Feature Selection based on F-score for Enhancing CTG Data Classification

Sevani, Nina; Hermawan, Indra; Jatmiko, Wisnu

doi:10.1109/cyberneticscom.2019.8875656

Cited by 12 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of author knowledge, most of the classi cation model studies have been carried out on the UCI machine learning repository CTG dataset [29], [30]. Thus, there were no studies addressing the derived dataset with the ve machine learning techniques.…”

Section: Results Analysismentioning

confidence: 99%

Intelligent SMOTE Based Machine learning Classification for Fetal State on Cardiotocography Dataset

Dutta¹,

Paul²,

Majumder³

2021

Preprint

View full text Add to dashboard Cite

A major contributor to under-five mortality is the death of children in the 1st month of life. Intrapartum complications are one of the major causes of perinatal mortality. Fetal cardiotocograph (CTGs) can be used as a monitoring tool to identify high-risk women during labor. The objective of this study was to study the precision of machine learning algorithm techniques on CTG data in identifying high-risk fetuses. CTG data of 2126 pregnant women were obtained from the University of California Irvine Machine Learning Repository. Out of 2126 CTG dataset 78% of them were normal, 14% were suspect, and 8 % had a pathological fetal state. To improve data imbalance SMOTE is applied followed by five different machine learning classification models were trained using CTG data. Sensitivity, precision, and F1 score for each class and overall accuracy of each model were obtained to predict normal, suspect, and pathological fetal states. For the model validity two statistical parameters MCC & kappa (k) are used. SMOTE based all the classification algorithm provides the higher degree of accuracy with minimum value is 96% and RF algorithm had the highest prediction accuracy about 98.01% which is quite satisfactory. Model validation statistical parameters MCC & kappa is maximum achieved by RF about 0.968 & 1 and for SVC is 0.977 & 1 respectively. Finally proposed work also compared with previous state of art techniques.

show abstract

Section: Results Analysismentioning

confidence: 99%

Intelligent SMOTE Based Machine learning Classification for Fetal State on Cardiotocography Dataset

Dutta¹,

Paul²,

Majumder³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The reported results are approaching 100% precision and 100% recall on both datasets, which confirms the stability of clustering algorithm. We also underline the importance of measuring the recall and precision at the same time using the F-score [27], as shown on Figure 4(a). Obviously, the improved BIRCH achieves higher F-scores than the basic BIRCH.…”

Section: Precision and Recall Evaluationmentioning

confidence: 93%

Automatic BIRCH thresholding with features transformation for hierarchical breast cancer clustering

Alzu’bi

Barham²

2022

IJECE

View full text Add to dashboard Cite

<p>Breast cancer is one of the most common diseases diagnosed in women over the world. The balanced iterative reducing and clustering using hierarchies (BIRCH) has been widely used in many applications. However, clustering the patient records and selecting an optimal threshold for the hierarchical clusters still a challenging task. In addition, the existing BIRCH is sensitive to the order of data records and influenced by many numerical and functional parameters. Therefore, this paper proposes a unique BIRCH-based algorithm for breast cancer clustering. We aim at transforming the medical records using the breast screening features into sub-clusters to group the subject cases into malignant or benign clusters. The basic BIRCH clustering is firstly fed by a set of normalized features then we automate the threshold initialization to enhance the tree-based sub-clustering procedure. Additionally, we present a thorough analysis on the performance impact of tuning BIRCH with various relevant linkage functions and similarity measures. Two datasets of the standard breast cancer wisconsin (BCW) benchmarking collection are used to evaluate our algorithm. The experimental results show a clustering accuracy of 97.7% in 0.0004 seconds only, thereby confirming the efficiency of the proposed method in clustering the patient records and making timely decisions.</p>

show abstract

“…Within each combination, including all study participants belonging to the corresponding study group pair, the average values were normalized using z-score [25]. Then for each combination, including all study participants belonging to the corresponding study group pair, the normalized average values of all features were applied as inputs to a cascade of one F-score algorithm [28] and one classical machine learning (cML) algorithm to select, according to the maximum classification accuracy, the best set of features. The F-score algorithm individually assesses and rates the features based on their F-score.…”

Section: Features Selection and Classificationmentioning

confidence: 99%

Structural MRI Texture Analysis for Detecting Alzheimer’s Disease

2023

View full text Add to dashboard Cite

Purpose: Alzheimer’s disease (AD) has the highest worldwide prevalence of all neurodegenerative disorders, no cure, and low ratios of diagnosis accuracy at its early stage where treatments have some effect and can give some years of life quality to patients. This work aims to develop an automatic method to detect AD in 3 different stages, namely, control (CN), mild-cognitive impairment (MCI), and AD itself, using structural magnetic resonance imaging (sMRI). Methods: A set of co-occurrence matrix and texture statistical measures (contrast, correlation, energy, homogeneity, entropy, variance, and standard deviation) were extracted from a two-level discrete wavelet transform decomposition of sMRI images. The discriminant capacity of the measures was analyzed and the most discriminant ones were selected to be used as features for feeding classical machine learning (cML) algorithms and a convolution neural network (CNN). Results: The cML algorithms achieved the following classification accuracies: 93.3% for AD vs CN, 87.7% for AD vs MCI, 88.2% for CN vs MCI, and 75.3% for All vs All. The CNN achieved the following classification accuracies: 82.2% for AD vs CN, 75.4% for AD vs MCI, 83.8% for CN vs MCI, and 64% for All vs All. Conclusion: In the evaluated cases, cML provided higher discrimination results than CNN. For the All vs All comparison, the proposed method surpasses by 4% the discrimination accuracy of the state-of-the-art methods that use structural MRI.

show abstract

Feature Selection based on F-score for Enhancing CTG Data Classification

Cited by 12 publications

References 12 publications

Intelligent SMOTE Based Machine learning Classification for Fetal State on Cardiotocography Dataset

Intelligent SMOTE Based Machine learning Classification for Fetal State on Cardiotocography Dataset

Automatic BIRCH thresholding with features transformation for hierarchical breast cancer clustering

Structural MRI Texture Analysis for Detecting Alzheimer’s Disease

Contact Info

Product

Resources

About