Machine learning assessment of myocardial ischemia using angiography: Development and retrospective validation

What characteristics of clinical decision support system implementations lead to adoption for regular use? A scoping review

INTRODUCTION

Digital healthcare innovation has yielded many prototype clinical decision support (CDS) systems, however, few are fully adopted into practice, despite successful research outcomes. We aimed to explore the characteristics of implementations in clinical practice to inform future innovation.

METHODS

Web of Science, Trip Database, PubMed, NHS Digital and the BMA website were searched for examples of CDS systems in May 2022 and updated in June 2023. Papers were included if they reported on a CDS giving pathway advice to a clinician, adopted into regular clinical practice and had sufficient published information for analysis. Examples were excluded if they were only used in a research setting or intended for patients. Articles found in citation searches were assessed alongside a detailed hand search of the grey literature to gather all available information, including commercial information. Examples were excluded if there was insufficient information for analysis. The normalisation process theory (NPT) framework informed analysis.

RESULTS

22 implemented CDS projects were included, with 53 related publications or sources of information (40 peer-reviewed publications and 13 alternative sources). NPT framework analysis indicated organisational support was paramount to successful adoption of CDS. Ensuring that workflows were optimised for patient care alongside iterative, mixed-methods implementation was key to engaging clinicians.

CONCLUSION

Extensive searches revealed few examples of CDS available for analysis, highlighting the implementation gap between research and healthcare innovation. Lessons from included projects include the need for organisational support, an underpinning mixed-methods implementation strategy and an iterative approach to address clinician feedback.

Patient-specific in silico 3D coronary model in cardiac catheterisation laboratories

Coronary artery disease is caused by the buildup of atherosclerotic plaque in the coronary arteries, affecting the blood supply to the heart, one of the leading causes of death around the world. X-ray coronary angiography is the most common procedure for diagnosing coronary artery disease, which uses contrast material and x-rays to observe vascular lesions. With this type of procedure, blood flow in coronary arteries is viewed in real-time, making it possible to detect stenoses precisely and control percutaneous coronary interventions and stent insertions. Angiograms of coronary arteries are used to plan the necessary revascularisation procedures based on the calculation of occlusions and the affected segments. However, their interpretation in cardiac catheterisation laboratories presently relies on sequentially evaluating multiple 2D image projections, which limits measuring lesion severity, identifying the true shape of vessels, and analysing quantitative data. In silico modelling, which involves computational simulations of patient-specific data, can revolutionise interventional cardiology by providing valuable insights and optimising treatment methods. This paper explores the challenges and future directions associated with applying patient-specific in silico models in catheterisation laboratories. We discuss the implications of the lack of patient-specific in silico models and how their absence hinders the ability to accurately predict and assess the behaviour of individual patients during interventional procedures. Then, we introduce the different components of a typical patient-specific in silico model and explore the potential future directions to bridge this gap and promote the development and utilisation of patient-specific in silico models in the catheterisation laboratories.

Non-invasive fractional flow reserve estimation using deep learning on intermediate left anterior descending coronary artery lesion angiography images

This study aimed to design an end-to-end deep learning model for estimating the value of fractional flow reserve (FFR) using angiography images to classify left anterior descending (LAD) branch angiography images with average stenosis between 50 and 70% into two categories: FFR > 80 and FFR ≤ 80. In this study 3625 images were extracted from 41 patients' angiography films. Nine pre-trained convolutional neural networks (CNN), including DenseNet121, InceptionResNetV2, VGG16, VGG19, ResNet50V2, Xception, MobileNetV3Large, DenseNet201, and DenseNet169, were used to extract the features of images. DenseNet169 indicated higher performance compared to other networks. AUC, Accuracy, Sensitivity, Specificity, Precision, and F1-score of the proposed DenseNet169 network were 0.81, 0.81, 0.86, 0.75, 0.82, and 0.84, respectively. The deep learning-based method proposed in this study can non-invasively and consistently estimate FFR from angiographic images, offering significant clinical potential for diagnosing and treating coronary artery disease by combining anatomical and physiological parameters.

