Machine Learning-Based Monitoring and Prognosis of Chronic Kidney Disease Patients DOI

Sandeep P. Abhang,

Manoj Tarambale,

Ali Esnaashariyeh

et al.

Published: Dec. 29, 2023

The study is aimed at examining the use of ML techniques for detecting and forecasting CKD progression. Experimental models were constructed utilizing whole datasets including demographic information, medical history, laboratory findings, clinical notes; then, they assessed. algorithms embedded are decision trees, random forests, support vector machines, gradient boosting. experimental results showed clear evidence good predictive accuracy across all algorithms, with boosting achieving highest 90 percent. Besides that, precision, recall, F1-score, area under receiver operating characteristic curve (AUC-ROC) assessed, having values from 0.82 to 0.92. performance proposed ensemble method which optimizes both two time-varying Markov chains among related works superior other methods. studied variables, such as serum creatinine, glomerular filtration rate, age, blood pressure urine protein levels linked progression in variable importance analysis. This emphasizes ability ML-based methods providing patients' care by means early diagnosis, personalized medicine, improved (outcomes).

Language: Английский

PREDICTING THE PREVALENCE OF CARDIOVASCULAR DISEASES USING MACHINE LEARNING ALGORITHMS DOI Creative Commons

Bernada E Sianga,

Maurice C. Y. Mbago,

Amina S. Msengwa

et al.

Intelligence-Based Medicine, Journal Year: 2025, Volume and Issue: unknown, P. 100199 - 100199

Published: Jan. 1, 2025

Language: Английский

Citations

1
Emergence of Artificial Intelligence and Machine Learning Models in Sudden Cardiac Arrest: A Comprehensive Review of Predictive Performance and Clinical Decision Support DOI
Hritvik Jain, Mohammed Dheyaa Marsool Marsool, Ramez M. Odat

et al.

Cardiology in Review, Journal Year: 2024, Volume and Issue: unknown

Published: June 5, 2024

Sudden cardiac death/sudden arrest (SCD/SCA) is an increasingly prevalent cause of mortality globally, particularly in individuals with preexisting conditions. The ambiguous premortem warnings and the restricted interventional window related to SCD account for complexity condition. Current reports suggest be accountable 20% all deaths hence accurately predicting risk imminent concern. Traditional approaches SCA, “track-and-trigger” warning systems have demonstrated considerable inadequacies, including low sensitivity, false alarms, decreased diagnostic liability, reliance on clinician involvement, human errors. Artificial intelligence (AI) machine learning (ML) models near-perfect accuracy SCA risk, allowing clinicians intervene timely. Given constraints current diagnostics, exploring benefits AI ML enhancing outcomes SCA/SCD imperative. This review article aims investigate efficacy managing SCD, targeting prediction.

Language: Английский

Citations

5
A Novel Approach for Performance Evaluation and Effectiveness of Data-Driven Heart Disease Diagnosis DOI
Md Aminul Islam, Anindya Nag,

Ayontika Das

et al.

Lecture notes in electrical engineering, Journal Year: 2025, Volume and Issue: unknown, P. 231 - 243

Published: Jan. 1, 2025

Language: Английский

Citations

0
A Predictive Model of Cardiovascular Aging by Clinical and Immunological Markers Using Machine Learning DOI Creative Commons
Мадина Сулейменова, Kuat Аbzaliyev, Мадина Мансурова

et al.

Diagnostics, Journal Year: 2025, Volume and Issue: 15(7), P. 850 - 850

Published: March 27, 2025

Background/Objectives: Aging and immune mechanisms play a key role in the development of cardiovascular disease (CVD), especially context chronic inflammation. Therefore, order to detect early aging elderly, we have developed prognostic model based on clinical immunological markers using machine learning. Methods: This paper analyzes relationships between markers, parameters, lifestyle factors individuals over 60 years age. A learning (ML) including random forest, logistic regression, k-nearest neighbors, XGBoost was predict rate risk CVD. Correlation anal is revealed significant associations (CD14+, HLA-DR, IL-10, CD8+), parameters (BMI, coronary heart disease, hypertension, diabetes), behavioral (physical activity, smoking, alcohol). Results: The results study confirm that systemic inflammation, as reflected by such CD14+, plays central pathogenesis related diseases. CD14+ shows moderate positive correlation with post-infarction cardiosclerosis, accounting for 37%. HLA-DR correlates body mass index at 39%. negative association IL-10 level BMI also found, where reaches 52% (r = -0.52). CD8+ cells smoking their number, being 40%. Training performed data models were evaluated accuracy, ROC-AUC, F1-score metrics. Among all trained models, best, achieving an accuracy 91% area under ROC curve (AUC) 0.8333. Conclusions: reveals correlations which allows assessment individual risks premature aging. R (version 4.3.0) specialized libraries matrix construction visualization used analysis, Python 3.11.11) training.

Language: Английский

Citations

0
Investigation of machine learning algorithms on heart disease through dominant feature detection and feature selection DOI
F. J. Turk

Signal Image and Video Processing, Journal Year: 2024, Volume and Issue: 18(4), P. 3943 - 3955

Published: March 4, 2024

Language: Английский

Citations

3
Medical diagnosis based on artificial intelligence and decision support system in the management of health development DOI

Kaipeng Chen,

Lizhuo Luo, Ye Tan

et al.

