Developing heart sound classification algorithms for murmur identification is critical for early screening of heart diseases. However, identifying murmurs in long-duration heart sound signals can be challenging due to their weak features and interference from noise. Considering the periodic patterns of heart sounds and murmurs, periodic priors can be introduced to enhance murmur identification, an approach that remains underutilized in current methods. In this study, we propose a novel Dual-level Periodic Pattern-Aware Transformer (DPPAT) to implicitly leverage the periodic priors of heart sound signals without requiring cycle segmentation. In the regional-level, an Adaptive Period-Aligned Window Selection algorithm is designed for the model to extract periodic components while suppressing random noise using a Periodic Pattern Attention module. In the global-level, the model further integrates these periodic features in global-modeling to enhance the identification of murmur-discriminative features. Validated on the dataset from 2022 George B. Moody PhysioNet Challenge, our proposed method achieves a weighted accuracy of 84.27% and an F1-score of 70.38% through 10-fold cross-validation. The generalizability of DPPAT is further verified on two additional public datasets, including both heart sound and respiratory sound signals. Furthermore, attention visualizations provide a clear understanding of the focus of the model, highlighting the decision-making basis for murmur identification.

Background Accurate evaluation of coronary artery disease is critical for guiding treatment decisions, particularly in complex coronary lesions. Fractional flow reserve (FFR) remains the gold standard for assessing lesion-specific ischemia but is invasive and requires pharmacological hyperemia. Noninvasive alternatives, such as quantitative flow ratio (QFR) from coronary angiography and ultrasonic flow ratio (UFR) from intravascular ultrasound (IVUS), offer promising diagnostic approaches. Objectives This study aimed to compare the diagnostic performance of UFR and QFR against FFR in assessing complex coronary lesions. Methods This retrospective multicenter study included 217 patients (220 vessels) who underwent IVUS and FFR. UFR was derived from IVUS imaging, and QFR was calculated using coronary angiography data. Correlation, agreement, and diagnostic metrics (sensitivity, specificity) were analyzed, with receiver operating characteristic curves assessing accuracy. Results UFR demonstrated stronger correlation with FFR (r = 0.79; 95% CI: 0.74-0.84; P < 0.001) compared with QFR (r = 0.68; 95% CI: 0.60-0.74; P < 0.001). UFR also showed better diagnostic performance, with an area under the receiver operating characteristic curve of 0.91 (95% CI: 0.86-0.94) compared with QFR’s 0.86 (95% CI: 0.81-0.90). In complex lesions (diffuse, bifurcation, calcified), UFR consistently outperformed QFR, particularly in bifurcation and lesions heavily calcified, where QFR accuracy dropped significantly (72.5% vs 86.8%, P = 0.001). Conclusions In this retrospective hypothesis-generating study, UFR showed numerically higher diagnostic accuracy than QFR in complex coronary lesions. These findings suggest UFR may have potential as a complementary tool for functional assessment, but definitive conclusions about superiority require validation in larger prospective studies. (Comparison of UFR With QFR in Stable Coronary Artery Disease; NCT06322355)

Aims:Coronary computed tomography angiography (CCTA) enables a non-invasive, comprehensive assessment of coronary artery disease, and artificial intelligence (AI) offers the potential to improve CCTA image interpretation. This study aimed to evaluate the performance of an AI-powered method for automatic plaque quantification from CCTA, with optical coherence tomography (OCT) as reference standard. Methods and results:Patients who underwent CCTA within 6 months prior to OCT were retrospectively enrolled. AI-assisted automatic plaque quantification was performed on CCTA with specific plaque composition classification based on adaptive Hounsfield unit thresholds. Qualitative high-risk plaque features were also assessed. Automated co-registration of CCTA and OCT was performed with the link of invasive coronary angiography. A total of 91 patients with 153 co-registered lesions were evaluated. The AI-assisted automatic CCTA analysis showed significant correlations with OCT for quantifying plaque volume/burden and different plaque compositions (all P values <0.001); of which, the correlation coefficient for plaque volume was 0.84. Vulnerable plaque, defined as lipid-to-cap ratio >0.33 on OCT, was identified in 39 (25.5%) lesions. CCTA-derived plaque volume >82.5 mm3 [odds ratio (OR), 9.39], maximal plaque burden >76.4% (OR, 3.70), lipidic tissue volume >16.3 mm³ (OR, 4.42), all P < 0.001, and high-risk plaque features ≥2 (OR, 2.70, P = 0.009) were independent predictors of OCT-derived vulnerable plaques. The average time for automatic CCTA plaque quantification was 1.8 min per patient. Conclusion:The novel AI-powered method facilitated fully automatic plaque quantification and correlated well with co-registered OCT.