Artificial intelligence in cardiovascular medicine: An updated review of the literature

Screening and early detection of cardiovascular disease (CVD) are crucial for managing progress and preventing related morbidity. In recent years, several studies have reported the important role of Artificial intelligence (AI) technology and its integration into various medical sectors. AI applications are able to deal with the massive amounts of data (medical records, ultrasounds, medications, and experimental results) generated in medicine and identify novel details that would otherwise be forgotten in the mass of healthcare data sets. Nowadays, AI algorithms are currently used to improve diagnosis of some CVDs including heart failure, atrial fibrillation, hypertrophic cardiomyopathy and pulmonary hypertension. This review summarized some AI concepts, critical execution requirements, obstacles, and new applications for CVDs.

Artificial intelligence in cardiovascular diseases: diagnostic and therapeutic perspectives

Artificial intelligence (AI), the technique of extracting information from complex database using sophisticated computer algorithms, has incorporated itself in medical field. AI techniques have shown the potential to accelerate the progression of diagnosis and treatment of cardiovascular diseases (CVDs), including heart failure, atrial fibrillation, valvular heart disease, hypertrophic cardiomyopathy, congenital heart disease and so on. In clinical scenario, AI have been proved to apply well in CVD diagnosis, enhance effectiveness of auxiliary tools, disease stratification and typing, and outcome prediction. Deeply developed to capture subtle connections from massive amounts of healthcare data, recent AI algorithms are expected to handle even more complex tasks than traditional methods. The aim of this review is to introduce current applications of AI in CVDs, which may allow clinicians who have limited expertise of computer science to better understand the frontier of the subject and put AI algorithms into clinical practice.

Can artificial intelligence and robotic nurses replace operating room nurses? The quasi-experimental research

Operating room nurses, who have an important place in the professional healthcare team, should be able to use technology effectively and adapt to innovations. This study is expected to shed light on how effective the development of robotic technologies and artificial intelligence and their integration into operating room nursing will be in fulfilling the requirements of contemporary nursing philosophy. This study was designed as a single group pre- and post-test quasi-experimental study. The quasi-experimental (pretest-posttest) research design was utilized to conduct the study in a Training and Research Hospital in Western Turkey. The nurses (n = 35) working in the operating room of the aforementioned hospital were included in the study. In this study, we aimed to determine whether operating room nurses experienced anxiety due to the use of artificial intelligence and robotic nurses, and the effectiveness of the training given to them in order to raise their awareness. The following three tools were used for data collection: The Nurses' Descriptive Characteristics Form, Artificial Intelligence Knowledge Questionnaire, and Artificial Intelligence Anxiety Scale. Data extraction and analysis were performed in a narrative and tabular way. According to this study, the training given to the operating room nurses significantly increased their knowledge levels about artificial intelligence and robotic nurses, and increased their artificial intelligence- and robotic nurse-related anxiety significantly (p < 0.05). The participating operating room nurses experienced limitations regarding current information, training programs and learning opportunities on robotic surgery. We recommend that the operating room nurses should be provided with trainings on artificial intelligence technologies and robotic nurses, and that they should be enabled to use these information technologies regarding future technologies actively.

Deep learning applications in coronary anatomy imaging: a systematic review and meta-analysis

Background

The application of deep learning on medical imaging is growing in prevalence in the recent literature. One of the most studied areas is coronary artery disease (CAD). Imaging of coronary artery anatomy is fundamental, which has led to a high number of publications describing a variety of techniques. The aim of this systematic review is to review the evidence behind the accuracy of deep learning applications in coronary anatomy imaging.

Methods

The search for the relevant studies, which applied deep learning on coronary anatomy imaging, was performed in a systematic approach on MEDLINE and EMBASE databases, followed by reviewing of abstracts and full texts. The data from the final studies was retrieved using data extraction forms. A meta-analysis was performed on a subgroup of studies, which looked at fractional flow reserve (FFR) prediction. Heterogeneity was tested using tau², I² and Q tests. Finally, a risk of bias was performed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) approach.

Results

A total of 81 studies met the inclusion criteria. The most common imaging modality was coronary computed tomography angiography (CCTA) (58%) and the most common deep learning method was convolutional neural network (CNN) (52%). The majority of studies demonstrated good performance metrics. The most common outputs were focused on coronary artery segmentation, clinical outcome prediction, coronary calcium quantification and FFR prediction, and most studies reported area under the curve (AUC) of ≥80%. The pooled diagnostic odds ratio (DOR) derived from 8 studies looking at FFR prediction using CCTA was 12.5 using the Mantel-Haenszel (MH) method. There was no significant heterogeneity amongst studies according to Q test (P=0.2496).

