Machine learning-based phenogrouping in heart failure to identify responders to cardiac resynchronization therapy

Establishing a differential diagnosis model between primary membranous nephropathy and non-primary membranous nephropathy by machine learning algorithms

CONTEXT

Four algorithms with relatively balanced complexity and accuracy in deep learning classification algorithm were selected for differential diagnosis of primary membranous nephropathy (PMN).

OBJECTIVE

This study explored the most suitable classification algorithm for PMN identification, and to provide data reference for PMN diagnosis research.

METHODS

A total of 500 patients were referred to Luo-he Central Hospital from 2019 to 2021. All patients were diagnosed with primary glomerular disease confirmed by renal biopsy, contained 322 cases of PMN, the 178 cases of non-PMN. Using the decision tree, random forest, support vector machine, and extreme gradient boosting (Xgboost) to establish a differential diagnosis model for PMN and non-PMN. Based on the true positive rate, true negative rate, false-positive rate, false-negative rate, accuracy, feature work area under the curve (AUC) of subjects, the best performance of the model was chosen.

RESULTS

The efficiency of the Xgboost model based on the above evaluation indicators was the highest, which the diagnosis of PMN of the sensitivity and specificity, respectively 92% and 96%.

CONCLUSIONS

The differential diagnosis model for PMN was established successfully and the efficiency performance of the Xgboost model was the best. It could be used for the clinical diagnosis of PMN.

Predicting angiographic coronary artery disease using machine learning and high-frequency QRS

AIM

Exercise stress ECG is a common diagnostic test for stable coronary artery disease, but its sensitivity and specificity need to be further improved. In this paper, we construct a machine learning model for the prediction of angiographic coronary artery disease by HFQRS analysis of cycling exercise ECG.

METHODS AND RESULTS

This study prospectively included 140 inpatients and 59 healthy volunteers undergoing cycling exercise ECG. The CHD group (N=104) and non-CHD group (N=95) were determined by coronary angiography gold standard. Automated HF QRS analysis was performed by the blinded method. The coronary group was predominantly male, with a higher prevalence of age, BMI, hypertension, and diabetes than the non-coronary group ( P < 0.001 ), higher lipid levels in the coronary group ( P < 0.005 ), significantly longer QRS duration during exercise testing ( P < 0.005 ), more positive leads ( P < 0.001 ), and a greater proportion of significant changes in HFQRS ( P < 0.001 ). Age, Gender, Hypertension, Diabetes, and HF QRS Conclusions were screened by correlation analysis and multifactorial retrospective analysis to construct the machine learning models of the XGBoost Classifier, Logistic Regression, LightGBM Classifier, RandomForest Classifier, Artificial Neural Network and Support Vector Machine, respectively.

CONCLUSION

Male, elderly, with hypertension, diabetes mellitus, and positive exercise stress test HFQRS conclusions suggested a high risk of CHD. The best performance of the Logistic Regression model was compared, and a column line graph for assessing the risk of CHD was further developed and validated.

Deciphering Popliteal Artery Aneurysm Patient Diversity: Insights From a Cluster Analysis of the POPART Registry

BACKGROUND

Popliteal artery aneurysms (PAAs) are the most common peripheral aneurysm. However, due to its rarity, the cumulative body of evidence regarding patient patterns, treatment strategies, and perioperative outcomes is limited. This analysis aims to investigate distinct phenotypical patient profiles and associated treatment and outcomes in patients with a PAA by performing an unsupervised clustering analysis of the POPART (Practice of Popliteal Artery Aneurysm Repair and Therapy) registry.

METHODS AND RESULTS

A cluster analysis (using k-means clustering) was performed on data obtained from the multicenter POPART registry (42 centers from Germany and Luxembourg). Sensitivity analyses were conducted to explore validity and stability. Using 2 clusters, patients were primarily separated by the absence or presence of clinical symptoms. Within the cluster of symptomatic patients, the main difference between patients with acute limb ischemia presentation and nonemergency symptomatic patients was PAA diameter. When using 6 clusters, patients were primarily grouped by comorbidities, with patients with acute limb ischemia forming a separate cluster. Despite markedly different risk profiles, perioperative complication rates appeared to be positively associated with the proportion of emergency patients. However, clusters with a higher proportion of patients having any symptoms before treatment experienced a lower rate of perioperative complications.

CONCLUSIONS

The conducted analyses revealed both an insight to the public health reality of PAA care as well as patients with PAA at elevated risk for adverse outcomes. This analysis suggests that the preoperative clinic is a far more crucial adjunct to the patient's preoperative risk assessment than the patient's epidemiological profile by itself.

Screening of lung cancer serum biomarkers based on Boruta-shap and RFC-RFECV algorithms

OBJECTIVE

This study aimed to identify a set of serum miRNAs as potential biomarkers for lung cancer diagnosis using algorithmic approaches.

METHODS

Serum miRNA expression data from lung cancer patients and non-tumor controls were obtained. The top six miRNAs were selected using Boruta-shap and RFC-RFECV algorithms. A Gaussian Naive Bayes (NB) classifier was trained and evaluated using cross-validation, ROC curve analysis, and evaluation metrics.

RESULTS

Six miRNAs (hsa-miRNA-144, hsa-miRNA-107, hsa-miRNA-484, hsa-miRNA-103, hsa-miRNA-26b, and hsa-miRNA-641) were identified as feature genes. The NB classifier achieved an area under curve (AUC) of 0.8966 and a mean AUC of 0.88 in cross-validation. Accuracy, recall, and F1 scores exhibited promising results, with an accuracy of 82%. In the validation set, the AUC values for the NB and SVC classifiers were 0.9345 and 0.9423, respectively, with a mean AUC of 0.95 in cross-validation. The classifiers demonstrated an accuracy of 89% in diagnosing lung cancer.

CONCLUSION

This study identified a panel of six serum miRNAs with potential as non-invasive biomarkers for lung cancer diagnosis. These miRNAs show promise in providing sensitive and specific tools for detecting lung cancer.

SIGNIFICANCE

Lung cancer is one of the top cancers worldwide, threatening the health and lives of tens of thousands of people. miRNA is a biomarker, which can be used as a potential clinical tool for diagnosis and prognosis of cancer patients. Therefore, the use of multiple miRNAs to construct diagnostic models may be one of the future methods of accurate diagnosis of lung cancer. In this study, we used the Boruta-shap and RFC-RFECV algorithms to automatically identify and extract characteristic miRNAs highly associated with lung cancer, thereby establishing an accurate classifier for the diagnosis of lung cancer with characteristic miRNAs.

Machine Learning and Deep Learning Methods for Fast and Accurate Assessment of Transthoracic Echocardiogram Image Quality

High-quality echocardiogram images are the cornerstone of accurate and reliable measurements of the heart. Therefore, this study aimed to develop, validate and compare machine learning and deep learning algorithms for accurate and automated assessment of transthoracic echocardiogram image quality. In total, 4090 single-frame two-dimensional transthoracic echocardiogram images were used from apical 4-chamber, apical 2-chamber and parasternal long-axis views sampled from 3530 adult patients. The data were extracted from CAMUS and Unity Imaging open-source datasets. For every raw image, additional grayscale block histograms were developed. For block histogram datasets, six classic machine learning algorithms were tested. Moreover, convolutional neural networks based on the pre-trained EfficientNetB4 architecture were developed for raw image datasets. Classic machine learning algorithms predicted image quality with 0.74 to 0.92 accuracy (AUC 0.81 to 0.96), whereas convolutional neural networks achieved between 0.74 and 0.89 prediction accuracy (AUC 0.79 to 0.95). Both approaches are accurate methods of echocardiogram image quality assessment. Moreover, this study is a proof of concept of a novel method of training classic machine learning algorithms on block histograms calculated from raw images. Automated echocardiogram image quality assessment methods may provide additional relevant information to the echocardiographer in daily clinical practice and improve reliability in clinical decision making.

Impact Exploration of Spatiotemporal Feature Derivation and Selection on Machine Learning-Based Predictive Models for Post-Embolization Cerebral Aneurysm Recanalization

PURPOSE

To enhance the performance of machine learning (ML) models for the post-embolization recanalization of cerebral aneurysms, we evaluated the impact of hemodynamic feature derivation and selection method on six ML algorithms.

METHODS

We utilized computational fluid dynamics (CFD) to simulate hemodynamics in 66 cerebral aneurysms from 65 patients, including 57 stable and nine recanalized aneurysms. We derived a total of 107 features for each aneurysm, encompassing four clinical features, 12 morphological features, and 91 hemodynamic features. To investigate the influence of feature derivation and selection methods on the ML models, we employed two derivation methods, simplified and fully derived, in combination with four selection methods: all features, statistically significant analysis, stepwise multivariate logistic regression analysis (stepwise-LR), and recursive feature elimination (RFE). Model performance was assessed using the area under the receiver operating characteristic curve (AUROC) and precision-recall curve (AUPRC) on both the training and testing datasets.

