Prognostic Value of Machine Learning-based Time-to-Event Analysis Using Coronary CT Angiography in Patients with Suspected Coronary Artery Disease.

Purpose: To assess the long-term prognostic value of a machine learning (ML) approach in time-to-event analyses incorporating coronary CT angiography (CCTA)-derived and clinical parameters in patients with suspected coronary artery disease. Materials and Methods: The retrospective analysis included patients with suspected coronary artery disease who underwent CCTA between October 2004 and December 2017. Major adverse cardiovascular events were defined as the composite of all-cause death, myocardial infarction, unstable angina, or late revascularization (>90 days after index scan). Clinical and CCTA-derived parameters were assessed as predictors of major adverse cardiovascular events and incorporated into two models: a Cox proportional hazards model with recursive feature elimination and an ML model based on random survival forests. Both models were trained and validated by employing repeated nested cross-validation. Harrell concordance index (C-index) was used to assess the predictive power. Results: A total of 5457 patients (mean age, 61 years ± 11 [SD]; 3648 male patients) were evaluated. The predictive power of the ML model (C-index, 0.74; 95% CI: 0.71, 0.76) was significantly higher than the Cox model (C-index, 0.71; 95% CI: 0.68, 0.74; P = .02). The ML model also outperformed the segment stenosis score (C-index, 0.69; 95% CI: 0.66, 0.72; P < .001), which was the best performing CCTA-derived parameter, and patient age (C-index, 0.66; 95% CI: 0.63, 0.69; P < .001), the best performing clinical parameter. Conclusion: An ML model for time-to-event analysis based on random survival forests had higher performance in predicting major adverse cardiovascular events compared with established clinical or CCTA-derived metrics and a conventional Cox model.Keywords: Machine Learning, CT Angiography, Cardiac, Arteries, Heart, Arteriosclerosis, Coronary Artery DiseaseSupplemental material is available for this article.© RSNA, 2023.

Convolutional neural networks on risk stratification of patients with suspected coronary artery disease undergoing coronary computed tomography angiography.

To assess the prognostic value of convolutional neural networks (CNN) on coronary computed tomography angiography (CCTA) in comparison to conventional computed tomography (CT) reporting and clinical risk scores. 5468 patients who underwent CCTA with suspected coronary artery disease (CAD) were included. Primary endpoint was defined as a composite of all-cause death, myocardial infarction, unstable angina or late revascularization (> 90 days after CCTA). Early revascularization was additionally included as a training endpoint for the CNN algorithm. Cardiovascular risk stratification was based on Morise score and the extent of CAD (eoCAD) as assessed on CCTA. Semiautomatic post-processing was performed for vessel delineation and annotation of calcified and non-calcified plaque areas. Using a two-step training of a DenseNet-121 CNN the entire network was trained with the training endpoint, followed by training the feature layer with the primary endpoint. During a median follow-up of 7.2 years, the primary endpoint occurred in 334 patients. CNN showed an AUC of 0.631 ± 0.015 for prediction of the combined primary endpoint, while combining it with conventional CT and clinical risk scores showed an improvement of AUC from 0.646 ± 0.014 (based on eoCAD only) to 0.680 ± 0.015 (p < 0.0001) and from 0.619 ± 0.0149 (based on Morise Score only) to 0.6812 ± 0.0145 (p < 0.0001), respectively. In a stepwise model including all prediction methods, it was found an AUC of 0.680 ± 0.0148. CNN analysis showed to improve conventional CCTA-derived and clinical risk stratification when evaluating CCTA of patients with suspected CAD.

Periprocedural myocardial injury according to optical characteristics of neointima and treatment modality of in-stent restenosis.

AIMS: Aim of the present study was to investigate the impact of increasing neointimal inhomogeneity and neoatherosclerosis as well as of treatment modality of in-stent restenosis (ISR) on the occurrence of periprocedural myocardial injury (PMI). METHODS AND RESULTS: Patients with normal or stable/falling increased baseline high-sensitivity troponin T (hs-cTnT) undergoing intravascular optical coherence tomography (OCT) and subsequent percutaneous coronary intervention (PCI) of ISR by means of drug-coated balloon (DCB) or drug-eluting stent (DES) were included. Overall, 128 patients were subdivided into low (n = 64) and high (n = 64) inhomogeneity groups, based on the median of distribution of non-homogeneous quadrants. No significant between-group differences were detected in terms of hs-cTnT changes (28.0 [12.0-65.8] vs. 25.5 [9.8-65.0] ng/L; p = 0.355), or the incidence of major PMI (31.2 vs. 31.2%; p = 1.000). Similarly, no differences were observed between DCB- and DES-treated groups in terms of hs-cTn changes (27.0 [10.0-64.0] vs. 28.0 [11.0-73.0] ng/L; p = 0.795), or the incidence of major PMI (28.9 vs. 35.6%; p = 0.566). Additionally, no significant interaction was present between optical neointimal characteristics and treatment modality in terms of changes in hs-cTnT (Pint = 0.432). No significant differences in PMI occurrence were observed between low and high neoatherosclerosis subgroups. CONCLUSIONS: In patients undergoing PCI for ISR, there was no association between increasing neointimal inhomogeneity, or increasing expression of neoatherosclerotic changes and occurrence of PMI. PMI occurrence was not influenced by the treatment modality (DCB vs. DES) of ISR lesions, a finding that supports the safety of DCB treatment for ISR.

Optical coherence tomography tissue coverage and characterization at six months after implantation of bioresorbable scaffolds versus conventional everolimus eluting stents in the ISAR-Absorb MI trial.

PURPOSE: Data regarding vessel healing by optical coherence tomography (OCT) after everolimus-eluting bioresorbable scaffolds (BRS) or everolimus-eluting metallic stent (EES) implantation in acute myocardial infarction (AMI) patients is scarce. We compared OCT findings after BRS or EES implantation in patients with AMI enrolled in a randomized trial. METHODS: In ISAR-Absorb MI, AMI patients were randomized to BRS or EES implantation, with 6-8 month angiographic follow-up. This analysis includes patients who underwent OCT during surveillance angiography. Tissue characterization was done using grey-scale signal intensity analysis. The association between OCT findings and target lesion failure (TLF) at 2 years was investigated. RESULTS: OCT was analyzed in 103 patients (2237 frames, 19,827 struts) at a median of 216 days post-implantation. Of these, 70 were treated with BRS versus 32 with EES. Pre-(92.8 vs. 68.7%, p = 0.002) and post-dilation (51.4 vs. 12.5%, p < 0.001) were more common in BRS as compared to EES. Strut coverage was higher in BRS vs. EES (97.5% vs. 90.9%, p < 0.001). Mean neointimal thickness was comparable in both groups [85.5 (61.9, 124.1) vs. 69.5 (32.7, 127.5) µm, respectively, p = 0.20]. Mature neointimal regions were numerically more common in BRS (43.0% vs. 24.6%; p = 0.35); this difference was statistically significant in ST-elevation myocardial infarction patients (40.9% vs. 21.1%, p = 0.03). At two-years, 8 (7.8%) patients experienced TLF. Mean neointimal area [0.61 (0.21, 1.33) vs. 0.41 (0.11, 0.75) mm2, p = 0.03] and mean neointimal coverage [106.1 (65.2, 214.8) vs. 80.5 (53.5, 122.1) µm, p < 0.01] were higher, with comparable tissue maturity, in lesions with versus without TLF. CONCLUSIONS: In selected patients who underwent OCT surveillance 6-8 months after coronary intervention for AMI with differing implantation characteristics depending on the device type used, vessel healing was more advanced in BRS compared with EES, particularly in the STEMI subgroup.

Clinical outcomes by optical characteristics of neointima and treatment modality in patients with coronary in-stent restenosis.

BACKGROUND: Drug-coated balloons (DCB) and drug-eluting stents (DES) represent the currently recommended treatments for in-stent restenosis (ISR). Optical coherence tomography (OCT) allows detailed neointimal characterisation which can guide treatment strategies. AIMS: The aims of this study were first, to assess the relation between neointimal pattern and clinical outcomes following in-stent restenosis (ISR) treatment, and second, to explore a potential interaction between neointimal pattern and treatment modality relative to clinical outcomes. METHODS: Patients undergoing OCT-guided treatment (DCB or DES) of ISR in three European centres were included. Based on the median of distribution of non-homogeneous neointima quadrants, patients were categorised into low and high inhomogeneity groups. RESULTS: A total of 197 patients (low inhomogeneity=100 and high inhomogeneity=97) were included. There were no significant differences in terms of major adverse cardiac events (MACE) (p=0.939) or target lesion revascularisation (TLR) (p=0.732) between the two groups. The exploratory analysis showed a significant interaction between neointimal pattern and treatment modality regarding MACE (pint=0.006) and TLR (pint=0.022). DES showed a significant advantage over DCB in the high (MACE: HR 0.26 [0.10-0.65], p=0.004; TLR: HR 0.28 [0.11-0.69], p=0.006), but not in the low inhomogeneity group (MACE: p=0.917; TLR: p=0.797). CONCLUSIONS: In patients with ISR treated with DCB or DES, there were no significant differences in terms of MACE or TLR between the low and high inhomogeneity groups. A significant interaction was observed between treatment modality and neointimal pattern with an advantage of DES over DCB in the high and no difference in the low inhomogeneity group. This warrants confirmation from prospective dedicated studies.