Conclusions

Deep learning has been used in many applications on coronary anatomy imaging, most of which are yet to be externally validated and prepared for clinical use. The performance of deep learning, especially CNN models, proved to be powerful and some applications have already translated into medical practice, such as computed tomography (CT)-FFR. These applications have the potential to translate technology into better care of CAD patients.

Metabolomics and random forests in patients with complex congenital heart disease

Introduction

It is increasingly common to simultaneously determine a large number of metabolites in order to assess the metabolic state of, or clarify biochemical pathways in, an organism (“metabolomics”). This approach is increasingly used in the investigation of the development of heart failure. Recently, the first reports with respect to a metabolomic approach for the assessment of patients with complex congenital heart disease have been published. Classical statistical analysis of such data is challenging.

Objective

This study aims to present an alternative to classical statistics with respect to identifying relevant metabolites in a classification task and numerically estimating their relative impact.

Methods

Data from two metabolomic studies on 20 patients with complex congenital heart disease and Fontan circulation and 20 controls were reanalysed using random forest (RF) methodology. Results were compared to those of classical statistics.

Results

RF analysis required no elaborate data pre-processing. The ranking of the variables with respect to classification impact (subject diseased, or not) was remarkably similar irrespective of the evaluation method used, leading to identical clinical interpretation.

Conclusion

In metabolomic classification in adult patients with complex congenital heart disease, RF analysis as a one-step method delivers the most adequate results with minimum effort. RF may serve as an adjunct to traditional statistics also in this small but crucial-to-monitor patient group.

Machine Learning Can Predict the Probability of Biologic Therapy in Patients with Inflammatory Bowel Disease

Background: Inflammatory bowel disease (IBD) is of high medical and socioeconomic relevance. Moderate and severe disease courses often require treatment with biologics. The aim of this study was to evaluate machine learning (ML)-based methods for the prediction of biologic therapy in IBD patients using a large prescription database. Methods: The present retrospective cohort study utilized a longitudinal prescription database (LRx). Patients with at least one prescription for an intestinal anti-inflammatory agent from a gastroenterologist between January 2015 and July 2021 were included. Patients who had received an initial biologic therapy prescription (infliximab, adalimumab, golimumab, vedolizumab, or ustekinumab) were categorized as the “biologic group”. The potential predictors included in the machine learning-based models were age, sex, and the 100 most frequently prescribed drugs within 12 months prior to the index date. Six machine learning-based methods were used for the prediction of biologic therapy. Results: A total of 122,089 patients were included in this study. Of these, 15,824 (13.0%) received at least one prescription for a biologic drug. The Light Gradient Boosting Machine had the best performance (accuracy = 74%) and was able to correctly identify 78.5% of the biologics patients and 72.6% of the non-biologics patients in the testing dataset. The most important variable was prednisolone, followed by lower age, mesalazine, budesonide, and ferric iron. Conclusions: In summary, this study reveals the advantages of ML-based models in predicting biologic therapy in IBD patients based on pre-treatment and demographic variables. There is a need for further studies in this regard that take into account individual patient characteristics, i.e., genetics and gut microbiota, to adequately address the challenges of finding optimal treatment strategies for patients with IBD.

A Cardiovascular Disease Prediction Model Based on Routine Physical Examination Indicators Using Machine Learning Methods: A Cohort Study

Background

Cardiovascular diseases (CVD) are currently the leading cause of premature death worldwide. Model-based early detection of high-risk populations for CVD is the key to CVD prevention. Thus, this research aimed to use machine learning (ML) algorithms to establish a CVD prediction model based on routine physical examination indicators suitable for the Xinjiang rural population.

Method

The research cohort data collection was divided into two stages. The first stage involved a baseline survey from 2010 to 2012, with follow-up ending in December 2017. The second-phase baseline survey was conducted from September to December 2016, and follow-up ended in August 2021. A total of 12,692 participants (10,407 Uyghur and 2,285 Kazak) were included in the study. Screening predictors and establishing variable subsets were based on least absolute shrinkage and selection operator (Lasso) regression, logistic regression forward partial likelihood estimation (FLR), random forest (RF) feature importance, and RF variable importance. The selected subset of variables was compared with L1 regularized logistic regression (L1-LR), RF, support vector machine (SVM), and AdaBoost algorithm to establish a CVD prediction model suitable for this population. The incidence of CVD in this population was then analyzed.