BACKGROUND:Ultrasonic flow ratio (UFR) is an artificial intelligence-powered method that derives fractional flow reserve (FFR) from intravascular ultrasound (IVUS) imaging. Although retrospective core laboratory studies have demonstrated its diagnostic accuracy, prospective on-site validation remains unexplored. OBJECTIVES:The aim of this study was to evaluate the diagnostic accuracy of on-site UFR for identifying hemodynamically significant coronary stenosis, using wire-based FFR as the reference standard. METHODS:Consecutive patients with ≥1 de novo lesion exhibiting 50% to 80% diameter stenosis and reference diameter ≥2.5 mm were prospectively enrolled. After FFR measurement, IVUS pull backs were acquired and analyzed on site using dedicated software, with analysts blinded to FFR results. Minimal luminal area (MLA) was simultaneously available during UFR computation. The prespecified primary endpoint was the on-site diagnostic accuracy of UFR for identifying FFR ≤0.80. RESULTS:Between February 2023 and November 2024, 112 patients (138 pull backs) were enrolled; after the exclusion of 6 patients (7 pull backs), 106 patients with 131 vessels remained for analysis. Median FFR was 0.84 (Q1-Q3: 0.78-0.90), with 43 of 131 lesions (32.8%) showing FFR ≤0.80. UFR achieved diagnostic accuracy of 94% (95% CI: 88%-97%), significantly exceeding the prespecified target of 78% (P < 0.001). Compared with IVUS-derived MLA, UFR demonstrated superior sensitivity (88% [95% CI: 75%-96%] vs 47% [95% CI: 31%-62%]; P < 0.001) and specificity (97% [95% CI: 90%-99%] vs 84% [95% CI: 75%-91%]; P = 0.003). The corresponding positive and negative predictive values were 93% (95% CI: 81%-97%) and 94% (95% CI: 88%-97%), respectively. CONCLUSIONS:This study achieved its prespecified primary goal by demonstrating high on-site diagnostic accuracy of UFR in identifying hemodynamically significant coronary stenosis. (Functional Comprehensive Assessment by IVUS Reconstruction in Patients With Suspected Ischemic Heart Disease [FEATURE]; NCT05694065).

The registration of coronary artery structures from preoperative coronary computed tomography angiography to intraoperative coronary angiography is of great interest to improve guidance in percutaneous coronary interventions. However, non-rigid deformation and discrepancies in both dimensions and topology between the two imaging modalities present a challenge in the 2D/3D coronary artery registration. In this study, we address this problem by formulating it as a centerline feature matching task and propose a GNN-based vessel matching network (GVM-Net) to establish dense correspondence between different image modalities in an end-to-end manner. GVM-Net considers centerline points as nodes in graphs and effectively models the complex topological relationships between them through attention mechanisms and message passing. Furthermore, by incorporating redundant rows and columns into the matching matrix, GVM-Net can effectively handle inconsistencies in vascular structures. We also introduce the query-based nodes grouping module, which clusters nodes in the feature space to further explore the topological relationships. GVM-Net achieves an average F1-score of 89.74% with a mean pixel distance of 0.48 pixels on the synthetic dataset with 276 data pairs and an average F1-score of 83.35% with a mean error of 1.52 mm in 55 manually labeled clinical cases, both exceeding existing feature matching methods.