RESULTS

The AUROC values on the testing dataset exhibited a wide-ranging spectrum, spanning from 0.373 to 0.863. Fully derived features and the RFE selection method demonstrated superior performance in intra-model comparisons. The multi-layer perceptron (MLP) model, trained with RFE-selected fully derived features, achieved the best performance on the testing dataset, with an AUROC value of 0.863 (95% CI: 0.684- 1.000).

CONCLUSION

Our study demonstrated the importance of feature derivation and selection in determining the performance of ML models. This enabled the development of accurate decision-making models without the need to invade the patient.

Role of Machine Learning and Artificial Intelligence in Arrhythmias and Electrophysiology

Machine learning (ML), a subset of artificial intelligence (AI) centered on machines learning from extensive datasets, stands at the forefront of a technological revolution shaping various facets of society. Cardiovascular medicine has emerged as a key domain for ML applications, with considerable efforts to integrate these innovations into routine clinical practice. Within cardiac electrophysiology, ML applications, especially in the automated interpretation of electrocardiograms, have garnered substantial attention in existing literature. However, less recognized are the diverse applications of ML in cardiac electrophysiology and arrhythmias, spanning basic science research on arrhythmia mechanisms, both experimental and computational, as well as contributions to enhanced techniques for mapping cardiac electrical function and translational research related to arrhythmia management. This comprehensive review delves into various ML applications within the scope of this journal, organized into 3 parts. The first section provides a fundamental understanding of general ML principles and methodologies, serving as a foundational resource for readers interested in exploring ML applications in arrhythmia research. The second part offers an in-depth review of studies in arrhythmia and electrophysiology that leverage ML methodologies, showcasing the broad potential of ML approaches. Each subject is thoroughly outlined, accompanied by a review of notable ML research advancements. Finally, the review delves into the primary challenges and future perspectives surrounding ML-driven cardiac electrophysiology and arrhythmias research.

Data-driven classification of left atrial morphology and its predictive impact on atrial fibrillation catheter ablation

INTRODUCTION

Various left atrial (LA) anatomical structures are correlated with postablative recurrence for atrial fibrillation (AF) patients. Comprehensively integrating anatomical structures, digitizing them, and implementing in-depth analysis, which may supply new insights, are needed. Thus, we aim to establish an interpretable model to identify AF patients' phenotypes according to LA anatomical morphology, using machine learning techniques.

METHODS AND RESULTS

Five hundred and nine AF patients underwent first ablation treatment in three centers were included and were followed-up for postablative recurrent atrial arrhythmias. Data from 369 patients were regarded as training set, while data from another 140 patients, collected from different centers, were used as validation set. We manually measured 57 morphological parameters on enhanced computed tomography with three-dimensional reconstruction technique and implemented unsupervised learning accordingly. Three morphological groups were identified, with distinct prognosis according to Kaplan-Meier estimator (p < .001). Multivariable Cox model revealed that morphological grouping were independent predictors of 1-year recurrence (Group 1: HR = 3.00, 95% CI: 1.51-5.95, p = .002; Group 2: HR = 4.68, 95% CI: 2.40-9.11, p < .001; Group 3 as reference). Furthermore, external validation consistently demonstrated our findings.

CONCLUSIONS

Our study illustrated the feasibility of employing unsupervised learning for the classification of LA morphology. By utilizing morphological grouping, we can effectively identify individuals at different risks of postablative recurrence and thereby assist in clinical decision-making.

Augmenting K-Means Clustering With Qualitative Data to Discover the Engagement Patterns of Older Adults With Multimorbidity When Using Digital Health Technologies: Proof-of-Concept Trial

BACKGROUND

Multiple chronic conditions (multimorbidity) are becoming more prevalent among aging populations. Digital health technologies have the potential to assist in the self-management of multimorbidity, improving the awareness and monitoring of health and well-being, supporting a better understanding of the disease, and encouraging behavior change.

OBJECTIVE

The aim of this study was to analyze how 60 older adults (mean age 74, SD 6.4; range 65-92 years) with multimorbidity engaged with digital symptom and well-being monitoring when using a digital health platform over a period of approximately 12 months.

METHODS

Principal component analysis and clustering analysis were used to group participants based on their levels of engagement, and the data analysis focused on characteristics (eg, age, sex, and chronic health conditions), engagement outcomes, and symptom outcomes of the different clusters that were discovered.

RESULTS

Three clusters were identified: the typical user group, the least engaged user group, and the highly engaged user group. Our findings show that age, sex, and the types of chronic health conditions do not influence engagement. The 3 primary factors influencing engagement were whether the same device was used to submit different health and well-being parameters, the number of manual operations required to take a reading, and the daily routine of the participants. The findings also indicate that higher levels of engagement may improve the participants' outcomes (eg, reduce symptom exacerbation and increase physical activity).

CONCLUSIONS

The findings indicate potential factors that influence older adult engagement with digital health technologies for home-based multimorbidity self-management. The least engaged user groups showed decreased health and well-being outcomes related to multimorbidity self-management. Addressing the factors highlighted in this study in the design and implementation of home-based digital health technologies may improve symptom management and physical activity outcomes for older adults self-managing multimorbidity.

Pulmonary vascular reactivity in growth restricted fetuses using computational modelling and machine learning analysis of fetal Doppler waveforms

The aim of this study was to investigate the pulmonary vasculature in baseline conditions and after maternal hyperoxygenation in growth restricted fetuses (FGR). A prospective cohort study of singleton pregnancies including 97 FGR and 111 normally grown fetuses was carried out. Ultrasound Doppler of the pulmonary vessels was obtained at 24-37 weeks of gestation and data were acquired before and after oxygen administration. After, Machine Learning (ML) and a computational model were used on the Doppler waveforms to classify individuals and estimate pulmonary vascular resistance (PVR). Our results showed lower mean velocity time integral (VTI) in the main pulmonary and intrapulmonary arteries in baseline conditions in FGR individuals. Delta changes of the main pulmonary artery VTI and intrapulmonary artery pulsatility index before and after hyperoxygenation were significantly greater in FGR when compared with controls. Also, ML identified two clusters: A (including 66% controls and 34% FGR) with similar Doppler traces over time and B (including 33% controls and 67% FGR) with changes after hyperoxygenation. The computational model estimated the ratio of PVR before and after maternal hyperoxygenation which was closer to 1 in cluster A (cluster A 0.98 ± 0.33 vs cluster B 0.78 ± 0.28, p = 0.0156). Doppler ultrasound allows the detection of significant changes in pulmonary vasculature in most FGR at baseline, and distinct responses to hyperoxygenation. Future studies are warranted to assess its potential applicability in the clinical management of FGR.

Left main coronary artery morphological phenotypes and its hemodynamic properties

BACKGROUND

Atherosclerosis may be linked to morphological defects that lead to variances in coronary artery hemodynamics. Few objective strategies exit at present for generalizing morphological phenotypes of coronary arteries in terms of hemodynamics. We used unsupervised clustering (UC) to classify the morphology of the left main coronary artery (LM) and looked at how hemodynamic distribution differed between phenotypes.

METHODS

In this study, 76 LMs were obtained from 76 patients. After LMs were reconstructed with coronary computed tomography angiography, centerlines were used to extract the geometric characteristics. Unsupervised clustering was carried out using these characteristics to identify distinct morphological phenotypes of LMs. The time-averaged wall shear stress (TAWSS) for each phenotype was investigated by means of computational fluid dynamics (CFD) analysis of the left coronary artery.

RESULTS

We identified four clusters (i.e., four phenotypes): Cluster 1 had a shorter stem and thinner branches (n = 26); Cluster 2 had a larger bifurcation angle (n = 10); Cluster 3 had an ostium at an angulation to the coronary sinus and a more curved stem, and thick branches (n = 10); and Cluster 4 had an ostium at an angulation to the coronary sinus and a flatter stem (n = 14). TAWSS features varied widely across phenotypes. Nodes with low TAWSS (L-TAWSS) were typically found around the branching points of the left anterior descending artery (LAD), particularly in Cluster 2.

CONCLUSION

Our findings demonstrated that UC is a powerful technique for morphologically classifying LMs. Different LM phenotypes exhibited distinct hemodynamic characteristics in certain regions. This morphological clustering method could aid in identifying people at high risk for developing coronary atherosclerosis, hence facilitating early intervention.

Fluid dynamic analysis in predicting the recanalization of intracranial aneurysms after coil embolization - A study of spatiotemporal characteristics

Purpose

Hemodynamics play a key role in the management of cerebral aneurysm recanalization after coil embolization; however, the most reliable hemodynamic parameter remains unknown. Previous studies have explored the use of both spatiotemporally averaged and maximal definitions for hemodynamic parameters, based on computational fluid dynamics (CFD) analysis, to build predictive models for aneurysmal recanalization. In this study, we aimed to assess the influence of different spatiotemporal characteristics of hemodynamic parameters on predictive performance.

