Data preprocessing for heart disease classification: A systematic literature review

Unraveling the role of the FHL family in cardiac diseases: Mechanisms, implications, and future directions

Heart diseases are a major cause of morbidity and mortality worldwide. Understanding the molecular mechanisms underlying these diseases is essential for the development of effective diagnostic and therapeutic strategies. The FHL family consists of five members: FHL1, FHL2, FHL3, FHL4, and FHL5/Act. These members exhibit different expression patterns in various tissues including the heart. FHL family proteins are implicated in cardiac remodeling, regulation of metabolic enzymes, and cardiac biomechanical stress perception. A large number of studies have explored the link between FHL family proteins and cardiac disease, skeletal muscle disease, and ovarian metabolism, but a comprehensive and in-depth understanding of the specific molecular mechanisms targeting FHL on cardiac disease is lacking. The aim of this review is to explore the structure and function of FHL family members, to comprehensively elucidate the mechanisms by which they regulate the heart, and to explore in depth the changes in FHL family members observed in different cardiac disorders, as well as the effects of mutations in FHL proteins on heart health.

Machine-Learning-Based Prediction of 1-Year Arrhythmia Recurrence after Ventricular Tachycardia Ablation in Patients with Structural Heart Disease

BACKGROUND

Ventricular tachycardia (VT) recurrence after catheter ablation remains a concern, emphasizing the need for precise risk assessment. We aimed to use machine learning (ML) to predict 1-month and 1-year VT recurrence following VT ablation.

METHODS

For 337 patients undergoing VT ablation, we collected 31 parameters including medical history, echocardiography, and procedural data. 17 relevant features were included in the ML-based feature selection, which yielded six and five optimal features for 1-month and 1-year recurrence, respectively. We trained several supervised machine learning models using 10-fold cross-validation for each endpoint.

RESULTS

We observed 1-month VT recurrence was observed in 60 (18%) cases and accurately predicted using our model with an area under the receiver operating curve (AUC) of 0.73. Input features used were hemodynamic instability, incessant VT, ICD shock, left ventricular ejection fraction, TAPSE, and non-inducibility of the clinical VT at the end of the procedure. A separate model was trained for 1-year VT recurrence (observed in 117 (35%) cases) with a mean AUC of 0.71. Selected features were hemodynamic instability, the number of inducible VT morphologies, left ventricular systolic diameter, mitral regurgitation, and ICD shock. For both endpoints, a random forest model displayed the highest performance.

CONCLUSIONS

Our ML models effectively predict VT recurrence post-ablation, aiding in identifying high-risk patients and tailoring follow-up strategies.

Epidemiological Determinants of Patient Non-Conveyance to the Hospital in an Emergency Medical Service Environment

BACKGROUND

The increasing prevalence of comorbidities worldwide has spurred the need for time-effective pre-hospital emergency medical services (EMS). Some pre-hospital emergency calls requesting EMS result in patient non-conveyance. Decisions for non-conveyance are sometimes driven by the patient or the clinician, which may jeopardize the patients' healthcare outcomes. This study aimed to explore the distribution and determinants of patient non-conveyance to hospitals in a Middle Eastern national Ambulance Service that promotes the transportation of all emergency call patients and does not adopt clinician-based non-conveyance decision.

METHODS

Using R Language, descriptive, bivariate, and binary logistic regression analyses were conducted for 334,392 multi-national patient non-conveyance emergency calls from June 2018 to July 2022, from a total of 1,030,228 calls to which a response unit was dispatched.

RESULTS

After data pre-processing, 237,862 cases of patient non-conveyance to hospital were retained, with a monthly average of 41.96% (n = 8799) of the emergency service demands and a standard deviation of 5.49% (n = 2040.63). They predominantly involved South Asians (29.36%, n = 69,849); 64.50% (n = 153,427) were of the age category from 14 to 44 years; 61.22% (n = 145,610) were male; 74.59% (n = 177,424) from the urban setting; and 71.28% (n = 169,552) had received on-scene treatment. Binary logistic regression with full variables and backward methods identified the final models of the determinants of patient non-conveyance decisions with an Akaike information criterion prediction estimator, respectively, of (250,200) and (250,169), indicating no significant difference between both models (Chi-square test; p-value = 0.63).