Background: Accurate assessment of coronary artery disease is essential for guiding clinical decision-making, particularly in cases involving complex coronary lesions. Fractional Flow Reserve (FFR) remains the reference standard for lesion-specific ischemia evaluation; however, it is invasive and requires pharmacologically induced hyperemia. Emerging non-invasive alternatives, including Quantitative Flow Ratio (QFR) derived from coronary angiography (CAG) and Ultrasonic Flow Ratio (UFR) derived from intravascular ultrasound (IVUS), offer promising diagnostic value. This study aimed to compare the diagnostic performance of UFR and QFR against FFR in the assessment of complex coronary lesions. Methods: In this retrospective, multicenter study, 217 patients (220 vessels) with coronary artery lesions who underwent both intravascular ultrasound (IVUS) and FFR measurement were included. UFR was derived from IVUS imaging, while QFR was computed using coronary angiography (CAG) data. Correlation coefficients, agreement analyses, and diagnostic metrics including sensitivity and specificity were employed to evaluate the performance of UFR and QFR against FFR, with receiver operating characteristic (ROC) curve analysis used to assess diagnostic accuracy. Results: UFR demonstrated a stronger correlation with FFR (r = 0.79, p < 0.001) compared to QFR (r = 0.68, p < 0.001). Moreover, UFR exhibited superior diagnostic performance, with an area under the ROC curve (AUC) of 0.91, exceeding that of QFR (AUC = 0.86). In subsets of complex lesions—specifically diffuse, bifurcation, and heavily calcified lesions—UFR consistently outperformed QFR, with the most pronounced difference observed in bifurcation and calcified lesions where QFR accuracy was significantly reduced (72.5% vs. 86.8%, p = 0.001). Conclusion: UFR provides enhanced diagnostic accuracy compared to QFR for complex coronary lesions and represents a reliable, non-invasive alternative to FFR, particularly in challenging anatomical scenarios such as bifurcation and heavily calcified lesions. The clinical integration of UFR may reduce the reliance on invasive FFR measurements while preserving diagnostic precision. ### Competing Interest Statement The authors have declared no competing interest. ### Clinical Trial ClinicalTrials.gov ([NCT06322355][1]). ### Funding Statement This work was supported by grant from Shanghai Pujiang Program (No. 22PJD011 and 22PJD012). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: he study received approval from the institutional review boards at Zhongshan Hospital in Shanghai, Shanghai 7th People's Hospital, and Zhengzhou 7th People's Hospital I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data will be made available on request. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT06322355&atom=%2Fmedrxiv%2Fearly%2F2025%2F04%2F19%2F2025.04.16.25325975.atom

The lumen and external elastic lamina contour delineation is crucial for quantitative analyses of intravascular ultrasound (IVUS) images. However, the various artifacts in IVUS images pose substantial challenges for accurate delineation. Existing mask-based methods often produce anatomically implausible contours in artifact-affected images, while contour-based methods suffer from the over-smooth problem within the artifact regions. In this paper, we directly regress the contour pairs instead of mask-based segmentation. A coupled contour representation is adopted to learn a low-dimensional contour signature space, where the embedded anatomical prior enables the model to avoid producing unreasonable results. Further, a PIoU loss is proposed to capture the overall shape of the contour points and maximize the similarity between the regressed contours and manually delineated contours with various irregular shapes, alleviating the over-smooth problem. For the images with severe artifacts, a difficulty-aware training strategy is designed for contour regression, which gradually guides the model focus on hard samples and improves contour localization accuracy. We evaluate the proposed framework on a large IVUS dataset, consisting of 7204 frames from 185 pullbacks. The mean Dice similarity coefficients of the method for the lumen and external elastic lamina are 0.951 and 0.967, which significantly outperforms other state-of-the-art (SOTA) models. All regressed contours in the test images are anatomically plausible. On the public IVUS-2011 dataset, the proposed method attains comparable performance to the SOTA models with the highest processing speed at 100 fps. The code is available at https://github.com/SMU-MedicalVision/ContourRegression.