Journal of Evaluation in Clinical Practice, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 21, 2024

Abstract Background Medical diagnosis plays a critical role in our daily lives. Every day, over 10 billion cases of both mental and physical health disorders are diagnosed reported worldwide. To diagnose these disorders, medical practitioners professionals employ various assessment tools. However, tools often face scrutiny due to their complexity, prompting researchers increase experimental parameters provide accurate justifications. Additionally, it is essential for properly justify, interpret, analyse the results from prediction Methods This research paper explores use artificial intelligence advanced analytics developing Clinical Decision Support Systems (CDSS). These systems capable diagnosing detecting patterns disorders. Various machine learning algorithms contribute building tools, with Network Pattern Recognition (NEPAR) algorithm being first aid CDSS. Over time, have recognised value learning‐based models successfully justifying diagnoses. Results The proposed CDSS demonstrated ability an accuracy up 89% using only 28 questions, without requiring human input. For issues, additional used enhance models. Conclusions Consequently, increasingly relying on models, utilising improve assist decision‐making. different cross‐validation values considered remove data biasness.

Language: Английский

Citations

2
The Comparative Early Prediction Model for Cardiovascular Disease Using Machine Learning DOI Open Access

Sri Sumarlinda,

Azizah Rahmat,

Zalizah binti Awang Long

et al.

International Journal of Scientific Research in Computer Science Engineering and Information Technology, Journal Year: 2024, Volume and Issue: unknown, P. 24 - 33

Published: Jan. 1, 2024

Cardiovascular disease (CVD) is a leading cause of death and major contributor to disability. Early detection cardiovascular using ANFIS has the potential reduce costs simplify treatment. This study aims develop prediction model (Adaptive Neuro-Fuzzy Inference System) for early disease. The dataset used consists 500 data with 12 features, including various risk factors such as blood sugar levels, cholesterol, uric acid, systolic pressure, diastolic body mass index (BMI), age, smoking habits, lifestyle, genetic factors, gender, one label feature. compares models machine learning methods, namely Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), ANFIS. development KNN algorithm involves value K=5 Euclidian distance measure. SVM kernel cache 200 convergence epsilon 0.001. was built sets divided into training (70%) testing (30%) data, rate variations 0.01, 0.05, 0.1, 0.2, 0.5. results show SVM, accuracy 0.760, precision 0.839, recall 0.671. For model, 0.758, 0.768, 0.771. As reaches 0.989, 0.996, 0.988. highest performance. Further shows that 0.1 provides most optimal

Language: Английский

Citations

0
A Hybrid Transfer Learning Approach Using Obesity Data for Predicting Cardiovascular Diseases Incorporating Lifestyle Factors DOI Creative Commons
Krishna Modi, Ishbir Singh, Yogesh Kumar

et al.

International Journal of experimental research and review, Journal Year: 2024, Volume and Issue: 46, P. 1 - 18

Published: Dec. 30, 2024

Cardiovascular Diseases (CVDs), particularly heart diseases, are becoming a significant global public health concern. This study enhances CVD detection through novel approach that integrates obesity prediction using machine learning (ML) models. Specifically, model trained on an dataset was used to add 'Obesity level' feature the disease dataset, leveraging relation of high with increased risk. We have also calculated BMI and added as in dataset. evaluated this transfer learning-based alongside eight ML Performance these models assessed precision, recall, accuracy F1-score metrics. Our research aims provide healthcare practitioners reliable tools for early diagnosis. Results indicate ensemble methods, which combine strengths multiple models, significantly improve compared other classifiers. able achieve 74% score along 0.72 F1 score, 0.77 precision 0.80 AUC XGBoost classifier, followed closely by DNN 73.7% 0.75 0.798 our proposed model. seek enhance efficiency promote integrating AI-based solutions into medical practice. The findings demonstrate potential techniques effectiveness incorporating obesity-related features optimized cardiovascular detection.

Language: Английский

Citations

0
Machine Learning-Based Monitoring and Prognosis of Chronic Kidney Disease Patients DOI

Sandeep P. Abhang,

Manoj Tarambale,

Ali Esnaashariyeh

et al.

Published: Dec. 29, 2023

The study is aimed at examining the use of ML techniques for detecting and forecasting CKD progression. Experimental models were constructed utilizing whole datasets including demographic information, medical history, laboratory findings, clinical notes; then, they assessed. algorithms embedded are decision trees, random forests, support vector machines, gradient boosting. experimental results showed clear evidence good predictive accuracy across all algorithms, with boosting achieving highest 90 percent. Besides that, precision, recall, F1-score, area under receiver operating characteristic curve (AUC-ROC) assessed, having values from 0.82 to 0.92. performance proposed ensemble method which optimizes both two time-varying Markov chains among related works superior other methods. studied variables, such as serum creatinine, glomerular filtration rate, age, blood pressure urine protein levels linked progression in variable importance analysis. This emphasizes ability ML-based methods providing patients' care by means early diagnosis, personalized medicine, improved (outcomes).

Language: Английский

Citations

0