CONCLUSIONS

Despite exercising a cautious protocol by encouraging patient transportation to hospital, patient non-conveyance seems to be a problem in the healthcare system that strains the pre-hospital medical response teams' resources. Policies and regulations should be adopted to encourage individuals to access other primary care centers when required rather than draining emergency services for non-emergency situations.

Securing health care data through blockchain enabled collaborative machine learning

Transferring of data in machine learning from one party to another party is one of the issues that has been in existence since the development of technology. Health care data collection using machine learning techniques can lead to privacy issues which cause disturbances among the parties and reduces the possibility to work with either of the parties. Since centralized way of information transfer between two parties can be limited and risky as they are connected using machine learning, this factor motivated us to use the decentralized way where there is no connection but model transfer between both parties will be in process through a federated way. The purpose of this research is to investigate a model transfer between a user and the client(s) in an organization using federated learning techniques and reward the client(s) for their efforts with tokens accordingly using blockchain technology. In this research, the user shares a model to organizations that are willing to volunteer their service to provide help to the user. The model is trained and transferred among the user and the clients in the organizations in a privacy preserving way. In this research, we found that the process of model transfer between user and the volunteered organizations works completely fine with the help of federated learning techniques and the client(s) is/are rewarded with tokens for their efforts. We used the COVID-19 dataset to test the federation process, which yielded individual results of 88% for contributor a, 85% for contributor b, and 74% for contributor c. When using the FedAvg algorithm, we were able to achieve a total accuracy of 82%.

Data-Centric AI for Healthcare Fraud Detection

Automated methods for detecting fraudulent healthcare providers have the potential to save billions of dollars in healthcare costs and improve the overall quality of patient care. This study presents a data-centric approach to improve healthcare fraud classification performance and reliability using Medicare claims data. Publicly available data from the Centers for Medicare & Medicaid Services (CMS) are used to construct nine large-scale labeled data sets for supervised learning. First, we leverage CMS data to curate the 2013-2019 Part B, Part D, and Durable Medical Equipment, Prosthetics, Orthotics, and Supplies (DMEPOS) Medicare fraud classification data sets. We provide a review of each data set and data preparation techniques to create Medicare data sets for supervised learning and we propose an improved data labeling process. Next, we enrich the original Medicare fraud data sets with up to 58 new provider summary features. Finally, we address a common model evaluation pitfall and propose an adjusted cross-validation technique that mitigates target leakage to provide reliable evaluation results. Each data set is evaluated on the Medicare fraud classification task using extreme gradient boosting and random forest learners, multiple complementary performance metrics, and 95% confidence intervals. Results show that the new enriched data sets consistently outperform the original Medicare data sets that are currently used in related works. Our results encourage the data-centric machine learning workflow and provide a strong foundation for data understanding and preparation techniques for machine learning applications in healthcare fraud.

A New ECG Data Processing Approach to Developing an Accurate Driving Fatigue Detection Framework with Heart Rate Variability Analysis and Ensemble Learning

More than 1.3 million people are killed in traffic accidents annually. Road traffic accidents are mostly caused by human error. Therefore, an accurate driving fatigue detection system is required for drivers. Most driving fatigue detection studies concentrated on improving feature engineering and classification methods. We propose a novel driving fatigue detection framework concentrating on the development of the preprocessing, feature extraction, and classification stages to improve the classification accuracy of fatigue states. The proposed driving fatigue detection framework measures fatigue using a two-electrode ECG. The resampling method and heart rate variability analysis were used to extract features from the ECG data, and an ensemble learning model was utilized to classify fatigue states. To achieve the best model performance, 40 possible scenarios were applied: a combination of 5 resampling scenarios, 2 feature extraction scenarios, and 4 classification model scenarios. It was discovered that the combination of a resampling method with a window duration of 300 s and an overlap of 270 s, 54 extracted features, and AdaBoost yielded an optimum accuracy of 98.82% for the training dataset and 81.82% for the testing dataset. Furthermore, the preprocessing resampling method had the greatest impact on the model’s performance; it is a new approach presented in this study.