Methods

Hemodynamics were simulated using CFD for 66 cerebral aneurysms from 65 patients. We evaluated 14 types of spatiotemporal definitions for two hemodynamic parameters in the pre-coiling model and five in virtual post-coiling model (VM) created by cutting the aneurysm from the pre-coiling model. A total of 91 spatiotemporal hemodynamic features were derived and utilized to develop univariate predictor (UP) and multivariate logistic regression (LR) models. The model's performance was assessed using two metrics: the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC).

Results

Different spatiotemporal hemodynamic features exhibited a wide range of AUROC values ranging from 0.224 to 0.747, with 22 feature pairs showing a significant difference in AUROC value (P-value <0.05), despite being derived from the same hemodynamic parameter. PDave,q1 was identified as the strongest UP with AUROC/AUPRC values of 0.747/0.385, yielding sensitivity and specificity value of 0.889 and 0.614 at the optimal cut-off value, respectively. The LR model further improved the prediction performance, having AUROC/AUPRC values of 0.890/0.903. At the optimal cut-off value, the LR model achieved a specificity of 0.877, sensitivity of 0.719, outperforming the UP model.

Conclusion

Our research indicated that the characteristics of hemodynamic parameters in terms of space and time had a significant impact on the development of predictive model. Our findings suggest that LR model based on spatiotemporal hemodynamic features could be clinically useful in predicting recanalization after coil embolization in patients, without the need for invasive procedures.

Construction of a predictive model for blood transfusion in patients undergoing total hip arthroplasty and identification of clinical heterogeneity

A precise forecast of the need for blood transfusions (BT) in patients undergoing total hip arthroplasty (THA) is a crucial step toward the implementation of precision medicine. To achieve this goal, we utilized supervised machine learning (SML) techniques to establish a predictive model for BT requirements in THA patients. Additionally, we employed unsupervised machine learning (UML) approaches to identify clinical heterogeneity among these patients. In this study, we recruited 224 patients undergoing THA. To identify factors predictive of BT during the perioperative period of THA, we employed LASSO regression and the random forest (RF) algorithm as part of supervised machine learning (SML). Using logistic regression, we developed a predictive model for BT in THA patients. Furthermore, we utilized unsupervised machine learning (UML) techniques to cluster THA patients who required BT based on similar clinical features. The resulting clusters were subsequently visualized and validated. We constructed a predictive model for THA patients who required BT based on six predictive factors: Age, Body Mass Index (BMI), Hemoglobin (HGB), Platelet (PLT), Bleeding Volume, and Urine Volume. Before surgery, 1 h after surgery, 1 day after surgery, and 1 week after surgery, significant differences were observed in HGB and PLT levels between patients who received BT and those who did not. The predictive model achieved an AUC of 0.899. Employing UML, we identified two distinct clusters with significantly heterogeneous clinical characteristics. Age, BMI, PLT, HGB, bleeding volume, and urine volume were found to be independent predictors of BT requirement in THA patients. The predictive model incorporating these six predictors demonstrated excellent predictive performance. Furthermore, employing UML enabled us to classify a heterogeneous cohort of THA patients who received BT in a meaningful and interpretable manner.

Artificial intelligence models in prediction of response to cardiac resynchronization therapy: a systematic review

The aim of the presented review is to summarize the literature data on the accuracy and clinical applicability of artificial intelligence (AI) models as a valuable alternative to the current guidelines in predicting cardiac resynchronization therapy (CRT) response and phenotyping of patients eligible for CRT implantation. This systematic review was performed according to the PRISMA guidelines. After a search of Scopus, PubMed, Cochrane Library, and Embase databases, 675 records were identified. Twenty supervised (prediction of CRT response) and 9 unsupervised (clustering and phenotyping) AI models were analyzed qualitatively (22 studies, 14,258 patients). Fifty-five percent of AI models were based on retrospective studies. Unsupervised AI models were able to identify clusters of patients with significantly different rates of primary outcome events (death, heart failure event). In comparison to the guideline-based CRT response prediction accuracy of 70%, supervised AI models trained on cohorts with > 100 patients achieved up to 85% accuracy and an AUC of 0.86 in their prediction of response to CRT for echocardiographic and clinical outcomes, respectively. AI models seem to be an accurate and clinically applicable tool in phenotyping of patients eligible for CRT implantation and predicting potential responders. In the future, AI may help to increase CRT response rates to over 80% and improve clinical decision-making and prognosis of the patients, including reduction of mortality rates. However, these findings must be validated in randomized controlled trials.

Application and Potential of Artificial Intelligence in Heart Failure: Past, Present, and Future

The prevalence of heart failure (HF) is increasing, necessitating accurate diagnosis and tailored treatment. The accumulation of clinical information from patients with HF generates big data, which poses challenges for traditional analytical methods. To address this, big data approaches and artificial intelligence (AI) have been developed that can effectively predict future observations and outcomes, enabling precise diagnoses and personalized treatments of patients with HF. Machine learning (ML) is a subfield of AI that allows computers to analyze data, find patterns, and make predictions without explicit instructions. ML can be supervised, unsupervised, or semi-supervised. Deep learning is a branch of ML that uses artificial neural networks with multiple layers to find complex patterns. These AI technologies have shown significant potential in various aspects of HF research, including diagnosis, outcome prediction, classification of HF phenotypes, and optimization of treatment strategies. In addition, integrating multiple data sources, such as electrocardiography, electronic health records, and imaging data, can enhance the diagnostic accuracy of AI algorithms. Currently, wearable devices and remote monitoring aided by AI enable the earlier detection of HF and improved patient care. This review focuses on the rationale behind utilizing AI in HF and explores its various applications.

Five or more gender- and size-diverse customizations of distal femur prostheses are needed to improve fit for Chinese knees

PURPOSE

Mismatch between partial imported prostheses and Chinese anatomy affects the clinical outcome of the procedure. The purpose of this study was to measure the anatomical dimensions of Chinese distal femurs to provide guidance for the design of more compatible distal femoral prostheses.

METHODS

A total of 406 healthy distal femurs were reconstructed and measured. Consistency of these measurements and differences in sides, gender, and populations were examined. Parameter correlations were analyzed, and pairs with strong correlations underwent linear regression analysis. The design of distal femoral prostheses was referenced from the results of K-means and hierarchical clustering analysis.

RESULTS

Ten parameters were measured, including the trans-epicondylar axis, width of the distal femur (ML), anteroposterior diameter of the distal femur (AP), etc. The intra-class correlation coefficient ranged from 0.795 to 0.999 for intra-observer consistency, and from 0.796 to 0.998 for inter-observer consistency. Males exhibited significantly larger parameters than females, except for the posterior condylar angle (all P values < 0.05). Compared to other populations, substantial differences were observed for most parameters, such as ML, AP, width of lateral femoral condyle, etc. (all P values < 0.05). Clustering analysis suggested that distal femoral prostheses should include at least five sizes to adequately accommodate the sampled population. ML sizes for males were 68, 70, 83, 73, and 89 mm, and for females 64, 65, 71, 67, and 77 mm. AP sizes for males were 56, 60, 60, 64, and 64 mm, and for females 48, 52, 54, 57, and 58 mm.

CONCLUSIONS

Chinese distal femur morphology, as analyzed using 3D techniques, varies significantly between genders and when compared with international data. For improved patient fit, the creation of five or more distal femur prostheses, diversified by gender and size and informed by the associated morphological parameters, is recommended.

LEVEL OF EVIDENCE

IV.

Virtual healthcare solutions in heart failure: a literature review

The widespread adoption of mobile technologies offers an opportunity for a new approach to post-discharge care for patients with heart failure (HF). By enabling non-invasive remote monitoring and two-way, real-time communication between the clinic and home-based patients, as well as a host of other capabilities, mobile technologies have a potential to significantly improve remote patient care. This literature review summarizes clinical evidence related to virtual healthcare (VHC), defined as a care team + connected devices + a digital solution in post-release care of patients with HF. Searches were conducted on Embase (06/12/2020). A total of 171 studies were included for data extraction and evidence synthesis: 96 studies related to VHC efficacy, and 75 studies related to AI in HF. In addition, 15 publications were included from the search on studies scaling up VHC solutions in HF within the real-world setting. The most successful VHC interventions, as measured by the number of reported significant results, were those targeting reduction in rehospitalization rates. In terms of relative success rate, the two most effective interventions targeted patient self-care and all-cause hospital visits in their primary endpoint. Among the three categories of VHC identified in this review (telemonitoring, remote patient management, and patient self-empowerment) the integrated approach in remote patient management solutions performs the best in decreasing HF patients' re-admission rates and overall hospital visits. Given the increased amount of data generated by VHC technologies, artificial intelligence (AI) is being investigated as a tool to aid decision making in the context of primary diagnostics, identifying disease phenotypes, and predicting treatment outcomes. Currently, most AI algorithms are developed using data gathered in clinic and only a few studies deploy AI in the context of VHC. Most successes have been reported in predicting HF outcomes. Since the field of VHC in HF is relatively new and still in flux, this is not a typical systematic review capturing all published studies within this domain. Although the standard methodology for this type of reviews was followed, the nature of this review is qualitative. The main objective was to summarize the most promising results and identify potential research directions.

