Anatomically aware dual-hop learning for pulmonary embolism detection in CT pulmonary angiograms.

Pulmonary Embolisms (PE) represent a leading cause of cardiovascular death. While medical imaging, through computed tomographic pulmonary angiography (CTPA), represents the gold standard for PE diagnosis, it is still susceptible to misdiagnosis or significant diagnosis delays, which may be fatal for critical cases. Despite the recently demonstrated power of deep learning to bring a significant boost in performance in a wide range of medical imaging tasks, there are still very few published researches on automatic pulmonary embolism detection. Herein we introduce a deep learning based approach, which efficiently combines computer vision and deep neural networks for pulmonary embolism detection in CTPA. Our method brings novel contributions along three orthogonal axes: (1) automatic detection of anatomical structures; (2) anatomical aware pretraining, and (3) a dual-hop deep neural net for PE detection. We obtain state-of-the-art results on the publicly available multicenter large-scale RSNA dataset.

Co-registered optical coherence tomography and X-ray angiography for the prediction of fractional flow reserve.

Cardiovascular disease (CVD) stands as the leading global cause of mortality, and coronary artery disease (CAD) has the highest prevalence, contributing to 42% of these fatalities. Recognizing the constraints inherent in the anatomical assessment of CAD, Fractional Flow Reserve (FFR) has emerged as a pivotal functional diagnostic metric. Herein, we assess the potential of employing an ensemble approach with deep neural networks (DNN) to predict invasively measured Fractional Flow Reserve (FFR) using raw anatomical data extracted from both optical coherence tomography (OCT) and X-ray coronary angiography (XA). In this study, we used a challenging dataset, with 46% of the lesions falling within the FFR range of 0.75 to 0.85. Despite this complexity, our model achieved an accuracy of 84.3%, demonstrating a sensitivity of 87.5% and a specificity of 81.4%. Our results demonstrate that incorporating both OCT and XA signals, co-registered, as inputs for the DNN model leads to an important increase in overall accuracy.

Assessment of the functional severity of coronary lesions from optical coherence tomography based on ensembled learning.

BACKGROUND: Atherosclerosis is one of the most frequent cardiovascular diseases. The dilemma faced by physicians is whether to treat or postpone the revascularization of lesions that fall within the intermediate range given by an invasive fractional flow reserve (FFR) measurement. The paper presents a monocentric study for lesions significance assessment that can potentially cause ischemia on the large coronary arteries. METHODS: A new dataset is acquired, comprising the optical coherence tomography (OCT) images, clinical parameters, echocardiography and FFR measurements collected from 80 patients with 102 lesions, with stable multivessel coronary artery disease. Having the ground truth given by the invasive FFR measurement, the dataset is challenging because almost 40% of the lesions are in the gray zone, having an FFR value between 0.75 and 0.85. Twenty-six features are extracted from OCT images, clinical characteristics, and echocardiography and the most relevant are identified by examining the models' accuracy. An ensembled learning is performed for solving the binary classification problem of lesion significance considering the leave-one-out cross-validation approach. RESULTS: Ensemble models are designed from the multi-features voting from 5 features models by prediction aggregation with a maximum accuracy of 81.37% and a maximum area under the curve score (AUC) of 0.856. CONCLUSIONS: The proposed explainable supervised learning-based lesion classification is a new method that can be improved by training with a larger multicenter dataset for further designing a tool for guiding the decision making of the clinician for the cases outside the gray zone and for the other situation extra clinical information about the lesion is needed.

In-stent restenosis in acute coronary syndrome-a classic and a machine learning approach.