Multi-modality data-driven analysis of diagnosis and treatment of psoriatic arthritis

Psoriatic arthritis (PsA) is associated with psoriasis, featured by its irreversible joint symptoms. Despite the significant impact on the healthcare system, it is still challenging to leverage machine learning or statistical models to predict PsA and its progression, or analyze drug efficacy. With 3961 patients' clinical records, we developed a machine learning model for PsA diagnosis and analysis of PsA progression risk, respectively. Furthermore, general additive models (GAMs) and the Kaplan-Meier (KM) method were applied to analyze the efficacy of various drugs on psoriasis treatment and inhibiting PsA progression. The independent experiment on the PsA prediction model demonstrates outstanding prediction performance with an AUC score of 0.87 and an AUPR score of 0.89, and the Jackknife validation test on the PsA progression prediction model also suggests the superior performance with an AUC score of 0.80 and an AUPR score of 0.83, respectively. We also identified that interleukin-17 inhibitors were the more effective drug for severe psoriasis compared to other drugs, and methotrexate had a lower effect in inhibiting PsA progression. The results demonstrate that machine learning and statistical approaches enable accurate early prediction of PsA and its progression, and analysis of drug efficacy.

Imbalanced data preprocessing techniques for machine learning: a systematic mapping study

Machine Learning (ML) algorithms have been increasingly replacing people in several application domains-in which the majority suffer from data imbalance. In order to solve this problem, published studies implement data preprocessing techniques, cost-sensitive and ensemble learning. These solutions reduce the naturally occurring bias towards the majority sample through ML. This study uses a systematic mapping methodology to assess 9927 papers related to sampling techniques for ML in imbalanced data applications from 7 digital libraries. A filtering process selected 35 representative papers from various domains, such as health, finance, and engineering. As a result of a thorough quantitative analysis of these papers, this study proposes two taxonomies-illustrating sampling techniques and ML models. The results indicate that oversampling and classical ML are the most common preprocessing techniques and models, respectively. However, solutions with neural networks and ensemble ML models have the best performance-with potentially better results through hybrid sampling techniques. Finally, none of the 35 works apply simulation-based synthetic oversampling, indicating a path for future preprocessing solutions.

Social Frailty and Meaningful Activities among Community-Dwelling Older Adults with Heart Disease

Patients with heart disease are more likely to experience social frailty due to physical inactivity, which may affect meaningful activities such as hobbies. This study aimed to investigate (1) the association between heart disease and social frailty in community-dwelling older adults and (2) the characteristics of meaningful activities in community-dwelling older adults with heart disease. Data from 630 older adults who participated in a community-based health survey were obtained, including clinical history, meaningful activities, social frailty and psychosomatic functions. Participants were divided into two groups: those with heart disease (n = 79) and those without (n = 551), and comparisons were made. Social frailty was observed in 23.7% of participants with heart disease, and logistic regression revealed significant associations with heart disease and social frailty after adjusting for potential covariates (OR, 1.97; 95% CI, 1.06 3.67; p = 0.032). Participants with heart disease did not differ significantly in terms of satisfaction or performance; their frequency of engagement in meaningful activities was significantly lower than without heart disease (p = 0.041). These results suggest that heart disease and social frailty are associated in community-dwelling older adults, and that this demographic is inclined to engage in meaningful activities less frequently.

E-Cardiac Care: A Comprehensive Systematic Literature Review

The Internet of Things (IoT) is a complete ecosystem encompassing various communication technologies, sensors, hardware, and software. IoT cutting-edge technologies and Artificial Intelligence (AI) have enhanced the traditional healthcare system considerably. The conventional healthcare system faces many challenges, including avoidable long wait times, high costs, a conventional method of payment, unnecessary long travel to medical centers, and mandatory periodic doctor visits. A Smart healthcare system, Internet of Things (IoT), and AI are arguably the best-suited tailor-made solutions for all the flaws related to traditional healthcare systems. The primary goal of this study is to determine the impact of IoT, AI, various communication technologies, sensor networks, and disease detection/diagnosis in Cardiac healthcare through a systematic analysis of scholarly articles. Hence, a total of 104 fundamental studies are analyzed for the research questions purposefully defined for this systematic study. The review results show that deep learning emerges as a promising technology along with the combination of IoT in the domain of E-Cardiac care with enhanced accuracy and real-time clinical monitoring. This study also pins down the key benefits and significant challenges for E-Cardiology in the domains of IoT and AI. It further identifies the gaps and future research directions related to E-Cardiology, monitoring various Cardiac parameters, and diagnosis patterns.