BACKGROUND Recent randomized clinical trials have demonstrated the benefits of intravascular imaging (IVI)-guided percutaneous coronary intervention (PCI) over angiography-guided PCI. However, the role of angiography-based physiological assessment during IVI-guided PCI remains unclear. OBJECTIVES This study aimed to explore the discrepancies and significance of angiography-based physiological assessments in IVI-guided PCI. METHODS In the international multicenter randomized FLAVOUR (Fractional Flow Reserve and Intravascular Ultrasound for Clinical Outcomes in Patients With Intermediate Stenosis) trial, angiography-based physiological assessment was retrospectively performed using the Murray law-based quantitative flow ratio (mu QFR). In this post hoc analysis, patients were categorized based on intravascular ultrasound (IVUS)-guided treatment decisions (PCI or deferral) and mu QFR as follows: negative mu QFR with deferral of PCI (DEFER), negative mu QFR with PCI (PERFORM), and positive mu QFR with PCI (REFERENCE). The primary outcome was major adverse cardiovascular events, defined as a composite of death, myocardial infarction, and target vessel revascularization at the 24-month follow-up. RESULTS Of the 784 patients, 34.4% (270/784), 29.3% (230/784), and 31.5% (247/784) were categorized into the DEFER, PERFORM, and REFERENCE groups, respectively. Physiological assessment led to substantial reclassification, encompassing 48.2% (230/477) of patients who underwent IVUS-guided PCI. The REFERENCE group showed a higher risk for major adverse cardiovascular events at 2 years compared with the PERFORM group (adjusted HR: 2.46; 95% CI: 1.13-5.35; P = 0.023). However, the primary outcomes in the DEFER and PERFORM groups were similar (adjusted HR: 0.88; 95% CI: 0.37-2.11; P = 0.779). The quality of life at 2 years was comparable among the 3 groups (P = 0.198). CONCLUSIONS Angiography-based physiological assessments can offer additional prognostic insights for patients undergoing IVI-guided PCI. IVUS-guided PCI may not be advantageous in patients with functionally insignificant lesions. (JACC Cardiovasc Interv. 2025;18:145-153) (c) 2025 by the American College of Cardiology Foundation.

Background Coronary physiology measured by fractional flow reserve (FFR) is superior to angiography for assessing the efficacy of percutaneous coronary intervention (PCI). Yet, the clinical adoption of post-PCI FFR is limited. Murray law-based quantitative flow ratio (μQFR) may represent a promising alternative, as it can quickly compute FFR from a single angiographic view. Objectives The authors aimed to investigate the potential role of post-PCI μQFR in predicting clinical outcomes. Methods This was a post hoc blinded analysis of the FLAVOUR trial. Patients with angiographically intermediate lesions randomized 1:1 to receive FFR or intravascular ultrasound-guided PCI were included. Post-PCI μQFR was assessed in successfully stented vessels, blinded to clinical outcomes. Suboptimal physiological outcome post-PCI was defined a priori as post-PCI μQFR <0.90. The primary endpoint was 2-year target vessel failure, including cardiac death, target vessel myocardial infarction, and target vessel revascularization. Secondary endpoints included the diagnostic concordance of pre-PCI μQFR with FFR in the FFR-guidance arm. Results Post-PCI μQFR was successfully analyzed in 806 vessels from 777 participants (feasibility 97.0% [806 of 831]). Suboptimal physiological outcome post-PCI was identified in 24.7% (199 of 806) of vessels and post-PCI μQFR <0.90 was associated with higher risk of 2-year target vessel failure (6.1% [12 of 199] vs 2.7% [16 of 607]; HR: 2.45 [95% CI: 1.14-5.26]; P = 0.022). Pre-PCI μQFR was obtained in 877 of 919 vessels (feasibility 95.4%), showing 90% accuracy, 82% sensitivity, and 94% specificity for identifying physiologically significant stenosis defined by pre-PCI FFR ≤0.80. Conclusions In patients with intermediate lesions who underwent PCI with contemporary imaging or physiology guidance, lower post-PCI μQFR values predict subsequent adverse events. (Fractional FLow Reserve And IVUS for Clinical OUtcomes in Patients With InteRmediate Stenosis [FLAVOUR]; NCT02673424)

Background: Optical flow ratio (OFR) is a novel computational fractional flow reserve derived from optical coherence tomography (OCT). However, the impact of combining post-stenting morphology (OCT) and physiology (OFR) remains largely unknown. Methods: OCT and OFR were analysed at an independent core laboratory. Target lesion failure (TLF) was defined as the composite of cardiac death, target lesion myocardial infarction, and target lesion revascularisation. Suboptimal stent deployment was identified with at least 1 TLF-related OCT or OFR characteristic. Results: A total of 448 patients with acute coronary syndrome (459 vessels) were assessed. Stent expansion < 80%, minimal stent area < 4.5 mm(2), stent edge lipid-rich plaque and OFR < 0.90 were independent predictors of TLF (all P < 0.001). Patients with OCTsuboptimal (adjusted hazard ratio [aHR] 7.88, 95% CI 2.73-22.72,-P < 0.001) or OFR-suboptimal (aHR 5.78, 95% CI 2.54-13.14; P < 0.001) stent deployment showed significantly higher risk of TLF compared with those with optimal stent deployment, with a significant interaction (Pinteraction < 0.001). OCT and OFR bothesuboptimal stent deployment was confirmed as an independent predictor of TLF (aHR 9.39, 95% CI 4.25-20.76; P < 0.001). Conclusions: Combined OCT and OFR conferred an optimal reclassification of stent deployment, which may aid in decision making regarding a tailored PCI strategy for optimal stent deployment.

Abstract Background Despite substantial advancements in coronary stent systems and implantation techniques, target vessel failure (TVF) remains a clinical issue to be addressed. Purpose We used deep learning algorithms for automatic stent analysis of intravascular ultrasound (IVUS) data to identify stent implantation characteristics associated with TVF. Methods In patients from the IVUS-guided group of the FLAVOUR trial who had undergone successful stenting, validated deep learning algorithms were applied to IVUS images to delineate plaque and stent strut for the quantification of stent expansion, apposition, and residual plaque burden. Stent overexpansion was defined as stent area exceeding 120% of the hypothetical healthy lumen area. Murray law-based quantitative flow ratio (μQFR) was also assessed from angiographic images. The primary endpoint was TVF at 2 years, defined as a composite of cardiac death, target vessel-related myocardial infarction, and ischemia-driven target vessel revascularization. Results A total of 459 vessels in 441 patients were included, with TVF events occurred in 15 vessels (3.3%) during the 2-year follow-up. Stent analysis was performed in 687,379 stent struts from 70,369 IVUS image frames of the interrogated vessels with the assistance of specific deep learning algorithms. On vessel-level multivariate analysis adjusting for procedural covariates, stent overexpansion length >5 mm (hazard ratio, 4.57; 95% confidence interval [CI], 1.31 to 15.92; P=0.017), residual plaque burden at proximal stent edge >50% (hazard ratio, 3.34; 95% CI, 1.19 to 9.33; P=0.022), and post-PCI μQFR <0.90 (hazard ratio, 5.09; 95% CI, 1.62 to 15.97; P=0.005) were more likely to be present in vessels with TVF events than in those without. Conclusions Besides the established indexes of post-procedural residual plaque burden and low μQFR, deep learning-powered IVUS stent overexpansion was significantly associated with increased risk of TVF.Suboptimal stenting results and TVF

BACKGROUND:The FORZA trial (FFR or OCT Guidance to Revascularize Intermediate Coronary Stenosis Using Angioplasty) prospectively compared the use of fractional flow reserve (FFR) or optical coherence tomography (OCT) for treatment decisions and percutaneous coronary intervention (PCI) optimization in patients with angiographically intermediate coronary lesions. Murray law-based quantitative-flow-ratio (μQFR) is a novel noninvasive method for the computation of FFR. In the present study, we evaluated the clinical impact of μQFR, FFR, or OCT guidance in FORZA trial lesions at 3-year follow-up. METHODS:μQFR was assessed at baseline and, in the case of a decision to intervene, after (FFR- or OCT-guided) PCI. The baseline μQFR was considered the final μQFR for deferred lesions, and post-PCI μQFR value was taken as final for stented lesions. The primary end point was target vessel failure ([TVF]; cardiac death, target-vessel-related myocardial infarction, and target-vessel-revascularization) at a 3-year follow-up. RESULTS:A total of 419 vessels (199 OCT-guided and 220 FFR-guided) were included in the FORZA trial. μQFR was evaluated in 256 deferred lesions and 159 treated lesions (98 OCT-guided PCI and 61 FFR-guided PCI). In treated lesions, post-PCI μQFR was higher in OCT-group compared with FFR-group (median, 0.93 versus 0.91; P=0.023), and the post-PCI μQFR improvement was greater in FFR-group (0.14 versus 0.08; P<0.0001). At 3-year follow-up, OCT- and FFR-guided treatment decisions resulted in comparable TVF rate (6.7% versus 7.9%; P=0.617). Final μQFR was the only predictor of TVF. μQFR ≤0.89 was associated with 3× increase in TVF (11.6% versus 3.7%; P=0.004). PCI was a predictor of higher final μQFR (odds ratio, 0.22 [95% CI, 0.14-0.34]; P<0.001). CONCLUSIONS:In vessels with angiographically intermediate coronary lesions, OCT-guided PCI resulted in comparable clinical outcomes as FFR-guided PCI. μQFR estimated at the end of diagnostic or interventional procedure predicted 3-year TVF. REGISTRATION:URL: https://www.clinicaltrials.gov; Unique identifier: NCT01824030.

