Machine Learning Model Comparison for Automatic Segmentation of Intracoronary Optical Coherence Tomography and Plaque Cap Thickness Quantification

Optical coherence tomography(OCT)is a new intravascular imaging technique with high resolution and could provide accurate morphological information for plaques in coronary arteries.However,its segmentation is still commonly performed manually by experts which is time-consuming.The aim of this study was to develop automatic techniques to characterize plaque components and quantify plaque cap thickness using 3 machine learning methods including convolutional neural network(CNN)with U-Net architecture,CNN with Fully convolutional DenseNet(FC-DenseNet)architecture and support vector machine(SVM).In vivo OCT and intravascular ultrasound(IVUS)images were acquired from two patients at Emory University with informed consent obtained.Eighteen OCT image slices which included lipid core and with acceptable image quality were selected for our study.Manual segmentation from imaging experts was used as the gold standard for model training and validation.Since OCT has limited penetration,virtual histology IVUS was combined with OCT data to improve reliability.A 3-fold cross-validation method was used for model training and validation.The overall tissue classification accuracy for the 18 slices studied(total classification database sample size was 8580096 pixels)was 96.36%and 92.72%for U-Net and FC-DenseNet,respectively.The best average prediction accuracy for lipid was 91.29%based on SVM,compared to 82.84%and 78.91%from U-Net and FC-DenseNet,respectively.The overall average accuracy(Acc)differentiating lipid and fibrous tissue were 95.58%,92.33%and 81.84%for U-Net,FC-DenseNet and SVM,respectively.The average errors of U-Net,FC-DenseNet and SVM from the 18 slices for cap thickness quantification were 8.83%,10.71%and 15.85%.The average relative errors of minimum cap thickness from 18 slices of U-Net,FC-DenseNet and SVM were 17.46%,13.06%and 22.20%,respectively.To conclude,CNN-based segmentation methods can better characterize plaque compositions and quantify plaque cap thickness on OCT images and are more likely to be used in the clinical arena.Large-scale studies are needed to further develop the methods and validate our findings.

Intensive statin versus low-dose statin + ezetimibe treatment for fibrous cap thickness of coronary vulnerable plaques

Background:Acute coronary syndromes mainly result from abrupt thrombotic occlusion caused by atherosclerotic vulnerable plaques (VPs) that suddenly rupture or erosion. Fibrous cap thickness (FCT) is a major determinant of the propensity of a VP to rupture and is recognized as a key factor. The intensive use of statins is known to have the ability to increase FCT;however, there is a risk of additional adverse effects. However, lower dose statin with ezetimibe is known to be tolerable by patients. The present study aimed to investigate the effect of intensive statin vs. low-dose stain + ezetimibe therapy on FCT, as evaluated using optical coherence tomography. Method:Patients who had VPs (minimum FCT <65 μm and lipid core >90°) and deferred from intervention in our single center from January 2014 to December 2018 were included in the trial. They were divided into the following two groups: intensive statin group (rosuvastatin 15-20 mg or atorvastatin 30-40 mg) and combination therapy group (rosuvastatin 5-10 mg or atorvastatin 10-20 mg + ezetimibe 10 mg). At the 12-month follow-up, we compared the change in the FCT (ΔFCT%) between the two groups and analyzed the association of ΔFCT% with risk factors. Fisher exact test was used for all categorical variables. Student’s t test or Mann-Whitney U-test was used for analyzing the continuous data. The relationship between ΔFCT% and risk factors was analyzed using linear regression analysis. Result:Total 53 patients were finally enrolled, including 26 patients who were in the intensive statin group and 27 who were in the combination therapy group. At the 12-month follow-up, the serum levels of total cholesterol (TC), total triglyceride, low-density lipoprotein (LDL-C), hypersensitive C-reactive protein (hs-CRP), and lipoprotein-associated phospholipase A2 (Lp-PLA2) levels were reduced in both the groups. The ΔTC%, ΔLDL-C%, and ΔLp-PLA2% were decreased further in the combination therapy group. FCT was increased in both the groups (combination treatment group vs. intensive statin group: 128.89 ± 7.64 vs. 110.19 ± 7.00 μm, t = -9.282, P < 0.001) at the 12-month follow-up. The increase in ΔFCT% was more in the combination therapy group (123.46% ± 14.05% vs. 91.14% ± 11.68%, t = -9.085, P < 0.001). Based on the multivariate linear regression analysis, only the serum Lp-PLA2 at the 12-month follow-up ( B = -0.203, t = -2.701, P = 0.010), ΔTC% ( B = -0.573, t = -2.048, P = 0.046), and Δhs-CRP% ( B = -0.302, t = -2.963, P = 0.005) showed an independent association with ΔFCT%. Conclusions:Low-dose statin combined with ezetimibe therapy maybe provide a profound and significant increase in FCT as compared to intensive statin monotherapy. The reductions in Lp-PLA2, ΔTC%, and Δhs-CRP% are independently associated with an increase in FCT.