Artificial intelligence and heart failure: A state-of-the-art review

Heart failure (HF) is a heterogeneous syndrome affecting more than 60 million individuals globally. Despite recent advancements in understanding of the pathophysiology of HF, many issues remain including residual risk despite therapy, understanding the pathophysiology and phenotypes of patients with HF and preserved ejection fraction, and the challenges related to integrating a large amount of disparate information available for risk stratification and management of these patients. Risk prediction algorithms based on artificial intelligence (AI) may have superior predictive ability compared to traditional methods in certain instances. AI algorithms can play a pivotal role in the evolution of HF care by facilitating clinical decision making to overcome various challenges such as allocation of treatment to patients who are at highest risk or are more likely to benefit from therapies, prediction of adverse outcomes, and early identification of patients with subclinical disease or worsening HF. With the ability to integrate and synthesize large amounts of data with multidimensional interactions, AI algorithms can supply information with which physicians can improve their ability to make timely and better decisions. In this review, we provide an overview of the AI algorithms that have been developed for establishing early diagnosis of HF, phenotyping HF with preserved ejection fraction, and stratifying HF disease severity. This review also discusses the challenges in clinical deployment of AI algorithms in HF, and the potential path forward for developing future novel learning-based algorithms to improve HF care.

Topological data analysis to identify cardiac resynchronization therapy patients exhibiting benefit from an implantable cardioverter-defibrillator

BACKGROUND

Current guidelines recommend considering multiple factors while deciding between cardiac resynchronization therapy with a defibrillator (CRT-D) or a pacemaker (CRT-P). Nevertheless, it is still challenging to pinpoint those candidates who will benefit from choosing a CRT-D device in terms of survival.

OBJECTIVE

We aimed to use topological data analysis (TDA) to identify phenogroups of CRT patients in whom CRT-D is associated with better survival than CRT-P.

METHODS

We included 2603 patients who underwent CRT-D (54%) or CRT-P (46%) implantation at Semmelweis University between 2000 and 2018. The primary endpoint was all-cause mortality. We applied TDA to create a patient similarity network using 25 clinical features. Then, we identified multiple phenogroups in the generated network and compared the groups' clinical characteristics and survival.

RESULTS

Five- and 10-year mortality were 43 (40-46)% and 71 (67-74)% in patients with CRT-D and 48 (45-50)% and 71 (68-74)% in those with CRT-P, respectively. TDA created a circular network in which we could delineate five phenogroups showing distinct patterns of clinical characteristics and outcomes. Three phenogroups (1, 2, and 3) included almost exclusively patients with non-ischemic etiology, whereas the other two phenogroups (4 and 5) predominantly comprised ischemic patients. Interestingly, only in phenogroups 2 and 5 were CRT-D associated with better survival than CRT-P (adjusted hazard ratio 0.61 [0.47-0.80], p < 0.001 and adjusted hazard ratio 0.84 [0.71-0.99], p = 0.033, respectively).

CONCLUSIONS

By simultaneously evaluating various clinical features, TDA may identify patients with either ischemic or non-ischemic etiology who will most likely benefit from the implantation of a CRT-D instead of a CRT-P. Topological data analysis to identify phenogroups of CRT patients in whom CRT-D is associated with better survival than CRT-P. AF atrial fibrillation, CRT cardiac resynchronization therapy, CRT-D cardiac resynchronization therapy defibrillator, CRT-P cardiac resynchronization therapy pacemaker, DM diabetes mellitus, HTN hypertension, LBBB left bundle branch block, LVEF left ventricular ejection fraction, MDS multidimensional scaling, MRA mineralocorticoid receptor antagonist, NYHA New York Heart Association.

Unsupervised machine learning model to predict cognitive impairment in subcortical ischemic vascular disease

INTRODUCTION

It is challenging to predict which patients who meet criteria for subcortical ischemic vascular disease (SIVD) will ultimately progress to subcortical vascular cognitive impairment (SVCI).

METHODS

We collected clinical information, neuropsychological assessments, T1 imaging, diffusion tensor imaging, and resting-state functional magnetic resonance imaging from 83 patients with SVCI and 53 age-matched patients with SIVD without cognitive impairment. We built an unsupervised machine learning model to isolate patients with SVCI. The model was validated using multimodal data from an external cohort comprising 45 patients with SVCI and 32 patients with SIVD without cognitive impairment.

RESULTS

The accuracy, sensitivity, and specificity of the unsupervised machine learning model were 86.03%, 79.52%, and 96.23% and 80.52%, 71.11%, and 93.75% for internal and external cohort, respectively.

DISCUSSION

We developed an accurate and accessible clinical tool which requires only data from routine imaging to predict patients at risk of progressing from SIVD to SVCI.

HIGHLIGHTS

Our unsupervised machine learning model provides an accurate and accessible clinical tool to predict patients at risk of progressing from subcortical ischemic vascular disease (SIVD) to subcortical vascular cognitive impairment (SVCI) and requires only data from imaging routinely used during the diagnosis of suspected SVCI. The model yields good accuracy, sensitivity, and specificity and is portable to other cohorts and to clinical practice to distinguish patients with SIVD at risk for progressing to SVCI. The model combines assessment of diffusion tensor imaging and functional magnetic resonance imaging measures in patients with SVCI to analyze whether the "disconnection hypothesis" contributes to functional and structural changes and to the clinical presentation of SVCI.

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.

Combination of personalized computational modeling and machine learning for optimization of left ventricular pacing site in cardiac resynchronization therapy

Introduction: The 30-50% non-response rate to cardiac resynchronization therapy (CRT) calls for improved patient selection and optimized pacing lead placement. The study aimed to develop a novel technique using patient-specific cardiac models and machine learning (ML) to predict an optimal left ventricular (LV) pacing site (ML-PS) that maximizes the likelihood of LV ejection fraction (LVEF) improvement in a given CRT candidate. To validate the approach, we evaluated whether the distance D_PS between the clinical LV pacing site (ref-PS) and ML-PS is associated with improved response rate and magnitude. Materials and methods: We reviewed retrospective data for 57 CRT recipients. A positive response was defined as a more than 10% LVEF improvement. Personalized models of ventricular activation and ECG were created from MRI and CT images. The characteristics of ventricular activation during intrinsic rhythm and biventricular (BiV) pacing with ref-PS were derived from the models and used in combination with clinical data to train supervised ML classifiers. The best logistic regression model classified CRT responders with a high accuracy of 0.77 (ROC AUC = 0.84). The LR classifier, model simulations and Bayesian optimization with Gaussian process regression were combined to identify an optimal ML-PS that maximizes the ML-score of CRT response over the LV surface in each patient. Results: The optimal ML-PS improved the ML-score by 17 ± 14% over the ref-PS. Twenty percent of the non-responders were reclassified as positive at ML-PS. Selection of positive patients with a max ML-score >0.5 demonstrated an improved clinical response rate. The distance D_PS was shorter in the responders. The max ML-score and D_PS were found to be strong predictors of CRT response (ROC AUC = 0.85). In the group with max ML-score > 0.5 and D_PS< 30 mm, the response rate was 83% compared to 14% in the rest of the cohort. LVEF improvement in this group was higher than in the other patients (16 ± 8% vs. 7 ± 8%). Conclusion: A new technique combining clinical data, personalized heart modelling and supervised ML demonstrates the potential for use in clinical practice to assist in optimizing patient selection and predicting optimal LV pacing lead position in HF candidates for CRT.

Robust Framework for Medical Time Series Classification and Application to Real Scenarios in Modern Bioengineering

In this study, a novel unsupervised classification framework for time series of medical nature is presented. This framework is based on the intersection of machine learning, Hilbert Spaces algebra, and signal theory. The methodology is illustrated through the resolution of three biomedical engineering problems: neuronal activity tracking, protein functional classification, and non-invasive diagnosis of atrial flutter (AFL). The results indicate that the proposed algorithms exhibit high proficiency in solving these tasks and demonstrate robustness in identifying damaged neuronal units while tracking healthy ones. Moreover, the application of the framework in protein functional classification provides a new perspective for the development of pharmaceutical products and personalised medicine. Additionally, the controlled environment of the framework in AFL simulation problem underscores the algorithm's ability to encode information efficiently. These results offer valuable insights into the potential of this framework and lay the groundwork for future studies.Clinical relevance- The framework proposed in this study has the potential to yield novel insights into the effects of newly implanted electrodes in the brain. Furthermore, the categorization of proteins by function could facilitate the development of personalised and efficient medicines, ultimately reducing both time and cost. The simulation of atrial flutter also demonstrates the framework's ability to encode information for arrhythmia diagnosis and treatment, which has the potential to lead to improved patient outcomes.

Are current clinical studies on artificial intelligence-based medical devices comprehensive enough to support a full health technology assessment? A systematic review

INTRODUCTION

Artificial Intelligence-based Medical Devices (AI-based MDs) are experiencing exponential growth in healthcare. This study aimed to investigate whether current studies assessing AI contain the information required for health technology assessment (HTA) by HTA bodies.

METHODS

We conducted a systematic literature review based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses methodology to extract articles published between 2016 and 2021 related to the assessment of AI-based MDs. Data extraction focused on study characteristics, technology, algorithms, comparators, and results. AI quality assessment and HTA scores were calculated to evaluate whether the items present in the included studies were concordant with the HTA requirements. We performed a linear regression for the HTA and AI scores with the explanatory variables of the impact factor, publication date, and medical specialty. We conducted a univariate analysis of the HTA score and a multivariate analysis of the AI score with an alpha risk of 5 %.

RESULTS

Of 5578 retrieved records, 56 were included. The mean AI quality assessment score was 67 %; 32 % of articles had an AI quality score ≥ 70 %, 50 % had a score between 50 % and 70 %, and 18 % had a score under 50 %. The highest quality scores were observed for the study design (82 %) and optimisation (69 %) categories, whereas the scores were lowest in the clinical practice category (23 %). The mean HTA score was 52 % for all seven domains. 100 % of the studies assessed clinical effectiveness, whereas only 9 % evaluated safety, and 20 % evaluated economic issues. There was a statistically significant relationship between the impact factor and the HTA and AI scores (both p = 0.046).

DISCUSSION

Clinical studies on AI-based MDs have limitations and often lack adapted, robust, and complete evidence. High-quality datasets are also required because the output data can only be trusted if the inputs are reliable. The existing assessment frameworks are not specifically designed to assess AI-based MDs. From the perspective of regulatory authorities, we suggest that these frameworks should be adapted to assess the interpretability, explainability, cybersecurity, and safety of ongoing updates. From the perspective of HTA agencies, we highlight that transparency, professional and patient acceptance, ethical issues, and organizational changes are required for the implementation of these devices. Economic assessments of AI should rely on a robust methodology (business impact or health economic models) to provide decision-makers with more reliable evidence.

CONCLUSION

Currently, AI studies are insufficient to cover HTA prerequisites. HTA processes also need to be adapted because they do not consider the important specificities of AI-based MDs. Specific HTA workflows and accurate assessment tools should be designed to standardise evaluations, generate reliable evidence, and create confidence.

Machine learning prediction model for treatment responders in patients with primary biliary cholangitis

Background and Aim

Treatment response to ursodeoxycholic acid may predict the prognosis of patients with primary biliary cholangitis (PBC). Recent studies have suggested the benefits of using machine learning (ML) to forecast complex medical predictions. We aimed to predict treatment response in patients with PBC using ML and pretreatment data.

Methods

We conducted a single-center retrospective study and collected data from 194 patients with PBC who were followed up for at least 12 months after treatment initiation. Patient data were analyzed with five ML models, namely random forest, extreme gradient boosting (XGB), decision tree, naïve Bayes, or logistic regression, to predict treatment response using the Paris II criteria. The established models were assessed using an out-of-sample validation. The area under the curve (AUC) was used to evaluate the efficacy of each algorithm. Overall survival and liver-related deaths were analyzed using Kaplan-Meier analysis.

Results

Compared to logistic regression (AUC = 0.595, P = 0.0219, 0.031 models), ML analyses showed significantly high AUC in the random forest (AUC = 0.84) and XGB (AUC = 0.83) models; however, the AUC was not significantly high for decision tree (AUC = 0.633) or naïve Bayes (AUC = 0.584) models. Kaplan-Meier analysis showed significantly improved prognoses in patients predicted to achieve the Paris II criteria by XGB (log-rank = 0.005 and 0.007).

Conclusion

ML algorithms could improve treatment response prediction using pretreatment data, which could lead to better prognoses. In addition, the ML model using XGB could predict the prognosis of patients before treatment initiation.

Unsupervised machine learning effectively clusters pediatric spastic cerebral palsy patients for determination of optimal responders to selective dorsal rhizotomy

Selective dorsal rhizotomy (SDR) can reduce the spasticity in patients with spastic cerebral palsy (SCP) and thus improve the motor function in these patients, but different levels of improvement in motor function were observed among patients after SDR. The aim of the present study was to subgroup patients and to predict the possible outcome of SDR based on the pre-operational parameters. A hundred and thirty-five pediatric patients diagnosed with SCP who underwent SDR from January 2015 to January 2021 were retrospectively reviewed. Spasticity of lower limbs, the number of target muscles, motor functions, and other clinical parameters were used as input variables for unsupervised machine learning to cluster all included patients. The postoperative motor function change is used to assess the clinical significance of clustering. After the SDR procedure, the spasticity of muscles in all patients was reduced significantly, and the motor function was promoted significantly at the follow-up duration. All patients were categorized into three subgroups by both hierarchical and K-means clustering methods. The three subgroups showed significantly different clinical characteristics except for the age at surgery, and the post-operational motor function change at the last follow-up in these three clusters was different. Three subgroups clustered by two methods could be identified as "best responders", "good responders" and "moderate responders" based on the increasement of motor function after SDR. Clustering results achieved by hierarchical and K-means algorithms showed high consistency in subgrouping the whole group of patients. These results indicated that SDR could relieve the spasticity and promote the motor function of patients with SCP. Unsupervised machine learning methods can effectively and accurately cluster patients into different subgroups suffering from SCP based on pre-operative characteristics. Machine learning can be used for the determination of optimal responders for SDR surgery.

Artificial Intelligence Technologies in Cardiology

As the world produces exabytes of data, there is a growing need to find new methods that are more suitable for dealing with complex datasets. Artificial intelligence (AI) has significant potential to impact the healthcare industry, which is already on the road to change with the digital transformation of vast quantities of information. The implementation of AI has already achieved success in the domains of molecular chemistry and drug discoveries. The reduction in costs and in the time needed for experiments to predict the pharmacological activities of new molecules is a milestone in science. These successful applications of AI algorithms provide hope for a revolution in healthcare systems. A significant part of artificial intelligence is machine learning (ML), of which there are three main types-supervised learning, unsupervised learning, and reinforcement learning. In this review, the full scope of the AI workflow is presented, with explanations of the most-often-used ML algorithms and descriptions of performance metrics for both regression and classification. A brief introduction to explainable artificial intelligence (XAI) is provided, with examples of technologies that have developed for XAI. We review important AI implementations in cardiology for supervised, unsupervised, and reinforcement learning and natural language processing, emphasizing the used algorithm. Finally, we discuss the need to establish legal, ethical, and methodical requirements for the deployment of AI models in medicine.

The saga of dyssynchrony imaging: Are we getting to the point

Cardiac resynchronisation therapy (CRT) has an established role in the management of patients with heart failure, reduced left ventricular ejection fraction (LVEF &lt; 35%) and widened QRS (&gt;130 msec). Despite the complex pathophysiology of left ventricular (LV) dyssynchrony and the increasing evidence supporting the identification of specific electromechanical substrates that are associated with a higher probability of CRT response, the assessment of LVEF is the only imaging-derived parameter used for the selection of CRT candidates.

This review aims to (1) provide an overview of the evolution of cardiac imaging for the assessment of LV dyssynchrony and its role in the selection of patients undergoing CRT; (2) highlight the main pitfalls and advantages of the application of cardiac imaging for the assessment of LV dyssynchrony; (3) provide some perspectives for clinical application and future research in this field.

Conclusion

the road for a more individualized approach to resynchronization therapy delivery is open and imaging might provide important input beyond the assessment of LVEF.

Identification of clinical heterogeneity and construction of a novel subtype predictive model in patients with ankylosing spondylitis: An unsupervised machine learning study

BACKGROUND

Accurate classification of patients with ankylosing spondylitis (AS) is the premise of precision medicine so as to perform different medical interventions for different patient types. AS pathology is closely related to the changes in the immune microenvironment. In this study, we used unsupervised machine learning (UML) to classify patients with AS based on clinical characteristics. We then constructed a novel subtype predictive model for AS based on the clinical classification, after which we investigated the difference in the immune microenvironment to unravel the AS pathogenesis.

METHODS

Overall, 196 patients with AS were enrolled. UML was used to cluster AS patients by similar clinical characteristics. Functional ability, disease status, and grading of radiologic features were assessed to verify the accuracy and heterogeneity of UML clustering. Least Absolute Shrinkage and Selection Operator (LASSO) regression and Random Forest algorithm were used to screen and identify predictive factors for the novel subtype of AS. Logistic regression was also performed to construct a predictive model of this novel subtype. Datasets were downloaded from the Gene Expression Omnibus database to assess immune cell infiltration, and the results were validated using data of routine blood tests from 3671 AS patients and 5720 non-AS patients. The differential expression of Fat Mass and Obesity-Associated Protein (FTO), an m⁶A regulator, between AS patients and healthy control subjects was confirmed using immunohistochemistry.