Background: In acute coronary syndrome (ACS), a number of previous studies tried to identify the risk factors that are most likely to influence the rate of in-stent restenosis (ISR), but the contribution of these factors to ISR is not clearly defined. Thus, the need for a better way of identifying the independent predictors of ISR, which comes in the form of Machine Learning (ML). Objectives: The aim of this study is to evaluate the relationship between ISR and risk factors associated with ACS and to develop and validate a nomogram to predict the probability of ISR through the use of ML in patients undergoing percutaneous coronary intervention (PCI). Methods: Consecutive patients presenting with ACS who were successfully treated with PCI and who had an angiographic follow-up after at least 3 months were included in the study. ISR risk factors considered into the study were demographic, clinical and peri-procedural angiographic lesion risk factors. We explored four ML techniques (Random Forest (RF), support vector machines (SVM), simple linear logistic regression (LLR) and deep neural network (DNN)) to predict the risk of ISR. Overall, 21 features were selected as input variables for the ML algorithms, including continuous, categorical and binary variables. Results: The total cohort of subjects included 340 subjects, in which the incidence of ISR observed was 17.68% (n = 87). The most performant model in terms of ISR prediction out of the four explored was RF, with an area under the receiver operating characteristic (ROC) curve of 0.726. Across the predictors herein considered, only three predictors were statistically significant, precisely, the number of affected arteries (≥2), stent generation and diameter. Conclusion: ML models applied in patients after PCI can contribute to a better differentiation of the future risk of ISR.

Prevalence and Predictors of Renal Disease in a National Representative Sample of the Romanian Adult Population: Data from the SEPHAR IV Survey.

Background: The prevalence of chronic kidney disease (CKD) correlates with the prevalence of hypertension (HT). We studied the prevalence and predictors of CKD in a representative sample of the Romanian adult population. Methods: A sample of 1470 subjects were enrolled in the SEPHAR IV (Study for the Evaluation of Prevalence of Hypertension and Cardiovascular Risk) survey. All subjects were evaluated for blood pressure (BP) and extensive evaluations of target organ damage, blood, and urine samples were undertaken. Results: A total of 883 subjects were included in the statistical analysis. Those experiencing CKD with an eGFR < 60 mL/min/1.73 m2 were older at 71.94 ± 7.4 years (n = 19, 2.15%) compared with those without renal impairment at 50.3 ± 16.21 years (n = 864, 97.85%), p < 0.0001. The prevalence of CKD among hypertensives (379 from 883) was 4.49% (17/379), while 17 out of 19 subjects with CKD had HT (89.47%). After adjusting for age, sex, and diabetic status, only serum uric acid (SUR) > 6.9 mg/dL (OR: 6.61; 95% CI: 2.063, 10.83; p = 0.004) was an independent risk factor and a predictor of CKD. Conclusions: The prevalence of CKD in hypertensive Romanian adults was more than ten times higher than in the normotensive population. Levels of SUR > 6.9 mg/dL were predictors of CKD.

Distinctive Morphological Patterns of Complicated Coronary Plaques in Acute Coronary Syndromes: Insights from an Optical Coherence Tomography Study.

Optical coherence tomography (OCT) is an ideal imaging technique for assessing culprit coronary plaque anatomy. We investigated the morphological features and mechanisms leading to plaque complication in a single-center observational retrospective study on 70 consecutive patients with an established diagnosis of acute coronary syndrome (ACS) who underwent OCT imaging after coronary angiography. Three prominent morphological entities were identified. Type I or intimal discontinuity, which was found to be the most common mechanism leading to ACS and was seen in 35 patients (50%), was associated with thrombus (68.6%; p = 0.001), mostly affected the proximal plaque segment (60%; p = 0.009), and had no distinctive underlying plaque features. Type II, a significant stenosis with vulnerability features (inflammation in 16 patients, 84.2%; thin-cap fibroatheroma (TCFA) in 10 patients, 52.6%) and a strong association with lipid-rich plaques (94.7%; p = 0.002), was observed in 19 patients (27.1%). Type III, a protrusive calcified nodule, which was found to be the dominant morphological pattern in 16 patients (22.9%), was found in longer plaques (20.8 mm vs. 16.8 mm ID vs. 12.4 mm SS; p = 0.04) and correlated well with TCFA (93.8%; p = 0.02) and inflammation (81.3%). These results emphasize the existence of a wide spectrum of coronary morphological patterns related to ACS.

Artificial intelligence and cloud based platform for fully automated PCI guidance from coronary angiography-study protocol.