Result

After 4.94 years of follow-up, a total of 1,176 people were diagnosed with CVD (cumulative incidence: 9.27%). In the comparison of discrimination and calibration, the prediction performance of the subset of variables selected based on FLR was better than that of other models. Combining the results of discrimination, calibration, and clinical validity, the prediction model based on L1-LR had the best prediction performance. Age, systolic blood pressure, low-density lipoprotein-L/high-density lipoproteins-C, triglyceride blood glucose index, body mass index, and body adiposity index were all important predictors of the onset of CVD in the Xinjiang rural population.

Conclusion

In the Xinjiang rural population, the prediction model based on L1-LR had the best prediction performance.

Non-Invasive Quantification of Fraction Flow Reserve Based on Steady-State Geometric Multiscale Models

Background: The underuse of invasive fraction flow reserve (FFR) in clinical practice has motivated research towards its non-invasive prediction. The early attempts relied on solving the incompressible three-dimensional Navier–Stokes equations in segmented coronary arteries. However, transient boundary condition has a high resource intensity in terms of computational time. Herein, a method for calculating FFR based on steady-state geometric multiscale (FFR_SS) is proposed.

Methods: A total of 154 moderately stenotic vessels (40–80% diameter stenosis) from 136 patients with stable angina were included in this study to validate the clinical diagnostic performance of FFR_SS. The method was based on the coronary artery model segmented from the patient’s coronary CTA image. The average pressure was used as the boundary condition for the inlet, and the microcirculation resistance calculated by the coronary flow was used as the boundary condition for the outlet to calculate the patient-specific coronary hyperemia. Then, the flow velocity and pressure distribution and the FFRss of each coronary artery branch were calculated to evaluate the degree of myocardial ischemia caused by coronary stenosis. Also, the FFR_SS and FFR_CT of all patients were calculated, and the clinically measured FFR was used as the “gold standard” to verify the diagnostic performance of FFR_SS and to compare the correlation between FFR_SS and FFR_CT.

Results: According to the FFR_SS calculation results of all patients, FFR_SS and FFR have a good correlation (r = 0.68, p < 0.001). Similarly, the correlation of FFR_SS and FFR_CT demonstrated an r of 0.75 (95%CI: 0.67–0.72) (p < 0.001). On receiver-operating characteristic analysis, the optimal FFR_SS cut point for FFR≤0.80 was 0.80 (AUC:0.85 [95% confidence interval: 0.79 to 0.90]; overall accuracy:88.3%). The overall sensitivity, specificity, PPV, and NPV for FFR_SS ≤0.80 versus FFR ≤0.80 was 68.18% (95% CI: 52.4–81.4), 93.64% (95% CI: 87.3–97.4), 82.9%, and 91.1%, respectively.

Conclusion: FFR_SS is a reliable diagnostic index for myocardial ischemia. This method was similar to the closed-loop geometric multiscale calculation of FFR accuracy but improved the calculation efficiency. It also improved the clinical applicability of the non-invasive computational FFR model, helped the clinicians diagnose myocardial ischemia, and guided percutaneous coronary intervention.

Development and validation pathways of artificial intelligence tools evaluated in randomised clinical trials

Objective

Given the complexities of testing the translational capability of new artificial intelligence (AI) tools, we aimed to map the pathways of training/validation/testing in development process and external validation of AI tools evaluated in dedicated randomised controlled trials (AI-RCTs).

Methods

We searched for peer-reviewed protocols and completed AI-RCTs evaluating the clinical effectiveness of AI tools and identified development and validation studies of AI tools. We collected detailed information, and evaluated patterns of development and external validation of AI tools.

