Non-Nutritive Suckling System for Real-Time Characterization of Intraoral Vacuum Profile in Full Term Neonates

Abstract

Infant breastfeeding diagnostics remain subjective due to the absence of instrumentation to objectively measure and understand infant oral motor skills and suckling characteristics. Qualitative diagnostic exams, such as the digital suck assessment which relies upon a clinician's gloved finger inserted into the infant's mouth, produce a diversity of diagnoses and intervention pathways due to their subjective nature. In this paper, we report on the design of a non-nutritive suckling (NNS) system which quantifies and analyzes quantitative intraoral vacuum and sucking patterns of full-term neonates in real time. In our study, we evaluate thirty neonate suckling profiles to demonstrate the technical and clinical feasibility of the system. We successfully extract the mean suck vacuum, maximum suck vacuum, frequency, burst duration, number of sucks per burst, number of sucks per minute, and number of bursts per minute. In addition, we highlight the discovery of three intraoral vacuum profile shapes that are found to be correlated to different levels of suckling characteristics. These results establish a framework for future studies to evaluate oromotor dysfunction that affect the appearance of these signals based on established normal profiles. Ultimately, with the ability to easily and quickly capture intraoral vacuum data, clinicians can more accurately perform suckling assessments to provide timely intervention and assist mothers and infants towards successful breastfeeding outcomes.

Keywords: Breastfeeding; diagnostics; instrument; intraoral vacuum; pacifier; pressure sensor.

FIGURE 1.

Image of the NNS system design with four major components: a modified pacifier, pressure sensor, data acquisition board, and a custom software interface. The design considers the intended clinical use and ease of adopting the system. The system can be used with minimal training and seamlessly integrated into the clinical workflow.

FIGURE 2.

Illustration of a typical intraoral vacuum waveform labeled with characterization parameters.

FIGURE 3.

An example of a suckling signal generated by an infant utilizing the modified pacifier (left). The figure on the right shows an example of a region of interest (a zoomed in of Burst 3) generated from the NNS software, providing immediate analysis of the signal within the region of interest defined by the clinician. Analysis includes peak and valley detection to characterize the signal parameters such as burst duration, maximum suck vacuum, and mean suck vacuum.

FIGURE 4.

Three distinguishable profiles appear in the NNS suckling signal of 30 infants: (a) group 1: smooth sinusoidal, (b) group 2: sharp valley, and (c) group 3: double valley. This figure shows representative examples of NNS signal from each group.

FIGURE 5.

Box and whisker plots comparing the extracted parameters from three classified groups. We observe statistically significant differences between groups 1 and 3 across mean suck vacuum, maximum suck vacuum, and frequency.

FIGURE 6.

Distribution plots of the three classified groups based on signal shape: (a) Mean suck vacuum (b) Max suck vacuum (c) Frequency (d) Burst duration (e) Number of sucks per burst. The distributions of mean suck vacuum, max suck vacuum and frequency of the three groups were confirmed to be normally distributed using the Shapiro-Wilk normality test. The Welch’s t-test, which requires the data to be normally distributed, was then used to compare these parameters across the three groups for any statistical differences. Burst duration (d) and Number of sucks per burst (e) appear to be right-skewed distributions.