Data pre-processing for cardiovascular disease classification: A systematic literature review

The important task in the medical field is the early detection of disease. Heart disease is one of the greatest challenging diseases in all other diseases subsequently 17.3 million people died once a year due to heart disease. A minute error in heart disease diagnosis is a risk for an individual lifespan. Precise heart disease diagnosis is consequently critical. Different approaches including data mining have been used for the prediction of heart disease. However, there are some solemn concerns related to the data quality for example inconsistencies, missing values, noise, high dimensionality, and imbalanced statistics. In order to improve the accuracy of Data Mining based prediction systems, techniques for data preparation were applied to increase the quality of the data. The foremost objective of this paper is to highlight and summarize the research work about (i) data preparation techniques mostly used, (ii) the impact of pre-processing procedures on the accuracy of a heart disease prediction system, (iii) classifier enactment with data pre-processing techniques, (4) comparison in terms of accuracy of the different pre-processing model. A systematic literature review on the use of data pre-processing in heart disease diagnosis is carried out from January 2001 to July 2021 by studying the published material. Almost 30 studies were designated and examined related to the above-mentioned benchmarks. The literature review concludes that data reduction and data cleaning pre-processing techniques are mostly used in heart disease prediction systems. Overall this study concludes that data pre-processing has improved the accuracy of models used for heart disease prediction. Some hybrid models including (ANN+CHI), (ANN+PCA), (DNN+CHI) and (SVM+PCA) have shown improved accuracy level. However, due to the lack of clarification, there is a number of limitations and challenges in order to implementing these models in the real world.

Significance of Visible Non-Invasive Risk Attributes for the Initial Prediction of Heart Disease Using Different Machine Learning Techniques

Introduction

Heart disease is emerging as the single most critical cause of death worldwide and is one of the costliest chronic conditions.

Purpose

Stimulated by the increasing heart disease mortality rate incidents, an effective, low-cost, and reliable heart disease risk evaluation model is developed using significant non-invasive risk attributes. The significant non-invasive risk attributes like (age, systolic BP, diastolic BP, BMI, hereditary factor, smoking, alcohol, and physical inactivity) are identified by the help of medical domain experts, and their reliability in heart disease prediction is investigated through different feature selection techniques. Methodology. The enhancements of applying specific investigated techniques like random forest, Naïve Bayes, decision tree, support vector machine, and K nearest neighbor to the risk factors are tested. The heart disease risk assessment model is developed using the Jupyter Notebook web application, and its performance is tested not only through medical domain measures but also through the model performance measures. Findings. To evaluate heart disease risk evaluation model, we calculated measures of discrimination like error rate, AUROC, sensitivity, specificity, accuracy, precision, and so on. Experimental results show that the random forest heart disease risk evaluation model outperforms other existing risk models with admirable predictive accuracy and minimum misclassification rate. Originality. The heart disease risk evaluation model is developed based on novel non-invasive heart disease dataset, which consists of 5776 records. This dataset is collected from different heterogeneous data sources of Kashmir (India) through quantitative data collection methods. Research Implications. The risk model is applicable where people lack the facilities of integrated primary medical care technologies for untimely heart disease risk prediction. Future Work. To investigate deep learning and study the significance of other controlled attributes on different age and sex groups in the risk estimation of heart disease.

