A machine learning system for automatic detection of preterm activity using artificial neural networks and uterine electromyography data

Abstract

Background: Preterm births are babies that are born before 37 weeks of gestation. The premature delivery of babies is regarded as a major global public health issue with those affected at greater risk of developing short and long-term complications. The care provided for premature infants has significantly improved. However, it has had no impact on reducing the prevalence of preterm birth. Therefore, a better understanding of why preterm births occur is needed. Methods: Electromyography is used to capture electrical activity in the uterus to help treat and understand the condition, which is time consuming and expensive. This has led to a recent interest in automated detection of the electromyography correlates of preterm activity. This paper explores this idea further using artificial neural networks to classify term and preterm records, using an open dataset containing 300 records of uterine electromyography signals. The Synthetic Minority Oversampling TEchnique is used to oversample the minority preterm class (38 records) to address the issues found in unbalanced datasets and classification. Results: Our approach shows an improvement on existing studies with 94.56% for sensitivity, 87.83% for specificity, and 94% for the area under the curve with 9% global error when using the Multilayer perceptron neural network trained using the Levenberg-Marquardt algorithm. Discussion: The Multilayer perceptron neural network trained using the Levenberg-Marquardt algorithm produced the best results, which is trained using Newton's method of least squares optimization and is an efficient learning algorithm for neural networks that have a few hundred weights, despite being computationally expensive.

A Critical Look at Studies Applying Over-Sampling on the TPEHGDB Dataset

“…While impressive (near-perfect) results on the TPEHGDB dataset are reported in many studies [10,20,33,2,16,19,12,27,1,34,21,29,18,17,11,15], these results should be interpreted cautiously as their evaluation methodology is based on applying over-sampling strategies before data partitioning. All these studies apply over-sampling in order to make the distribution of classes more uniform.…”

A Proposal for observing Conceived Ladies having High Risk of Premature Delivery using WHSN

“…However, the quality of EHG signals is based on the suitable skin preparation and accurate placement of electrodes on abdomen of the pregnant lady. Hence with electrohysterography as noninvasive alternative for uterine monitoring, will be able to identify the incidence of normal and premature labor up to the pregnancy period of 27 weeks [20]. By utilizing WHSNs to gather EHG flags and examining them with a smart device, we can know if the pregnant lady is in the process of giving birth and alert her if required.…”