RESULTS

UML clustering identified two clusters. The clinical characteristics of the two clusters were significantly heterogeneous. For the novel subtype of AS identified in UML clustering, a predictive model was built using three predictive factors, namely, C-reactive protein (CRP), absolute value of neutrophils (NEU), and absolute value of monocytes (MONO). The area under the curve of the predictive model was 0.983. Heterogeneity in the neutrophil and monocyte counts in AS was verified through immune cell infiltration analysis. Data from routine blood tests revealed that NEU and MONO were significantly higher in AS patients than in non-AS patients (p < 0.001). FTO expression was negatively correlated with both NEU and MONO. Immunohistochemistry analysis confirmed the downregulated expression of FTO.

CONCLUSIONS

UML provides an explicable and remarkable classification of a heterogeneous cohort of AS patients. A novel subtype of AS was identified in UML clustering. CRP, NEU, and MONO were the independent predictive factors for the novel subtype of AS. FTO expression was correlated with immune cell infiltration in AS patients.

A streamlined, machine learning-derived approach to risk-stratification in heart failure patients with secondary tricuspid regurgitation

AIMS

Secondary tricuspid regurgitation (sTR) is the most frequent valvular heart disease and has a significant impact on mortality. A high burden of comorbidities often worsens the already dismal prognosis of sTR, while tricuspid interventions remain underused and initiated too late. The aim was to examine the most powerful predictors of all-cause mortality in moderate and severe sTR using machine learning techniques and to provide a streamlined approach to risk-stratification using readily available clinical, echocardiographic and laboratory parameters.

METHODS AND RESULTS

This large-scale, long-term observational study included 3359 moderate and 1509 severe sTR patients encompassing the entire heart failure spectrum (preserved, mid-range and reduced ejection fraction). A random survival forest was applied to investigate the most important predictors and group patients according to their number of adverse features.The identified predictors and thresholds, that were associated with significantly worse mortality were lower glomerular filtration rate (<60 mL/min/1.73m2), higher NT-proBNP, increased high sensitivity C-reactive protein, serum albumin < 40 g/L and hemoglobin < 13 g/dL. Additionally, grouping patients according to the number of adverse features yielded important prognostic information, as patients with 4 or 5 adverse features had a fourfold risk increase in moderate sTR [4.81(3.56-6.50) HR 95%CI, P < 0.001] and fivefold risk increase in severe sTR [5.33 (3.28-8.66) HR 95%CI, P < 0.001].

CONCLUSION

This study presents a streamlined, machine learning-derived and internally validated approach to risk-stratification in patients with moderate and severe sTR, that adds important prognostic information to aid clinical-decision-making.

Predicting transient ischemic attack risk in patients with mild carotid stenosis using machine learning and CT radiomics

Objective

This study aims to establish a radiomics-based machine learning model that predicts the risk of transient ischemic attack in patients with mild carotid stenosis (30-50% North American Symptomatic Carotid Endarterectomy Trial) using extracted computed tomography radiomics features and clinical information.

Methods

A total of 179 patients underwent carotid computed tomography angiography (CTA), and 219 carotid arteries with a plaque at the carotid bifurcation or proximal to the internal carotid artery were selected. The patients were divided into two groups; patients with symptoms of transient ischemic attack after CTA and patients without symptoms of transient ischemic attack after CTA. Then we performed random sampling methods stratified by the predictive outcome to obtain the training set (N = 165) and testing set (N = 66). 3D Slicer was employed to select the site of plaque on the computed tomography image as the volume of interest. An open-source package PyRadiomics in Python was used to extract radiomics features from the volume of interests. The random forest and logistic regression models were used to screen feature variables, and five classification algorithms were used, including random forest, eXtreme Gradient Boosting, logistic regression, support vector machine, and k-nearest neighbors. Data on radiomic feature information, clinical information, and the combination of these pieces of information were used to generate the model that predicts the risk of transient ischemic attack in patients with mild carotid artery stenosis (30-50% North American Symptomatic Carotid Endarterectomy Trial).

Results

The random forest model that was built based on the radiomics and clinical feature information had the highest accuracy (area under curve = 0.879; 95% confidence interval, 0.787-0.979). The combined model outperformed the clinical model, whereas the combined model showed no significant difference from the radiomics model.

Conclusion

The random forest model constructed with both radiomics and clinical information can accurately predict and improve discriminative power of computed tomography angiography in identifying ischemic symptoms in patients with carotid atherosclerosis. This model can aid in guiding the follow-up treatment of patients at high risk.

A method using deep learning to discover new predictors from left-ventricular mechanical dyssynchrony for CRT response

BACKGROUND

Studies have shown that the conventional parameters characterizing left ventricular mechanical dyssynchrony (LVMD) measured on gated SPECT myocardial perfusion imaging (MPI) have their own statistical limitations in predicting cardiac resynchronization therapy (CRT) response. The purpose of this study is to discover new predictors from the polarmaps of LVMD by deep learning to help select heart failure patients with a high likelihood of response to CRT.

METHODS

One hundred and fifty-seven patients who underwent rest gated SPECT MPI were enrolled in this study. CRT response was defined as an increase in left ventricular ejection fraction (LVEF) > 5% at 6 [Formula: see text] 1 month follow up. The autoencoder (AE) technique, an unsupervised deep learning method, was applied to the polarmaps of LVMD to extract new predictors characterizing LVMD. Pearson correlation analysis was used to explain the relationships between new predictors and existing clinical parameters. Patients from the IAEA VISION-CRT trial were used for an external validation. Heatmaps were used to interpret the AE-extracted feature.

RESULTS

Complete data were obtained in 130 patients, and 68.5% of them were classified as CRT responders. After variable selection by feature importance ranking and correlation analysis, one AE-extracted LVMD predictor was included in the statistical analysis. This new AE-extracted LVMD predictor showed statistical significance in the univariate (OR 2.00, P = .026) and multivariate (OR 1.11, P = .021) analyses, respectively. Moreover, the new AE-extracted LVMD predictor not only had incremental value over PBW and significant clinical variables, including QRS duration and left ventricular end-systolic volume (AUC 0.74 vs 0.72, LH 7.33, P = .007), but also showed encouraging predictive value in the 165 patients from the IAEA VISION-CRT trial (P < .1). The heatmaps for calculation of the AE-extracted predictor showed higher weights on the anterior, lateral, and inferior myocardial walls, which are recommended as LV pacing sites in clinical practice.

CONCLUSIONS

AE techniques have significant value in the discovery of new clinical predictors. The new AE-extracted LVMD predictor extracted from the baseline gated SPECT MPI has the potential to improve the prediction of CRT response.

Cardiac Structure and Function Phenogroups and Risk of Incident Heart Failure (from the Multi-ethnic Study of Atherosclerosis)

Indices of cardiac structure and function, such as left ventricular (LV) mass and ejection fraction, have been associated with risk of incident heart failure (HF), but the clinical relevance of data-driven grouping of a comprehensive set of cardiac parameters is unclear. In Multi-Ethnic Study of Atherosclerosis participants, latent class analysis was applied in the sample stratified by gender to define phenogroups on the basis of cardiovascular magnetic resonance imaging parameters of right ventricular and LV structure and function at baseline. Cox proportional hazard models in gender-stratified analyses were used to assess the association between phenogroup membership and risk of HF subtypes adjusting for potential confounders. In the 4,204 participants (mean age 61 ± 10 years, 53% women), the mean follow-up time was 14 ± 4 years for men and 15 ± 4 years for women. For both genders, 4 distinct phenogroups were identified: (1) ideal cardiac mechanics; (2) higher output/hypertrophied LV; (3) impaired ejection fraction/dilated LV; and (4) higher output/hyperdynamic (LV). Men in phenogroups 4 (hazard ratio [HR] 2.91, 95% confidence interval [CI] 1.60 to 5.31, p = 0.0005), 3 (HR 3.52, 95% CI 1.90 to 6.53, p <0.0001), and 2 (HR 3.49, 95% CI 1.94 to 6.28, p <0.0001) had higher rates of incident HF than did men in phenogroup 1, in fully adjusted models. No significant associations were found between phenogroup membership and incident HF in women. In conclusion, phenogroup membership based on cardiac structure and function in men was significantly associated with incident HF. Integration of cardiac magnetic resonance imaging variables may help identify differential risk for HF in men.

Electrocardiogram-based deep learning improves outcome prediction following cardiac resynchronization therapy

AIMS

This study aims to identify and visualize electrocardiogram (ECG) features using an explainable deep learning-based algorithm to predict cardiac resynchronization therapy (CRT) outcome. Its performance is compared with current guideline ECG criteria and QRSAREA.