Results

We found 23 AI-RCTs evaluating the clinical impact of 18 unique AI tools (2009–2021). Standard-of-care interventions were used in the control arms in all but one AI-RCT. Investigators did not provide access to the software code of the AI tool in any of the studies. Considering the primary outcome, the results were in favour of the AI intervention in 82% of the completed AI-RCTs (14 out of 17). We identified significant variation in the patterns of development, external validation and clinical evaluation approaches among different AI tools. A published development study was found only for 10 of the 18 AI tools. Median time from the publication of a development study to the respective AI-RCT was 1.4 years (IQR 0.2–2.2).

Conclusions

We found significant variation in the patterns of development and validation for AI tools before their evaluation in dedicated AI-RCTs. Published peer-reviewed protocols and completed AI-RCTs were also heterogeneous in design and reporting. Upcoming guidelines providing guidance for the development and clinical translation process aim to improve these aspects.

Application of machine learning to predict the occurrence of arrhythmia after acute myocardial infarction

Background

Early identification of the occurrence of arrhythmia in patients with acute myocardial infarction plays an essential role in clinical decision-making. The present study attempted to use machine learning (ML) methods to build predictive models of arrhythmia after acute myocardial infarction (AMI).

Methods

A total of 2084 patients with acute myocardial infarction were enrolled in this study. (All data is available on Github: https://github.com/wangsuhuai/AMI-database1.git). The primary outcome is whether tachyarrhythmia occurred during admission containing atrial arrhythmia, ventricular arrhythmia, and supraventricular tachycardia. All data is randomly divided into a training set (80%) and an internal testing set (20%). Apply three machine learning algorithms: decision tree, random forest (RF), and artificial neural network (ANN) to learn the training set to build a model, then use the testing set to evaluate the prediction performance, and compare it with the model built by the Global Registry of Acute Coronary Events (GRACE) risk variable set.

Results

Three ML models predict the occurrence of tachyarrhythmias after AMI. After variable selection, the artificial neural network (ANN) model has reached the highest accuracy rate, which is better than the model constructed using the Grace variable set. After applying SHapley Additive exPlanations (SHAP) to make the model interpretable, the most important features are abnormal wall motion, lesion location, bundle branch block, age, and heart rate. Among them, RBBB (odds ratio [OR]: 4.21; 95% confidence interval [CI]: 2.42–7.02), ≥ 2 ventricular walls motion abnormal (OR: 3.26; 95% CI: 2.01–4.36) and right coronary artery occlusion (OR: 3.00; 95% CI: 1.98–4.56) are significant factors related to arrhythmia after AMI.

Conclusions

We used advanced machine learning methods to build prediction models for tachyarrhythmia after AMI for the first time (especially the ANN model that has the best performance). The current study can supplement the current AMI risk score, provide a reliable evaluation method for the clinic, and broaden the new horizons of ML and clinical research.

Trial registration Clinical Trial Registry No.: ChiCTR2100041960.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-021-01667-8.

Social Determinants in Machine Learning Cardiovascular Disease Prediction Models: A Systematic Review

INTRODUCTION

Cardiovascular disease is the leading cause of death worldwide, and cardiovascular disease burden is increasing in low-resource settings and for lower socioeconomic groups. Machine learning algorithms are being developed rapidly and incorporated into clinical practice for cardiovascular disease prediction and treatment decisions. Significant opportunities for reducing death and disability from cardiovascular disease worldwide lie with accounting for the social determinants of cardiovascular outcomes. This study reviews how social determinants of health are being included in machine learning algorithms to inform best practices for the development of algorithms that account for social determinants.

METHODS

A systematic review using 5 databases was conducted in 2020. English language articles from any location published from inception to April 10, 2020, which reported on the use of machine learning for cardiovascular disease prediction that incorporated social determinants of health, were included.

RESULTS

Most studies that compared machine learning algorithms and regression showed increased performance of machine learning, and most studies that compared performance with or without social determinants of health showed increased performance with them. The most frequently included social determinants of health variables were gender, race/ethnicity, marital status, occupation, and income. Studies were largely from North America, Europe, and China, limiting the diversity of the included populations and variance in social determinants of health.

DISCUSSION

Given their flexibility, machine learning approaches may provide an opportunity to incorporate the complex nature of social determinants of health. The limited variety of sources and data in the reviewed studies emphasize that there is an opportunity to include more social determinants of health variables, especially environmental ones, that are known to impact cardiovascular disease risk and that recording such data in electronic databases will enable their use.