Bioinspired Self-Powered Piezoresistive Sensors for Simultaneous Monitoring of Human Health and Outdoor UV Light Intensity

The exact fabrication of precise three-dimensional structures for piezoresistive sensors necessitates superior manufacturing methods or tooling, which are accompanied by time-consuming processes and the potential for environmental harm. Herein, we demonstrated a method for in situ synthesis of zinc oxide nanorod (ZnO NR) arrays on graphene-treated cotton and paper substrates and constructed highly sensitive, flexible, wearable, and chemically stable strain sensors. Based on the structure of pine trees and needles in nature, the hybrid sensing layer consisted of graphene-attached cotton or paper fibers and ZnO NRs, and the results showed a high sensitivity of 0.389, 0.095, and 0.029 kPa^-1 and an ultra-wide linear range of 0-100 kPa of this sensor under optimal conditions. Our study found that water absorption and swelling of graphene fibers and the associated reduction of pore size and growth of zinc oxide were detrimental to pressure sensor performance. A random line model was developed to examine the effects of different hydrothermal times on sensor performance. Meanwhile, pulse detection, respiration detection, speech recognition, and motion detection, including finger movements, walking, and throat movements, were used to show their practical application in human health activity monitoring. In addition, monolithically grown ZnO NRs on graphene cotton sheets had been integrated into a flexible sensing platform for outdoor UV photo-indication, which is, to our knowledge, the first successful case of an integrated UV photo-detector and motion sensor. Due to its excellent strain detection and UV detection abilities, these strategies are a step forward in developing wearable sensors that are cost-controllable and high-performance.

Epigenetics of Mitochondria-Associated Genes in Striated Muscle

Striated muscle has especially large energy demands. We identified 97 genes preferentially expressed in skeletal muscle and heart, but not in aorta, and found significant enrichment for mitochondrial associations among them. We compared the epigenomic and transcriptomic profiles of the 27 genes associated with striated muscle and mitochondria. Many showed strong correlations between their tissue-specific transcription levels, and their tissue-specific promoter, enhancer, or open chromatin as well as their DNA hypomethylation. Their striated muscle-specific enhancer chromatin was inside, upstream, or downstream of the gene, throughout much of the gene as a super-enhancer (CKMT2, SLC25A4, and ACO2), or even overlapping a neighboring gene (COX6A2, COX7A1, and COQ10A). Surprisingly, the 3' end of the 1.38 Mb PRKN (PARK2) gene (involved in mitophagy and linked to juvenile Parkinson's disease) displayed skeletal muscle/myoblast-specific enhancer chromatin, a myoblast-specific antisense RNA, as well as brain-specific enhancer chromatin. We also found novel tissue-specific RNAs in brain and embryonic stem cells within PPARGC1A (PGC-1α), which encodes a master transcriptional coregulator for mitochondrial formation and metabolism. The tissue specificity of this gene's four alternative promoters, including a muscle-associated promoter, correlated with nearby enhancer chromatin and open chromatin. Our in-depth epigenetic examination of these genes revealed previously undescribed tissue-specific enhancer chromatin, intragenic promoters, regions of DNA hypomethylation, and intragenic noncoding RNAs that give new insights into transcription control for this medically important set of genes.

Functional Coronary Artery Assessment: a Systematic Literature Review

Cardiovascular diseases represent the number one cause of death in the world, including the most common disorders in the heart's health, namely coronary artery disease (CAD). CAD is mainly caused by fat accumulated in the arteries' internal walls, creating an atherosclerotic plaque that impacts the blood flow functional behavior. Anatomical plaque characteristics are essential but not sufficient for a complete functional assessment of CAD. In fact, plaque analysis and visual inspection alone have proven insufficient to determine the lesion severity and hemodynamic repercussion. Furthermore, the fractional flow reserve (FFR) exam, which is considered the gold standard for stenosis functional impair determination, is invasive and contains several limitations. Such a panorama evidences the need for new techniques applied to image exams to improve CAD functional assessment. In this article, we perform a systematic literature review on emerging methods determining CAD significance, thus delivering a unique base for comparing these methods, qualitatively and quantitatively. Our goal is to guide further studies with evidence from the most promising methods, highlighting the benefits from both areas. We summarize benchmarks, metrics for evaluation, and challenges already faced, thus shedding light on the requirements for a valid, meaningful, and accepted technique for functional assessment evaluation. We create a base of comparison based on quantitative and qualitative indicators and highlight the most relevant geometrical metrics that correlate with lesion significance. Finally, we point out future benchmarks based on recent literature.