4; P<0.001) after clinical adoption of FFRCT. Replacing adjunctive myocardial perfusion imaging with FFRCT testing for functional assessment of intermediate stenosis determined by coronary computed tomography angiography in stable coronary artery disease was associated with less ICA utilization, and a higher ICA diagnostic yield. The findings in this observational study needs confirmation in prospective, randomized trials.

Dataset Information

Angiography-Based Machine Learning for Predicting Fractional Flow Reserve in Intermediate Coronary Artery Lesions.

Publications

Angiography-Based Machine Learning for Predicting Fractional Flow Reserve in Intermediate Coronary Artery Lesions.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets