Deep Analysis of EIT Dataset to Classify Apnea and Non-Apnea Cases in Neonatal Patients

Vahabi, Nafiseh; Yerworth, Rebecca J.; Miedema, Martijn; Kaam, Anton van; Bayford, Richard; Demosthenous, Andreas

doi:10.1109/access.2021.3056558

Cited by 8 publications

(3 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An example of a classifier model which is built on limited training images was reported by [28]. Nevertheless, this dataset will be a valuable framework in developing tomographic sensing interpretation using ML [29] [30].…”

Section: Machine Learning Methods Using Mutual Induction Datamentioning

confidence: 99%

Interior Void Classification in Liquid Metal Using Multi-Frequency Magnetic Induction Tomography With a Machine Learning Approach

Muttakin

Soleimani

2021

IEEE Sensors J.

View full text Add to dashboard Cite

Identification of gas bubble, void detection and porosity estimation are important factors in many liquid metal processes. In steel casting, the importance of flow condition and phase distribution in crucial parts, such as submerged entry nozzle (SEN) and mould raises the needs to observe the phenomena. Cross-section of flow shapes can be visualised using the magnetic induction tomography (MIT) technique. However, the inversion procedure in the image reconstruction has either limited resolution or postprocessing stages. Additionally, in some cases, the displayed image may not be essential when quantifying the void fraction or porosity. This work proposes an interior void classifier based on multi-frequency mutual induction measurements with eutectic alloy GaInSn as a cold liquid metal model contained in a 3D printed plastic miniature of an SEN. The sensors consist of eight coils arranged in a circle encapsulating the column, providing combinatorial detection on conductive surface and depth. The datasets are induced voltage collections of several non-metallic inclusions (NMI) patterns in liquid metal static test and used to train a machine learning model. The model architectures are a fully connected neural network (FCNN) for 1D; and a convolutional neural network (CNN) for 2D data. The classifier using 1D data has been trained providing 98% accuracy on this dataset. On the other hand, CNN classification using multi-dimensional data produces 96% of test accuracy. Refined with representative flow scenarios, the trained model could be deployed for an intelligent online control system of the liquid metal process.

show abstract

Section: Machine Learning Methods Using Mutual Induction Datamentioning

confidence: 99%

Interior Void Classification in Liquid Metal Using Multi-Frequency Magnetic Induction Tomography With a Machine Learning Approach

Muttakin

Soleimani

2021

IEEE Sensors J.

View full text Add to dashboard Cite

show abstract

“…37 A study conducted on 15 premature newborns took data from an electrical impedance tomography (EIT) and used it in a hybrid classification model combining convoluted neural networks and SVM, resulting in an accuracy of 71% to 97%. 38 Another study collected data from 229 neonates who had apnea, with 23 features and processed it using decision trees and SVM. For the SVM-based model, a radial kernel was chosen with 10-fold cross-validation.…”

Section: Apneamentioning

confidence: 99%

Support vector machines in neonatal mortality detection: a comprehensive scoping review with disease-specific emphasis

Zoya Aamir,

Mahrosh Kasbati,

Arusha Hasan

et al. 2023

Int J Sci Rep

View full text Add to dashboard Cite

Neonatal mortality is a widely significant problem since diseases such as sepsis, apnea and jaundice have claimed the lives of 2.3 million neonates in 2021. As such, better tools need to be developed to reduce its rate. While traditional methods like Clinical risk index (CRIB) and Score for neonatal acute physiology (SNAP) have proven helpful in predicting neonatal mortality, there is a need for more efficient measures. One such approach is support vector machine (SVM), a supervised machine-learning algorithm that is primarily used for classification. SVM can perform both linear and non-linear classification; it conducts the latter with the assistance of the kernel trick and functions such as polynomial, gaussian, RBF and sigmoid functions. This narrative review aims to explore the potential and limitations of SVM in predicting major global causes of neonatal mortality. We searched through articles employing SVM to predict different diseases and symptoms such as sepsis, seizures, fetal heart rate, low birth weight, hypoxic-ischemic encephalopathy, apnea, jaundice and neonatal respiratory distress syndrome, and concluded that while SVM has its merits and has shown promising results in many aspects, it also has its demerits such as requiring an extensive training time to achieve higher accuracy and precision.

show abstract

“…LR is based on the statistical model that employs the logistic function to learn the data. Following the efficacy of traditional machine learning methods, Hansen et al(26) employed the hidden Markov Model (HMM) coupling with the higher-order features obtained from the Minkowski and Mahalanobis distances on multi-tag RFID measurements from abdominal belts for neonatal respiratory monitoring. the optimised threshold parameters above and below the zero value of the data for the classification.…”

mentioning

confidence: 99%

Artificial intelligence-driven wearable technologies for neonatal cardiorespiratory monitoring. Part 2: artificial intelligence

et al. 2022

View full text Add to dashboard Cite

show abstract

Deep Analysis of EIT Dataset to Classify Apnea and Non-Apnea Cases in Neonatal Patients

Cited by 8 publications

References 35 publications

Interior Void Classification in Liquid Metal Using Multi-Frequency Magnetic Induction Tomography With a Machine Learning Approach

Interior Void Classification in Liquid Metal Using Multi-Frequency Magnetic Induction Tomography With a Machine Learning Approach

Support vector machines in neonatal mortality detection: a comprehensive scoping review with disease-specific emphasis

Artificial intelligence-driven wearable technologies for neonatal cardiorespiratory monitoring. Part 2: artificial intelligence

Contact Info

Product

Resources

About