Acute coronary syndromes (ACS) are one of the leading causes of mortality worldwide, with atherosclerotic plaque rupture and subsequent thrombus formation as the main underlying substrate. Thrombus burden evaluation is important for tailoring treatment therapy and predicting prognosis. Coronary optical coherence tomography (OCT) enables in-vivo visualization of thrombus that cannot otherwise be achieved by other image modalities. However, automatic quantification of thrombus on OCT has not been implemented. The main challenges are due to the variation in location, size and irregularities of thrombus in addition to the small data set. In this paper, we propose a novel dual-coordinate cross-attention transformer network, termed DCCAT, to overcome the above challenges and achieve the first automatic segmentation of thrombus on OCT. Imaging features from both Cartesian and polar coordinates are encoded and fused based on long-range correspondence via multi-head cross-attention mechanism. The dual-coordinate cross-attention block is hierarchically stacked amid convolutional layers at multiple levels, allowing comprehensive feature enhancement. The model was developed based on 5,649 OCT frames from 339 patients and tested using independent external OCT data from 548 frames of 52 patients. DCCAT achieved Dice similarity score (DSC) of 0.706 in segmenting thrombus, which is significantly higher than the CNN-based (0.656) and Transformer-based (0.584) models. We prove that the additional input of polar image not only leverages discriminative features from another coordinate but also improves model robustness for geometrical transformation.Experiment results show that DCCAT achieves competitive performance with only 10% of the total data, highlighting its data efficiency. The proposed dual-coordinate cross-attention design can be easily integrated into other developed Transformer models to boost performance.

BACKGROUND:The recently introduced ultrasonic flow ratio (UFR), is a novel fast computational method to derive fractional flow reserve (FFR) from intravascular ultrasound (IVUS) images. In the present study, we evaluate the diagnostic performance of UFR in patients with intermediate left main (LM) stenosis. METHODS:This is a prospective, single center study enrolling consecutive patients with presence of intermediated LM lesions (diameter stenosis of 30%-80% by visual estimation) underwent IVUS and FFR measurement. An independent core laboratory assessed offline UFR and IVUS-derived minimal lumen area (MLA) in a blinded fashion. RESULTS:Both UFR and FFR were successfully achieved in 41 LM patients (mean age, 62.0 ± 9.9 years, 46.3% diabetes). An acceptable correlation between UFR and FFR was identified (r = 0.688, P < 0.0001), with an absolute numerical difference of 0.03 (standard difference: 0.01). The area under the curve (AUC) in diagnosis of physiologically significant coronary stenosis for UFR was 0.94 (95% CI: 0.87-1.01), which was significantly higher than angiographic identified stenosis > 50% (AUC = 0.66, P < 0.001) and numerically higher than IVUS-derived MLA (AUC = 0.82; P = 0.09). Patient level diagnostic accuracy, sensitivity and specificity for UFR to identify FFR ≤ 0.80 was 82.9% (95% CI: 70.2-95.7), 93.1% (95% CI: 82.2-100.0), 58.3% (95% CI: 26.3-90.4), respectively. CONCLUSION:In patients with intermediate LM diseases, UFR was proved to be associated with acceptable correlation and high accuracy with pressure wire-based FFR as standard reference. The present study supports the use of UFR for functional evaluation of intermediate LM stenosis.