METHODS AND RESULTS

A deep learning algorithm, trained on 1.1 million ECGs from 251 473 patients, was used to compress the median beat ECG, thereby summarizing most ECG features into only 21 explainable factors (FactorECG). Pre-implantation ECGs of 1306 CRT patients from three academic centres were converted into their respective FactorECG. FactorECG predicted the combined clinical endpoint of death, left ventricular assist device, or heart transplantation [c-statistic 0.69, 95% confidence interval (CI) 0.66-0.72], significantly outperforming QRSAREA and guideline ECG criteria [c-statistic 0.61 (95% CI 0.58-0.64) and 0.57 (95% CI 0.54-0.60), P < 0.001 for both]. The addition of 13 clinical variables was of limited added value for the FactorECG model when compared with QRSAREA (Δ c-statistic 0.03 vs. 0.10). FactorECG identified inferolateral T-wave inversion, smaller right precordial S- and T-wave amplitude, ventricular rate, and increased PR interval and P-wave duration to be important predictors for poor outcome. An online visualization tool was created to provide interactive visualizations (https://crt.ecgx.ai).

CONCLUSION

Requiring only a standard 12-lead ECG, FactorECG held superior discriminative ability for the prediction of clinical outcome when compared with guideline criteria and QRSAREA, without requiring additional clinical variables. End-to-end automated visualization of ECG features allows for an explainable algorithm, which may facilitate rapid uptake of this personalized decision-making tool in CRT.

Interpretable machine learning predicts cardiac resynchronization therapy responses from personalized biochemical and biomechanical features

BACKGROUND

Cardiac Resynchronization Therapy (CRT) is a widely used, device-based therapy for patients with left ventricle (LV) failure. Unfortunately, many patients do not benefit from CRT, so there is potential value in identifying this group of non-responders before CRT implementation. Past studies suggest that predicting CRT response will require diverse variables, including demographic, biomarker, and LV function data. Accordingly, the objective of this study was to integrate diverse variable types into a machine learning algorithm for predicting individual patient responses to CRT.

METHODS

We built an ensemble classification algorithm using previously acquired data from the SMART-AV CRT clinical trial (n = 794 patients). We used five-fold stratified cross-validation on 80% of the patients (n = 635) to train the model with variables collected at 0 months (before initiating CRT), and the remaining 20% of the patients (n = 159) were used as a hold-out test set for model validation. To improve model interpretability, we quantified feature importance values using SHapley Additive exPlanations (SHAP) analysis and used Local Interpretable Model-agnostic Explanations (LIME) to explain patient-specific predictions.

RESULTS

Our classification algorithm incorporated 26 patient demographic and medical history variables, 12 biomarker variables, and 18 LV functional variables, which yielded correct prediction of CRT response in 71% of patients. Additional patient stratification to identify the subgroups with the highest or lowest likelihood of response showed 96% accuracy with 22 correct predictions out of 23 patients in the highest and lowest responder groups.

CONCLUSION

Computationally integrating general patient characteristics, comorbidities, therapy history, circulating biomarkers, and LV function data available before CRT intervention can improve the prediction of individual patient responses.

Big Data in cardiac surgery: real world and perspectives

Big Data, and the derived analysis techniques, such as artificial intelligence and machine learning, have been considered a revolution in the modern practice of medicine. Big Data comes from multiple sources, encompassing electronic health records, clinical studies, imaging data, registries, administrative databases, patient-reported outcomes and OMICS profiles. The main objective of such analyses is to unveil hidden associations and patterns. In cardiac surgery, the main targets for the use of Big Data are the construction of predictive models to recognize patterns or associations better representing the individual risk or prognosis compared to classical surgical risk scores. The results of these studies contributed to kindle the interest for personalized medicine and contributed to recognize the limitations of randomized controlled trials in representing the real world. However, the main sources of evidence for guidelines and recommendations remain RCTs and meta-analysis. The extent of the revolution of Big Data and new analytical models in cardiac surgery is yet to be determined.

Artificial intelligence in cardiology: Hope for the future and power for the present

Cardiovascular disease (CVD) is the principal cause of mortality and morbidity globally. With the pressures for improved care and translation of the latest medical advances and knowledge to an actionable plan, clinical decision-making for cardiologists is challenging. Artificial Intelligence (AI) is a field in computer science that studies the design of intelligent agents which take the best feasible action in a situation. It incorporates the use of computational algorithms which simulate and perform tasks that traditionally require human intelligence such as problem solving and learning. Whilst medicine is arguably the last to apply AI in its everyday routine, cardiology is at the forefront of AI revolution in the medical field. The development of AI methods for accurate prediction of CVD outcomes, non-invasive diagnosis of coronary artery disease (CAD), detection of malignant arrythmias through wearables, and diagnosis, treatment strategies and prediction of outcomes for heart failure (HF) patients, demonstrates the potential of AI in future cardiology. With the advancements of AI, Internet of Things (IoT) and the promotion of precision medicine, the future of cardiology will be heavily based on these innovative digital technologies. Despite this, ethical dilemmas regarding the implementation of AI technologies in real-world are still unaddressed.

Emerging role of artificial intelligence in cardiac electrophysiology

Artificial intelligence (AI) and machine learning (ML) have significantly impacted the field of cardiovascular medicine, especially cardiac electrophysiology (EP), on multiple fronts. The goal of this review is to familiarize readers with the field of AI and ML and their emerging role in EP. The current review is divided into 3 sections. In the first section, we discuss the definitions and basics of AI, ML, and big data. In the second section, we discuss their application to EP in the context of detection, prediction, and management of arrhythmias. Finally, we discuss the regulatory issues, challenges, and future directions of AI in EP.

An Artificial Intelligence Approach to Guiding the Management of Heart Failure Patients Using Predictive Models: A Systematic Review

Heart failure (HF) is one of the leading causes of mortality and hospitalization worldwide. The accurate prediction of mortality and readmission risk provides crucial information for guiding decision making. Unfortunately, traditional predictive models reached modest accuracy in HF populations. We therefore aimed to present predictive models based on machine learning (ML) techniques in HF patients that were externally validated. We searched four databases and the reference lists of the included papers to identify studies in which HF patient data were used to create a predictive model. Literature screening was conducted in Academic Search Ultimate, ERIC, Health Source Nursing/Academic Edition and MEDLINE. The protocol of the current systematic review was registered in the PROSPERO database with the registration number CRD42022344855. We considered all types of outcomes: mortality, rehospitalization, response to treatment and medication adherence. The area under the receiver operating characteristic curve (AUC) was used as the comparator parameter. The literature search yielded 1649 studies, of which 9 were included in the final analysis. The AUCs for the machine learning models ranged from 0.6494 to 0.913 in independent datasets, whereas the AUCs for statistical predictive scores ranged from 0.622 to 0.806. Our study showed an increasing number of ML predictive models concerning HF populations, although external validation remains infrequent. However, our findings revealed that ML approaches can outperform conventional risk scores and may play important role in HF management.

Contemporary Applications of Machine Learning for Device Therapy in Heart Failure

Despite a better understanding of the underlying pathogenesis of heart failure (HF), pharmacotherapy, surgical, and percutaneous interventions do not prevent disease progression in all patients, and a significant proportion of patients end up requiring advanced therapies. Machine learning (ML) is gaining wider acceptance in cardiovascular medicine because of its ability to incorporate large, complex, and multidimensional data and to potentially facilitate the creation of predictive models not constrained by many of the limitations of traditional statistical approaches. With the coexistence of "big data" and novel advanced analytic techniques using ML, there is ever-increasing research into applying ML in the context of HF with the goal of improving patient outcomes. Through this review, the authors describe the basics of ML and summarize the existing published reports regarding contemporary applications of ML in device therapy for HF while highlighting the limitations to widespread implementation and its future promises.

Development and Validation of a Machine Learning Score for Readmissions After Transcatheter Aortic Valve Implantation

Background

Identifying predictors of readmissions after transcatheter aortic valve implantation (TAVI) is an important unmet need.

Objectives

We sought to explore the role of machine learning (ML) in predicting readmissions after TAVI.

Methods

We included patients who underwent TAVI between 2016 and 2019 in the Nationwide Readmission Database. A total of 917 candidate predictors representing all International Classification of Diseases, Tenth Revision, diagnosis and procedure codes were included. First, we used lasso regression to remove noninformative variables and rank informative ones. Next, we used an unsupervised ML model (K-means) to identify patterns/clusters in the data. Furthermore, we used Light Gradient Boosting Machine and Shapley Additive exPlanations to specify the impact of individual predictors. Finally, we built a parsimonious model to predict 30-day readmission.