Ischemic heart disease represent a heavy burden for the medical systems irrespective of the methods used for diagnosis and treatment of such patients in the daily medical routine. The present paper depicts the protocol of a study whose main aim is to develop, implement and test an artificial intelligence algorithm and cloud based platform for fully automated PCI guidance using coronary angiography images. We propose the utilisation of multiple artificial intelligence based models to produce three-dimensional coronary anatomy reconstruction and assess function- post-PCI FFR computation- for developing an extensive report describing and motivating the optimal PCI strategy selection. All the relevant artificial intelligence model outputs (anatomical and functional assessment-pre- and post-PCI) are presented to the clinician via a cloud platform, who can then take the utmost treatment decision. The physician will be provided with multiple scenarios and treatment possibilities for the same case allowing a real-time evaluation of the most appropriate PCI strategy planning and follow-up. The artificial intelligence algorithms and cloud based PCI selection workflow will be verified and validated in a pilot clinical study including subjects prospectively to compare the artificial intelligence services and results against annotations and invasive measurements.

Nine-Year Trends in Atrial Fibrillation Prevalence among Romanian Adult Hypertensives: A Post-Hoc Analysis of SEPHAR II-IV Surveys.

Objectives: There are limited epidemiological data regarding atrial fibrillation (AF) in hypertensive (HT) Romanian adults. We sought to evaluate AF prevalence trends in the SEPHAR surveys (Study for Evaluation of Prevalence of Hypertension and Cardiovascular Risk in an Adult Population in Romania) during a nine-year interval (2012−2016−2021). Methods: Three consecutive editions of a national epidemiological survey regarding HT included representative samples of subjects stratified by age, gender and area of residence (SEPHAR II-IV­in total, 5422 subjects, mean age 48.69 ± 16.65 years, 57.5% (n = 3116) females). A post-hoc analysis of AF prevalence and oral anticoagulation (OAC) rates was performed. AF definition was based on a documented medical history of AF and/or AF documentation by study electrocardiogram. Results: General AF prevalence was 5.5% (n = 297). AF prevalence in HT subjects was 8.9% (n = 209) and has risen since SEPHAR II­7.2% (n = 57) and SEPHAR III­8.1% (n = 72) to SEPHAR IV­11.8% (n = 80), respectively (p = 0.001). AF prevalence has increased in HT males (SEPHAR II­5.3% (n = 19), SEPHAR III­7.6% (n = 26) and SEPHAR IV­11.7% (n = 35) (p = 0.010)) and in HT from urban areas (SEPHAR II­7.8% (n = 37), SEPHAR III­7.8% (n = 40), SEPHAR IV­14.7% (n = 50), p < 0.001). In SEPHAR III-IV, only 19.3% (n = 23) of HT AF patients with OAC indication were anticoagulated. Conclusions: AF prevalence has increased by ~64% in hypertensive Romanian adults between 2012 and 2021. However, anticoagulation strategies may be suboptimal in patients with cardioembolic risk.

Improving robustness of automatic cardiac function quantification from cine magnetic resonance imaging using synthetic image data.

Although having been the subject of intense research over the years, cardiac function quantification from MRI is still not a fully automatic process in the clinical practice. This is partly due to the shortage of training data covering all relevant cardiovascular disease phenotypes. We propose to synthetically generate short axis CINE MRI using a generative adversarial model to expand the available data sets that consist of predominantly healthy subjects to include more cases with reduced ejection fraction. We introduce a deep learning convolutional neural network (CNN) to predict the end-diastolic volume, end-systolic volume, and implicitly the ejection fraction from cardiac MRI without explicit segmentation. The left ventricle volume predictions were compared to the ground truth values, showing superior accuracy compared to state-of-the-art segmentation methods. We show that using synthetic data generated for pre-training a CNN significantly improves the prediction compared to only using the limited amount of available data, when the training set is imbalanced.

Personalized Pre- and Post-Operative Hemodynamic Assessment of Aortic Coarctation from 3D Rotational Angiography.