Introduction: Ultrasonic flow ratio (UFR) is a novel intravascular ultrasound (IVUS)-derived method for fast computation of fractional flow reserve (FFR) without pressure wires and adenosine. Aims: The aim of this study was to evaluate the diagnostic performance of UFR and compare it with angiography-based quantitative flow ratio (QFR), using FFR as the reference standard. Methods: Patients who underwent coronary angiography, IVUS and FFR were analyzed. UFR and QFR was computed offline in a central core-lab by independent analysts blinded to FFR. Results: A total of 111 paired comparisons between UFR, QFR and FFR from 56 patients were analyzed. UFR showed a numerically better correlation (r=0.83 vs. 0.79; p=0.27) and significant better agreement (standard deviation of the difference=0.07 vs. 0.08; p=0.02) with FFR than QFR. Diagnostic accuracy, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio and negative likelihood ratio for UFR to predict FFR was 92% (95% CI: 85%-96%), 85% (95% CI: 71%-94%), 97% (95% CI: 89%-99%), 95% (95% CI: 83%-99%), 90% (95% CI: 82%-95%), 27.6 (95% CI: 7.0-108.4), and 0.16 (95% CI: 0.08-0.31), respectively. The area under curve for UFR to predict an FFR ≤ 0.80 was 0.94, is equivalent to QFR (difference=0.04; p=0.23), and higher than minimal lumen area (difference=0.09, p=0.006). Diagnostic accuracy was not significantly different in bifurcation lesions, nor in non-bifurcation lesions. Conclusions: The UFR showed good diagnostic concordance with FFR, equivalent to QFR. The good diagnostic performance of UFR provides a potentiality for the integration of physiological assessment and intravascular imaging in real-world clinical practice.

Coronary artery disease (CAD) is the leading cause of death globally. The 3D fusion of coronary X-ray angiography (XA) and optical coherence tomography (OCT) provides complementary information to appreciate coronary anatomy and plaque morphology. This significantly improve CAD diagnosis and prognosis by enabling precise hemodynamic and computational physiology assessments. The challenges of fusion lie in the potential misalignment caused by the foreshortening effect in XA and non-uniform acquisition of OCT pullback. Moreover, the need for reconstructions of major bifurcations is technically demanding. This paper proposed an automated 3D fusion framework AutoFOX, which consists of deep learning model TransCAN for 3D vessel alignment. The 3D vessel contours are processed as sequential data, whose features are extracted and integrated with bifurcation information to enhance alignment via a multi-task fashion. TransCAN shows the highest alignment accuracy among all methods with a mean alignment error of 0.99 ± 0.81 mm along the vascular sequence, and only 0.82 ± 0.69 mm at key anatomical positions. The proposed AutoFOX framework uniquely employs an advanced side branch lumen reconstruction algorithm to enhance the assessment of bifurcation lesions. A multi-center dataset is utilized for independent external validation, using the paired 3D coronary computer tomography angiography (CTA) as the reference standard. Novel morphological metrics are proposed to evaluate the fusion accuracy. Our experiments show that the fusion model generated by AutoFOX exhibits high morphological consistency with CTA. AutoFOX framework enables automatic and comprehensive assessment of CAD, especially for the accurate assessment of bifurcation stenosis, which is of clinical value to guiding procedure and optimization.

BACKGROUND:The combination of coronary imaging assessment and blood flow perturbation estimation has the potential to improve percutaneous coronary intervention (PCI) guidance.OBJECTIVES:We aimed to evaluate a novel method for fast computation of Murray law-based quantitative flow ratio (μQFR) from coregistered optical coherence tomography (OCT) and angiography (OCT-modulated μQFR, OCT-μQFR) in predicting physiological efficacy of PCI.METHODS:Patients treated by OCT-guided PCI in the OCT-arm of the Fractional Flow Reserve versus Optical Coherence Tomography to Guide RevasculariZAtion of Intermediate Coronary Stenoses trial (FORZA, NCT01824030) were included. Based on angiography and OCT before PCI, simulated residual OCT-μQFR was computed by assuming full stent expansion to the intended-to-treat segment. Plaque composition was automatically characterized using a validated artificial intelligence algorithm. Actual post-PCI OCT-μQFR pullback was computed based on coregistration of angiography and OCT acquired immediately after PCI. Suboptimal functional stenting result was defined as OCT-μQFR ≤ 0.90.RESULTS:Paired simulated residual OCT-μQFR and actual post-PCI OCT-μQFR were obtained in 76 vessels from 74 patients. Simulated residual OCT-μQFR showed good correlation (r = 0.80, p < 0.001), agreement (mean difference = -0.02 ± 0.02, p < 0.001), and diagnostic concordance (79%, 95% confidence interval: 70%-88%) with actual post-PCI OCT-μQFR. Actual post-PCI in-stent OCT-μQFR had a median value of 0.02 and was associated with left anterior descending artery lesion location (β = 0.38, p < 0.001), higher baseline total plaque burden (β = 0.25, p = 0.031), and fibrous plaque volume (β = 0.24, p = 0.026).CONCLUSIONS:This study based on patients enrolled in a prospective OCT-guidance PCI trial shows that simulated residual OCT-μQFR had good correlation, agreement, and diagnostic concordance with actual post-PCI OCT-μQFR. In OCT-guided procedures, OCT-μQFR in-stent pressure drop was low and was significantly predicted by pre-PCI vessel/plaque characteristics.