Cardiovascular Disease Prediction by Machine Learning Algorithms Based on Cytokines in Kazakhs of China

Background

Cardiovascular disease (CVD) is the leading cause of mortality worldwide. Accurately identifying subjects at high-risk of CVD may improve CVD outcomes. We sought to systematically examine the feasibility and performance of 7 widely used machine learning (ML) algorithms in predicting CVD risks.

Methods

The final analysis included 1508 Kazakh subjects in China without CVD at baseline who completed follow-up. All subjects were randomly divided into the training set (80%) and the test set (20%). L1-penalized logistic regression (LR), support vector machine with radial basis function (SVM), decision tree (DT), random forest (RF), k-nearest neighbors (KNN), Gaussian naive Bayes (NB), and extreme gradient boosting (XGB) were employed for prediction CVD outcomes. Ten-fold cross-validation was used during model developing and hyperparameters tuning in the training set. Model performance was evaluated in the test set in light of discrimination, calibration, and clinical usefulness. RF was applied to obtain the variable importance of included variables. Twenty-two variables, including sociodemographic characteristics, medical history, cytokines, and synthetic indices, were used for model development.

Results

Among 1508 subjects, 203 were diagnosed with CVD over a median follow-up of 5.17 years. All 7 models had moderate to excellent discrimination (AUC ranged from 0.770 to 0.872) and were well calibrated. LR and SVM performed identically with an AUC of 0.872 (95% CI: 0.829–0.907) and 0.868 (95% CI: 0.825–0.904), respectively. LR had the lowest Brier score (0.078) and the highest sensitivity (97.1%). Decision curve analysis indicated that SVM was slightly better than LR. The inflammatory cytokines, such as hs-CRP and IL-6, were identified as strong predictors of CVD.

Conclusion

SVM and LR can be applied to guide clinical decision-making in the Kazakh Chinese population, and further study is required to ensure their accuracies.

Clinical prediction models of fractional flow reserve: an exploration of the current evidence and appraisal of model performance

Background

Invasive fractional flow reserve (FFR) is a standard indicator of coronary stenoses' hemodynamic severity. Clinical prediction models (CPMs) may help differentiate ischemic from non-ischemic lesions without using a pressure wire but by integrating related variables. This approach differs from that of physics-based models. However, it is not yet known which CPMs are the most reliable at detecting hemodynamic significance.

Methods

A systematic review was performed of relevant publications that developed or validated any FFR CPMs from inception to April 2019 in the PubMed, EMBASE, and Cochrane Library databases by two independent authors. The risk of bias and applicability were assessed using the prediction model risk of the bias assessment tool (PROBAST).

Results

A total of 11 unique CPMs and 5 subsequent external validation studies were identified. The prevalence of hemodynamically significant lesions (FFR ≤0.80) across the studies had a median of 37.1% (range: 20.7-68.0%). Lesion length, percent diameter stenosis, and minimal lumen diameter were the three most frequently used variables in the CPMs. Of the 11 FFR CPMs, 9 (82%) exhibited strong discrimination [area under the curve (AUC) >0.75], and 5 (45%) had been subject to external validation; however, calibration was only available for 3 models (27%). There was a high degree of applicability; however, none of the studies was assessed as having a low risk of bias. A CPM was identified that had undergone rigorous validation and calibration: the DILEMMA score (three validations; median AUC, 0.83).

Conclusions

Almost half of the existing FFR CPMs had been externally validated. Due to their good discrimination abilities, these FFR CPMs are useful tools that could reduce the need for invasive hemodynamic measurements. Future research that adheres to methodological guidelines should be undertaken to develop high-quality models in this setting. (PROSPERO registration number: CRD42019125011).

Machine Learning-Based Prediction of Small Intracranial Aneurysm Rupture Status Using CTA-Derived Hemodynamics: A Multicenter Study

BACKGROUND AND PURPOSE

Small intracranial aneurysms are being increasingly detected while the rupture risk is not well-understood. We aimed to develop rupture-risk models of small aneurysms by combining clinical, morphologic, and hemodynamic information based on machine learning techniques and to test the models in external validation datasets.