PURPOSE: Coarctation of Aorta (CoA) is a congenital disease consisting of a narrowing that obstructs the systemic blood flow. This proof-of-concept study aimed to develop a framework for automatically and robustly personalizing aortic hemodynamic computations for the assessment of pre- and post-intervention CoA patients from 3D rotational angiography (3DRA) data. METHODS: We propose a framework that combines hemodynamic modelling and machine learning (ML) based techniques, and rely on 3DRA data for non-invasive pressure computation in CoA patients. The key features of our framework are a parameter estimation method for calibrating inlet and outlet boundary conditions, and regional mechanical wall properties, to ensure that the computational results match the patient-specific measurements, and an improved ML based pressure drop model capable of predicting the instantaneous pressure drop for a wide range of flow conditions and anatomical CoA variations. RESULTS: We evaluated the framework by investigating 6 patient datasets, under pre- and post-operative setting, and, since all calibration procedures converged successfully, the proposed approach is deemed robust. We compared the peak-to-peak and the cycle-averaged pressure drop computed using the reduced-order hemodynamic model with the catheter based measurements, before and after virtual and actual stenting. The mean absolute error for the peak-to-peak pressure drop, which is the most relevant measure for clinical decision making, was 2.98 mmHg for the pre- and 2.11 mmHg for the post-operative setting. Moreover, the proposed method is computationally efficient: the average execution time was of only [Formula: see text] minutes on a standard hardware configuration. CONCLUSION: The use of 3DRA for hemodynamic modelling could allow for a complete hemodynamic assessment, as well as virtual interventions or surgeries and predictive modeling. However, before such an approach can be used routinely, significant advancements are required for automating the workflow.

Importance of Visual Estimation of Coronary Artery Stenoses and Use of Functional Evaluation for Appropriate Guidance of Coronary Revascularization-Multiple Operator Evaluation.

BACKGROUND: Visual estimation (VE) of coronary stenoses is the first step during invasive coronary angiography. The aim of this study was to evaluate the accuracy of VE together with invasive functional assessment (IFA) in defining the functional significance (FS) of coronary stenoses based on the opinion of multiple operators. METHODS: Fourteen independent operators visually evaluated 133 coronary lesions which had a previous FFR measurement, indicating the degree of stenosis (DS), FS and IFA intention. We determined the accuracy of FS prediction using several scenarios combining individual and group decision, considering IFA as deemed necessary by the operator or only in intermediate lesions. RESULTS: The accuracy of VE in predicting FS was largely variable between operators (average 66.1%); it improved significantly when IFA was used either as per operator's opinion (86.3%; p < 0.0001) or only in intermediate DS (82.9; p < 0.0001). There was no significant difference between using IFA per observer's opinion or only in intermediate DS lesions (p = 0.166). The poorest accuracy of VE for FS was obtained in intermediate DS lesions (59.1%). CONCLUSIONS: There are significant inter-observer differences in reporting the degree of DS, while the accuracy of VE prediction of FS is also largely dependent on the operator, and the worst performance is obtained in the evaluation of intermediate DS.

A framework for validating AI in precision medicine: considerations from the European ITFoC consortium.

BACKGROUND: Artificial intelligence (AI) has the potential to transform our healthcare systems significantly. New AI technologies based on machine learning approaches should play a key role in clinical decision-making in the future. However, their implementation in health care settings remains limited, mostly due to a lack of robust validation procedures. There is a need to develop reliable assessment frameworks for the clinical validation of AI. We present here an approach for assessing AI for predicting treatment response in triple-negative breast cancer (TNBC), using real-world data and molecular -omics data from clinical data warehouses and biobanks. METHODS: The European "ITFoC (Information Technology for the Future Of Cancer)" consortium designed a framework for the clinical validation of AI technologies for predicting treatment response in oncology. RESULTS: This framework is based on seven key steps specifying: (1) the intended use of AI, (2) the target population, (3) the timing of AI evaluation, (4) the datasets used for evaluation, (5) the procedures used for ensuring data safety (including data quality, privacy and security), (6) the metrics used for measuring performance, and (7) the procedures used to ensure that the AI is explainable. This framework forms the basis of a validation platform that we are building for the "ITFoC Challenge". This community-wide competition will make it possible to assess and compare AI algorithms for predicting the response to TNBC treatments with external real-world datasets. CONCLUSIONS: The predictive performance and safety of AI technologies must be assessed in a robust, unbiased and transparent manner before their implementation in healthcare settings. We believe that the consideration of the ITFoC consortium will contribute to the safe transfer and implementation of AI in clinical settings, in the context of precision oncology and personalized care.