Results

A total of 117,398 and 93,800 index TAVI hospitalizations were included in the 30- and 90-day analyses, respectively. Lasso regression identified 138 and 199 informative predictors for the 30- and 90-day readmission, respectively. Next, K-means recognized 2 distinct clusters: low risk and high risk. In the 30-day cohort, the readmission rate was 10.1% in the low risk group and 23.3% in the high risk group. In the 90-day cohort, the rates were 17.4% and 35.3%, respectively. The top predictors were the length of stay, frailty score, total discharge diagnoses, acute kidney injury, and Elixhauser score. These predictors were incorporated into a risk score (TAVI readmission score), which exhibited good performance in an external validation cohort (area under the curve 0.74 [0.7-0.78]).

Conclusions

ML methods can leverage widely available administrative databases to identify patients at risk for readmission after TAVI, which could inform and improve post-TAVI care.

Machine learning-based clustering in cervical spondylotic myelopathy patients to identify heterogeneous clinical characteristics

Background

Anterior cervical decompression and fusion can effectively treat cervical spondylotic myelopathy (CSM). Accurately classifying patients with CSM who have undergone anterior cervical decompression and fusion is the premise of precision medicine. In this study, we used machine learning algorithms to classify patients and compare the postoperative efficacy of each classification.

Methods

A total of 616 patients with cervical spondylotic myelopathy who underwent anterior cervical decompression and fusion were enrolled. Unsupervised machine learning algorithms (UMLAs) were used to cluster subjects according to similar clinical characteristics. Then, the results of clustering were visualized. The surgical outcomes were used to verify the accuracy of machine learning clustering.

Results

We identified two clusters in these patients who had significantly different baseline clinical characteristics, preoperative complications, the severity of neurological symptoms, and the range of decompression required for surgery. UMLA divided the CSM patients into two clusters according to the severity of their illness. The repose to surgical treatment between the clusters was significantly different.

Conclusions

Our results showed that UMLA could be used to rationally classify a heterogeneous cohort of CSM patients effectively, and thus, it might be used as the basis for a data-driven platform for identifying the cluster of patients who can respond to a particular treatment method.

Clinical Phenotypes of Heart Failure across the spectrum of Ejection Fraction: A Cluster Analysis

INTRODUCTION

Heart failure (HF), and especially HF with preserved ejection fraction (HFpEF), remains a challenging condition to define. The heterogenous nature of this population may be related to a variety of underlying etiologies interacting myocardial dysfunction.

METHOD

Alberta HEART study was a prospective, observational cohort that enrolled participants along the spectrum of heart failure including: healthy controls, people at risk of HF, and patients with HF and preserved (HFpEF) or reduced ejection fraction (HFrEF). We aimed to explore phenotypes of patients with HF and at-risk of developing HF. Utilising 27 detailed clinical, echocardiographic and biomarker variables, latent class analysis with and without multiple imputation was undertaken to identify distinct clinical phenotypes.

RESULTS

Of 621 participants, 191 (30.8%) and 169 (27.2%) were adjudicated by cardiologists to have HFpEF and HFrEF respectively. In the overall cohort, latent class analysis identified four distinct phenotypes. Phenotype A (n=152, 24.5%) was a healthy and low risk group. Phenotype B (n=129, 20.8%) demonstrated increased left ventricular mass and end-diastolic volumes, with elevated natriuretic peptides and clinical features of congestion. Phenotype C (n=128, 20.6%) was primarily characterised by obesity (80%) and normal indexed cardiac chamber sizes, low natriuretic peptide levels and minimal features of congestion. Phenotype D (n=212, 34.1%) consisted of elderly patients with clinical features of congestions. Phenotypes B and D demonstrated the highest risk of mortality and hospitalization over a median follow-up of 3.7 years.

CONCLUSION

Phenotypes with congestive features demonstrated increased risk profiles. Heart failure is a heterogenous classification which requires further work to appropriately categorise patients based on the underlying etiology or mechanism of impairment.

Identifying novel subgroups in heart failure patients with unsupervised machine learning: A scoping review

Background

Heart failure is currently divided into three main forms, HFrEF, HFpEF, and HFmrEF, but its etiology is diverse and highly heterogeneous. Many studies reported a variety of novel subgroups in heart failure patients, with unsupervised machine learning methods. The aim of this scoping review is to provide insights into how these techniques can diagnose and manage HF faster and better, thus providing direction for future research and facilitating its routine use in clinical practice.

Methods

The review was performed following PRISMA-SCR guideline. We searched the PubMed database for eligible publications. Studies were included if they defined new subgroups in HF patients using clustering analysis methods, and excluded if they are (1) Reviews, commentary, or editorials, (2) Studies not about defining new sub-types, or (3) Studies not using unsupervised algorithms. All study screening and data extraction were conducted independently by two investigators and narrative integration of data extracted from included studies was performed.

Results

Of the 498 studies identified, 47 were included in the analysis. Most studies (61.7%) were published in 2020 and later. The largest number of studies (46.8%) coming from the United States, and most of the studies were authored and included in the same country. The most commonly used machine learning method was hierarchical cluster analysis (46.8%), the most commonly used cluster variable type was comorbidity (61.7%), and the least used cluster variable type was genomics (12.8%). Most of the studies used data sets of less than 500 patients (48.9%), and the sample size had negative correlation with the number of clustering variables. The majority of studies (85.1%) assessed the association between cluster grouping and at least one outcomes, with death and hospitalization being the most commonly used outcome measures.

Conclusion

This scoping review provides an overview of recent studies proposing novel HF subgroups based on clustering analysis. Differences were found in study design, study population, clustering methods and variables, and outcomes of interests, and we provided insights into how these studies were conducted and identify the knowledge gaps to guide future research.

Correlation between Cardiac Ultrasound-Related Indicators and Cardiac Function in Patients with Coronary Heart Disease and Heart Failure

Objective

The purpose of this paper is to analyse the correlation between cardiac ultrasound-related indicators and cardiac function in patients with coronary heart disease and heart failure.

Methods

In this experiment, a total of 160 patients with coronary heart disease and heart failure who were diagnosed and treated in our hospital from June 2019 to March 2021 were recruited as the study group. All were examined by colour Doppler ultrasound instrument, SPSS statistical software was used to analyse the data obtained, and Spearman correlation was used to analyse the correlation between cardiac ultrasound-related indicators and cardiac function in patients with coronary heart disease and heart failure.

Results

In the study group, there were 68 patients with grade II cardiac function, accounting for 42.50%; 74 patients with grade III, accounting for 46.25%; and 18 patients with grade IV, accounting for 11.25%. The ultrasound parameters of the patients in the study group were profiled and calculated, and then statistically analysed with cardiac function grading. Cardiac function classification was significantly positively correlated with LVMI, LAD, and LVEDd (r = 0.689/0.915/0.928, P=0.001) and significantly negatively correlated with CI, LVFS, and LVEF (r = −0.689/−0.878/−0.912), P=0.001).

Conclusion

Cardiac ultrasound-related indicators are associated with patients with coronary heart disease and heart failure. With the decline of cardiac function in patients with coronary heart disease and heart failure, the patient's condition is aggravated. Therefore, cardiac ultrasound-related indicators play a major role in the diagnosis of clinical disease progression.

A multimodal deep learning model for cardiac resynchronisation therapy response prediction

We present a novel multimodal deep learning framework for cardiac resynchronisation therapy (CRT) response prediction from 2D echocardiography and cardiac magnetic resonance (CMR) data. The proposed method first uses the 'nnU-Net' segmentation model to extract segmentations of the heart over the full cardiac cycle from the two modalities. Next, a multimodal deep learning classifier is used for CRT response prediction, which combines the latent spaces of the segmentation models of the two modalities. At test time, this framework can be used with 2D echocardiography data only, whilst taking advantage of the implicit relationship between CMR and echocardiography features learnt from the model. We evaluate our pipeline on a cohort of 50 CRT patients for whom paired echocardiography/CMR data were available, and results show that the proposed multimodal classifier results in a statistically significant improvement in accuracy compared to the baseline approach that uses only 2D echocardiography data. The combination of multimodal data enables CRT response to be predicted with 77.38% accuracy (83.33% sensitivity and 71.43% specificity), which is comparable with the current state-of-the-art in machine learning-based CRT response prediction. Our work represents the first multimodal deep learning approach for CRT response prediction.

The Role of Cardiac Imaging in Heart Failure with Reduced Ejection Fraction

Heart failure (HF) is a major health burden associated with significant morbidity and mortality. Approximately half of all HF patients have reduced ejection fraction (left ventricular ejection fraction <40%) at rest (HF with reduced ejection fraction). The aetiology of HF is complex, and encompasses a wide range of cardiac conditions, hereditary defects and systemic diseases. Early identification of aetiology is important to allow personalised treatment and prognostication. Cardiac imaging has a major role in the assessment of patients with HF with reduced ejection fraction, and typically incorporates multiple imaging modalities, each with unique but complimentary roles. In this review, the comprehensive role of cardiac imaging in the diagnosis, assessment of aetiology, treatment planning and prognostication of HF with reduced ejection fraction is discussed.