MATERIALS AND METHODS

From January 2010 to December 2016, five hundred four consecutive patients with only small aneurysms (<5 mm) detected by CTA and invasive cerebral angiography (or surgery) were retrospectively enrolled and randomly split into training (81%) and internal validation (19%) sets to derive and validate the proposed machine learning models (support vector machine, random forest, logistic regression, and multilayer perceptron). Hemodynamic parameters were obtained using computational fluid dynamics simulation. External validation was performed in other hospitals to test the models.

RESULTS

The support vector machine performed the best with areas under the curve of 0.88 (95% CI, 0.85-0.92) and 0.91 (95% CI, 0.74-0.98) in the training and internal validation datasets, respectively. Feature ranks suggested hemodynamic parameters, including stable flow pattern, concentrated inflow streams, and a small (<50%) flow-impingement zone, and the oscillatory shear index coefficient of variation, were the best predictors of aneurysm rupture. The support vector machine showed an area under the curve of 0.82 (95% CI, 0.69-0.94) in the external validation dataset, and no significant difference was found for the areas under the curve between internal and external validation datasets (P = .21).

CONCLUSIONS

This study revealed that machine learning had a good performance in predicting the rupture status of small aneurysms in both internal and external datasets. Aneurysm hemodynamic parameters were regarded as the most important predictors.

Machine Learning Quantitation of Cardiovascular and Cerebrovascular Disease: A Systematic Review of Clinical Applications

Research into machine learning (ML) for clinical vascular analysis, such as those useful for stroke and coronary artery disease, varies greatly between imaging modalities and vascular regions. Limited accessibility to large diverse patient imaging datasets, as well as a lack of transparency in specific methods, are obstacles to further development. This paper reviews the current status of quantitative vascular ML, identifying advantages and disadvantages common to all imaging modalities. Literature from the past 8 years was systematically collected from MEDLINE^® and Scopus database searches in January 2021. Papers satisfying all search criteria, including a minimum of 50 patients, were further analysed and extracted of relevant data, for a total of 47 publications. Current ML image segmentation, disease risk prediction, and pathology quantitation methods have shown sensitivities and specificities over 70%, compared to expert manual analysis or invasive quantitation. Despite this, inconsistencies in methodology and the reporting of results have prevented inter-model comparison, impeding the identification of approaches with the greatest potential. The clinical potential of this technology has been well demonstrated in Computed Tomography of coronary artery disease, but remains practically limited in other modalities and body regions, particularly due to a lack of routine invasive reference measurements and patient datasets.

Predicting the Risk of Adverse Events in Pregnant Women With Congenital Heart Disease

Background

Women with congenital heart disease are considered at high risk for adverse events. Therefore, we aim to establish 2 prediction models for mothers and their offspring, which can predict the risk of adverse events occurred in pregnant women with congenital heart disease.

Methods and Results

A total of 318 pregnant women with congenital heart disease were included; 213 women were divided into the development cohort, and 105 women were divided into the validation cohort. Least absolute shrinkage and selection operator was used for predictor selection. After validation, multivariate logistic regression analysis was used to develop the model. Machine learning algorithms (support vector machine, random forest, AdaBoost, decision tree, k‐nearest neighbor, naïve Bayes, and multilayer perceptron) were used to further verify the predictive ability of the model. Forty‐one (12.9%) women experienced adverse maternal events, and 93 (29.2%) neonates experienced adverse neonatal events. Seven high‐risk factors were discovered in the maternal model, including New York Heart Association class, Eisenmenger syndrome, pulmonary hypertension, left ventricular ejection fraction, sinus tachycardia, arterial blood oxygen saturation, and pregnancy duration. The machine learning–based algorithms showed that the maternal model had an accuracy of 0.76 to 0.86 (area under the receiver operating characteristic curve=0.74–0.87) in the development cohort, and 0.72 to 0.86 (area under the receiver operating characteristic curve=0.68–0.80) in the validation cohort. Three high‐risk factors were discovered in the neonatal model, including Eisenmenger syndrome, preeclampsia, and arterial blood oxygen saturation. The machine learning–based algorithms showed that the neonatal model had an accuracy of 0.75 to 0.80 (area under the receiver operating characteristic curve=0.71–0.77) in the development cohort, and 0.72 to 0.79 (area under the receiver operating characteristic curve=0.69–0.76) in the validation cohort.