An Automated Workflow for Hemodynamic Computations in Cerebral Aneurysms.

In recent years, computational fluid dynamics (CFD) has become a valuable tool for investigating hemodynamics in cerebral aneurysms. CFD provides flow-related quantities, which have been shown to have a potential impact on aneurysm growth and risk of rupture. However, the adoption of CFD tools in clinical settings is currently limited by the high computational cost and the engineering expertise required for employing these tools, e.g., for mesh generation, appropriate choice of spatial and temporal resolution, and of boundary conditions. Herein, we address these challenges by introducing a practical and robust methodology, focusing on computational performance and minimizing user interaction through automated parameter selection. We propose a fully automated pipeline that covers the steps from a patient-specific anatomical model to results, based on a fast, graphics processing unit- (GPU-) accelerated CFD solver and a parameter selection methodology. We use a reduced order model to compute the initial estimates of the spatial and temporal resolutions and an iterative approach that further adjusts the resolution during the simulation without user interaction. The pipeline and the solver are validated based on previously published results, and by comparing the results obtained for 20 cerebral aneurysm cases with those generated by a state-of-the-art commercial solver (Ansys CFX, Canonsburg PA). The automatically selected spatial and temporal resolutions lead to results which closely agree with the state-of-the-art, with an average relative difference of only 2%. Due to the GPU-based parallelization, simulations are computationally efficient, with a median computation time of 40 minutes per simulation.

Deep neural networks for ECG-free cardiac phase and end-diastolic frame detection on coronary angiographies.

Invasive coronary angiography (ICA) is the gold standard in Coronary Artery Disease (CAD) imaging. Detection of the end-diastolic frame (EDF) and, in general, cardiac phase detection on each temporal frame of a coronary angiography acquisition is of significant importance for the anatomical and non-invasive functional assessment of CAD. This task is generally performed via manual frame selection or semi-automated selection based on simultaneously acquired ECG signals - thus introducing the requirement of simultaneous ECG recordings. In this paper, we evaluate the performance of a purely image based workflow relying on deep neural networks for fully automated cardiac phase and EDF detection on coronary angiographies. A first deep neural network (DNN), trained to detect coronary arteries, is employed to preselect a subset of frames in which coronary arteries are well visible. A second DNN predicts cardiac phase labels for each frame. Only in the training and evaluation phases for the second DNN, ECG signals are used to provide ground truth labels for each angiographic frame. The networks were trained on 56,655 coronary angiographies from 6820 patients and evaluated on 20,780 coronary angiographies from 6261 patients. No exclusion criteria related to patient state (stable or acute CAD), previous interventions (PCI or CABG), or pathology were formulated. Cardiac phase detection had an accuracy of 98.8 %, a sensitivity of 99.3 % and a specificity of 97.6 % on the evaluation set. EDF prediction had a precision of 98.4 % and a recall of 97.9 %. Several sub-group analyses were performed, indicating that the cardiac phase detection performance is largely independent from acquisition angles, the heart rate of the patient, and the angiographic view (LCA / RCA). The average execution time of cardiac phase detection for one angiographic series was on average less than five seconds on a standard workstation. We conclude that the proposed image based workflow potentially obviates the need for manual frame selection and ECG acquisition, representing a relevant step towards automated CAD assessment.

Applying Deep Neural Networks over Homomorphic Encrypted Medical Data.

In recent years, powered by state-of-the-art achievements in a broad range of areas, machine learning has received considerable attention from the healthcare sector. Despite their ability to provide solutions within personalized medicine, strict regulations on the confidentiality of patient health information have in many cases hindered the adoption of deep learning-based solutions in clinical workflows. To allow for the processing of sensitive health information without disclosing the underlying data, we propose a solution based on fully homomorphic encryption (FHE). The considered encryption scheme, MORE (Matrix Operation for Randomization or Encryption), enables the computations within a neural network model to be directly performed on floating point data with a relatively small computational overhead. We consider the well-known MNIST digit recognition problem to evaluate the feasibility of the proposed method and show that performance does not decrease when deep learning is applied on MORE homomorphic data. To further evaluate the suitability of the method for healthcare applications, we first train a model on encrypted data to estimate the outputs of a whole-body circulation (WBC) hemodynamic model and then provide a solution for classifying encrypted X-ray coronary angiography medical images. The findings highlight the potential of the proposed privacy-preserving deep learning methods to outperform existing approaches by providing, within a reasonable amount of time, results equivalent to those achieved by unencrypted models. Lastly, we discuss the security implications of the encryption scheme and show that while the considered cryptosystem promotes efficiency and utility at a lower security level, it is still applicable in certain practical use cases.

