Early Diagnosis and Classification of Fetal Health Status from a Fetal Cardiotocography Dataset Using Ensemble Learning

Kuzu, Adem; SANTUR, Yunus

doi:10.3390/diagnostics13152471

Cited by 12 publications

(5 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The importance of ensemble learning techniques is further highlighted in a research paper concluded by [11], which achieved remarkable accuracy levels above 99.5%. This approach significantly improves prediction performance, demonstrating the potential of combining multiple classifiers in fetal health diagnostics.…”

Section: B Literature Reviewmentioning

confidence: 99%

Comprehensive Classification of Fetal Health Using Cardiotocogram Data Based on Machine Learning

Alkurdi,

Abdulazeez

2024

ijcs

View full text Add to dashboard Cite

In the realm of obstetrics, the evaluation of fetal health remains a paramount yet challenging endeavor. Traditional approaches, such as electronic fetal monitoring (EFM), despite their widespread adoption, continue to grapple with uncertainties regarding their impact on neonatal outcomes and the reduction of emergency cesarean deliveries. This ambiguity is compounded by a prevailing confusion within the obstetric community about interpreting fetal heart rate patterns, often leading to inconsistent and subjective assessments. Addressing these complexities, our study presents an innovative machine learning-based techniques for the comprehensive classification of fetal health using cardiotocogram (CTG) data, offering a more objective and nuanced alternative to conventional methods. The core of our proposed solution is a novel model employing a sophisticated ensemble of machine learning classifiers, including Multi-Support Vector Machine (Multi-SVM), Decision Tree, Random Forest with Hyperparameter Tuning, XGBoost, and Neural Networks. This model is unique in its application, processing datasets in four different forms: raw datasets, datasets processed with MinMaxScaler, datasets subjected to feature selection using SelectKBest, and a combination of MinMaxScaler processing and SelectKBest feature selection. Such meticulous preprocessing, encompassing normalization and feature selection, is pivotal in ensuring equitable contribution from each feature, thereby optimizing the model's learning process and predictive accuracy. The effectiveness of our model is rigorously evaluated using a dataset comprising 2126 individual records from CTG exams, classified by specialist obstetricians into three types: Normal, Suspect, and Pathological. These records are exhaustively analyzed using various metrics, including Accuracy, Precision, Recall, F1-Score, ROC AUC, and Confusion Matrix. Among the classifiers, XGBoost emerged as the most proficient, consistently outperforming others across multiple metrics. This indicates its superior ability to accurately identify and categorize the different states of fetal health. Our findings thus underscore the significant promise of machine learning in revolutionizing fetal health monitoring, offering a more reliable, objective, and comprehensive method for assessing fetal well-being, with profound implications for prenatal care and clinical decision-making.

show abstract

Section: B Literature Reviewmentioning

confidence: 99%

Comprehensive Classification of Fetal Health Using Cardiotocogram Data Based on Machine Learning

Alkurdi,

Abdulazeez

2024

ijcs

View full text Add to dashboard Cite

show abstract

“…Equation (1) below succinctly illustrates the precise methods used in the normalization and standardization process. In this equation, '𝑥' represents the original input values, and '𝑥′' denotes the transformed values, reflecting the enhanced, range-bound, and standardized data [17].…”

Section: Data Preprocessingmentioning

confidence: 99%

An Explainable Artificial Intelligence Based Early Lung Cancer Risk Prediction Using LightGBM

2023

IJANSER

View full text Add to dashboard Cite

Lung cancer is the most common cancer in the world and is also the deadliest. Early diagnosis can improve patients' life expectancy and reduce the cost of treatment. The aim of this study is to help physicians and patients by using explainable artificial intelligence methods to early diagnose risk of lung cancer. This study will create an opportunity for physicians to make early diagnosis and treatment strategies for patients. In this study, a machine learning model was developed to predict the risk of lung cancer by using the LightGBM algorithm. Furthermore, the SHAP method is used to explain why and how the model's predictions are made, thus making the AI model reliable. These explanations increase the acceptability and reliability of the predictions made by model, while helping physicians and patients understand the model's decisions. The results obtained show that the developed LightGBM model can predict the risk of lung cancer with a 100% accuracy rate. The model has achieved great results in terms of accuracy rate and sensitivity. In addition, the SHAP analyses explain which features each of the model's predictions is based on, which will increase the confidence of physicians and patients in the decisions of artificial intelligence.

show abstract

“…Kuzu et al proposed a batch learning-based prediction method for the classification of fetal health (normal, suspicious, pathological) using fetal heart rate acceleration obtained from a cardiotocography (CTG) dataset and NST tests in their study (22). In this research, they developed binary and multiclass classification models using LR, RF, and extreme gradient boosting (XGBoost) algorithms, and applied these models to the UCI dataset.…”

Section: Literature Reviewmentioning

confidence: 99%

“…It was observed that 22 features were used in two studies and 23 features were used in two studies using cross validation (CV) along with accuracy, precision, recall and F 1 score metrics. Kuzu et al achieved 100% accuracy as a result of their experiments using the polynomial expansion (PE) function for feature extraction along with CV, entropy and normalisation (22). Das et al on the other hand, achieved 99.91% accuracy by using standard dimensionality reduction algorithms such as PCA, correlation-based feature subset selection, Chi-squared feature selection, MinMax, and CV methods (64).…”

Section: Artificial Intelligence In Fetal Healthmentioning

confidence: 99%

Artificial Intelligence in Fetal Health Diagnosis: A Systematic Literature Review

Kuzu,

Santur

2023

TUSEB

View full text Add to dashboard Cite

Obstetricians commonly use Non-Stress Test to evaluate fetal well-being during the antepartum and intrapartum periods by measuring fetal heart rate and uterine contractions of the mother. Non-Stress Test is also used to diagnose fetal distress at an early stage. Early diagnosis and treatment can increase the fetus's survival rate and improve their quality of life. The fetal heart rate and uterine contraction signals obtained from the Non-Stress Test are recorded on a paper called a trace. Obstetricians interpret the trace to make decisions about the fetus's condition. However, traditional analysis of Non-Stress Test takes time, and there are differences in interpretation among experts. Newly qualified doctors and midwives are more prone to making mistakes and incorrect decisions. To overcome the differences in the interpretation of Non-Stress Test analysis and to automate the process to minimize diagnostic errors, machine learning and deep learning models have been increasingly used in recent years. In this study, the literature of the past five years is researched, and numerical expressions, tables, and graphs related to this are presented.

show abstract

Early Diagnosis and Classification of Fetal Health Status from a Fetal Cardiotocography Dataset Using Ensemble Learning

Cited by 12 publications

References 38 publications

Comprehensive Classification of Fetal Health Using Cardiotocogram Data Based on Machine Learning

Comprehensive Classification of Fetal Health Using Cardiotocogram Data Based on Machine Learning

An Explainable Artificial Intelligence Based Early Lung Cancer Risk Prediction Using LightGBM

Artificial Intelligence in Fetal Health Diagnosis: A Systematic Literature Review

Contact Info

Product

Resources

About