Conclusions

Two prenatal risk assessment models for both adverse maternal and neonatal events were established, which might assist clinicians in tailoring precise management and therapy in pregnant women with congenital heart disease.

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.

Machine Learning for Assessment of Coronary Artery Disease in Cardiac CT: A Survey

Cardiac computed tomography (CT) allows rapid visualization of the heart and coronary arteries with high spatial resolution. However, analysis of cardiac CT scans for manifestation of coronary artery disease is time-consuming and challenging. Machine learning (ML) approaches have the potential to address these challenges with high accuracy and consistent performance. In this mini review, we present a survey of the literature on ML-based analysis of coronary artery disease in cardiac CT. We summarize ML methods for detection and characterization of atherosclerotic plaque as well as anatomically and functionally significant coronary artery stenosis.

Prediction of pulmonary pressure after Glenn shunts by computed tomography-based machine learning models

OBJECTIVES

This study aimed to develop non-invasive machine learning classifiers for predicting post-Glenn shunt patients with low and high risks of a mean pulmonary arterial pressure (mPAP) > 15 mmHg based on preoperative cardiac computed tomography (CT).

METHODS

This retrospective study included 96 patients with functional single ventricle who underwent a bidirectional Glenn procedure between November 1, 2009, and July, 31, 2017. All patients underwent post-procedure CT, followed by cardiac catheterization. Overall, 23 morphologic parameters were manually extracted from cardiac CT images for each patient. The Mann-Whitney U or chi-square test was applied to select the most significant predictors. Six machine learning algorithms including logistic regression, Naive Bayes, random forest (RF), linear discriminant analysis, support vector machine, and K-nearest neighbor were used for modeling. These algorithms were independently trained on 100 train-validation random splits with a 3:1 ratio. Their average performance was evaluated by area under the curve (AUC), accuracy, sensitivity, and specificity.

RESULTS

Seven CT morphologic parameters were selected for modeling. RF obtained the best performance, with mean AUC of 0.840 (confidence interval [CI] 0.832-0.850) and 0.787 (95% CI 0.780-0.794); sensitivity of 0.815 (95% CI 0.797-0.833) and 0.778 (95% CI 0.767-0.788), specificity of 0.766 (95% CI 0.748-0.785) and 0.746 (95% CI 0.735-0.757); and accuracy of 0.782 (95% CI 0.771-0.793) and 0.756 (95% CI 0.748-0.764) in the training and validation cohorts, respectively.

CONCLUSIONS

The CT-based RF model demonstrates a good performance in the prediction of mPAP, which may reduce the need for right heart catheterization in post-Glenn shunt patients with suspected mPAP > 15 mmHg.

KEY POINTS

• Twenty-three candidate descriptors were manually extracted from cardiac computed tomography images, and seven of them were selected for subsequent modeling. • The random forest model presents the best predictive performance for pulmonary pressure among all methods. • The computed tomography-based machine learning model could predict post-Glenn shunt pulmonary pressure non-invasively.

The Current Research Landscape of the Application of Artificial Intelligence in Managing Cerebrovascular and Heart Diseases: A Bibliometric and Content Analysis

The applications of artificial intelligence (AI) in aiding clinical decision-making and management of stroke and heart diseases have become increasingly common in recent years, thanks in part to technological advancements and the heightened interest of the research and medical community. This study aims to provide a comprehensive picture of global trends and developments of AI applications relating to stroke and heart diseases, identifying research gaps and suggesting future directions for research and policy-making. A novel analysis approach that combined bibliometrics analysis with a more complex analysis of abstract content using exploratory factor analysis and Latent Dirichlet allocation, which uncovered emerging research domains and topics, was adopted. Data were extracted from the Web of Science database. Results showed topics with the most compelling growth to be AI for big data analysis, robotic prosthesis, robotics-assisted stroke rehabilitation, and minimally invasive surgery. The study also found an emerging landscape of research that was centered on population-specific and early detection of stroke and heart disease. Application of AI in health behavior tracking and improvement as well as the use of robotics in medical diagnostics and prognostication have also been found to attract significant research attention. In light of these findings, it is suggested that the currently under-researched issues of data management, AI model reliability, as well as validation of its clinical utility, need to be further explored in future research and policy decisions to maximize the benefits of AI applications in stroke and heart diseases.