Use of 3D rotational angiography to perform computational fluid dynamics and virtual interventions in aortic coarctation.

Computational fluid dynamics (CFD) can be used to analyze blood flow and to predict hemodynamic outcomes after interventions for coarctation of the aorta and other cardiovascular diseases. We report the first use of cardiac 3-dimensional rotational angiography for CFD and show not only feasibility but also validation of its hemodynamic computations with catheter-based measurements in three patients.

Privacy-Preserving Artificial Intelligence: Application to Precision Medicine.

Motivated by state-of-the-art performances across a wide variety of areas, over the last few years Machine Learning has drawn a significant amount of attention from the healthcare domain. Despite their potential in enabling person-alized medicine applications, the adoption of Deep Learning based solutions in clinical workflows has been hindered in many cases by the strict regulations concerning the privacy of patient health data. We propose a solution that relies on Fully Homomorphic Encryption, particularly on the MORE scheme, as a mechanism for enabling computations on sensitive health data, without revealing the underlying data. The chosen variant of the encryption scheme allows for the computations in the Neural Network model to be directly performed on floating point numbers, while incurring a reasonably small computational overhead. For feasibility evaluation, we demonstrate on the MNIST digit recognition task that Deep Learning can be performed on encrypted data without compromising the accuracy. We then address a more complex task by training a model on encrypted data to estimate the outputs of a whole-body circulation (WBC) model. These results underline the potential of the proposed approach to outperform current solutions by delivering comparable results to the unencrypted Deep Learning based solutions, in a reasonable amount of time. Lastly, the security aspects of the encryption scheme are analyzed, and we show that, even though the chosen encryption scheme favors performance and utility at the cost of weaker security, it can still be used in certain practical applications.

Diagnostic Accuracy of a Machine-Learning Approach to Coronary Computed Tomographic Angiography-Based Fractional Flow Reserve: Result From the MACHINE Consortium.

BACKGROUND: Coronary computed tomographic angiography (CTA) is a reliable modality to detect coronary artery disease. However, CTA generally overestimates stenosis severity compared with invasive angiography, and angiographic stenosis does not necessarily imply hemodynamic relevance when fractional flow reserve (FFR) is used as reference. CTA-based FFR (CT-FFR), using computational fluid dynamics (CFD), improves the correlation with invasive FFR results but is computationally demanding. More recently, a new machine-learning (ML) CT-FFR algorithm has been developed based on a deep learning model, which can be performed on a regular workstation. In this large multicenter cohort, the diagnostic performance ML-based CT-FFR was compared with CTA and CFD-based CT-FFR for detection of functionally obstructive coronary artery disease. METHODS AND RESULTS: At 5 centers in Europe, Asia, and the United States, 351 patients, including 525 vessels with invasive FFR comparison, were included. ML-based and CFD-based CT-FFR were performed on the CTA data, and diagnostic performance was evaluated using invasive FFR as reference. Correlation between ML-based and CFD-based CT-FFR was excellent (R=0.997). ML-based (area under curve, 0.84) and CFD-based CT-FFR (0.84) outperformed visual CTA (0.69; P<0.0001). On a per-vessel basis, diagnostic accuracy improved from 58% (95% confidence interval, 54%-63%) by CTA to 78% (75%-82%) by ML-based CT-FFR. The per-patient accuracy improved from 71% (66%-76%) by CTA to 85% (81%-89%) by adding ML-based CT-FFR as 62 of 85 (73%) false-positive CTA results could be correctly reclassified by adding ML-based CT-FFR. CONCLUSIONS: On-site CT-FFR based on ML improves the performance of CTA by correctly reclassifying hemodynamically nonsignificant stenosis and performs equally well as CFD-based CT-FFR.