BACKGROUND Optical flow ratio (OFR) is a novel method for the fast computation of fractional flow reserve (FFR) from optical coherence tomography. AIMS We aimed to evaluate the diagnostic accuracy of OFR in assessing intermediate coronary stenosis using wire-based FFR as the reference. METHODS We performed an individual patient-level meta-analysis of all available studies with paired OFR and FFR assessments. The primary outcome was vessel-level diagnostic concordance of the OFR and FFR, using a cut-off of ≤0.80 to define ischaemia and ≤0.90 to define suboptimal post-percutaneous coronary intervention (PCI) physiology. This meta-analysis was registered in PROSPERO (CRD42021287726). RESULTS Five studies were finally included, providing 574 patients and 626 vessels (404 pre-PCI and 222 post-PCI) with paired OFR and FFR from 9 international centres. Vessel-level diagnostic concordance of the OFR and FFR was 91% (95% confidence interval [CI]: 88%-94%), 87% (95% CI: 82%-91%), and 90% (95% CI: 87%-92%) in pre-PCI, post-PCI, and overall, respectively. The overall sensitivity, specificity, and positive and negative predictive values were 84% (95% CI: 79%-88%), 94% (95% CI: 92%-96%), 90% (95% CI: 86%-93%), and 89% (95% CI: 86%-92%), respectively. Multivariate logistic regression indicated that a low pullback speed (odds ratio [OR] 7.02, 95% CI: 1.68-29.43; p=0.008) was associated with a higher risk of obtaining OFR values at least 0.10 higher than FFR. Increasing the minimal lumen area was associated with a lower risk of obtaining an OFR at least 0.10 lower than FFR (OR 0.39, 95% CI: 0.18-0.82; p=0.013). CONCLUSIONS This individual patient data meta-analysis demonstrated a high diagnostic accuracy of OFR. OFR has the potential to provide an improved integration of intracoronary imaging and physiological assessment for the accurate evaluation of coronary artery disease.

Automatic delineation of the lumen and vessel contours in intravascular ultrasound (IVUS) images is crucial for the subsequent IVUS-based analysis. Existing methods usually address this task through mask-based segmentation, which cannot effectively handle the anatomical plausibility of the lumen and external elastic lamina (EEL) contours and thus limits their performance. In this article, we propose a contour encoding based method called coupled contour regression network (CCRNet) to directly predict the lumen and EEL contour pairs. The lumen and EEL contours are resampled, coupled, and embedded into a low-dimensional space to learn a compact contour representation. Then, we employ a convolutional network backbone to predict the coupled contour signatures and reconstruct the signatures to the object contours by a linear decoder. Assisted by the implicit anatomical prior of the paired lumen and EEL contours in the signature space and contour decoder, CCRNet has the potential to avoid producing unreasonable results. We evaluated our proposed method on a large IVUS dataset consisting of 7204 cross-sectional frames from 185 pullbacks. The CCRNet can rapidly extract the contours at 100 fps. Without any post-processing, all produced contours are anatomically reasonable in the test 19 pullbacks. The mean Dice similarity coefficients of our CCRNet for the lumen and EEL are 0.940 and 0.958, which are comparable to the mask-based models. In terms of the contour metric Hausdorff distance, our CCRNet achieves 0.258 mm for lumen and 0.268 mm for EEL, which outperforms the mask-based models.