New Tools for a Technology-aided Assessment of Newborns' Oral-Motor Behavior

Thanks to technology advances, the survival of high-risk newborns is increasing, and there is a rising concern about the potential problems that this population may encounter. As a result, both medical and research communities are calling for new and more reliable tools to support health care professional in assessing the newborn's neuro-motor status, his readiness for discharge and planning for subsequent care. The newborn's oral-motor behavior during feeding is the earliest window in which care providers and parents can view and assess the well-being of the infant. One of the major developmental milestones of high-risk newborns in fact is the attainment of adequate Nutritive Sucking (NS) skills for feeding: they often show oral-motor dyscoordination and are unable to suck and feed orally. This is why the importance of monitoring newborns' oral-motor activity has been acknowledged over the years, and the need for reliable instruments has been recently stressed. Over the last decades significant advances occurred in the technology-assisted monitoring of NS disorders in newborns, however there are no accepted standards for their implementation in clinical practice, and so no objective tool is regularly and widely used. The goal of this PhD thesis was to provide new tools for a technology-aided assessment of newborns' oral-motor behavior during feeding. The first objective has been to overcome the technological challenges emerging from the state of the art, designing a hardware platform that was suitable for a wide and regular application in both clinical (pre-discharge) and domestic (post-discharge) setting: i) A low-cost device for monitoring sucking pressures has been designed and developed in order to be easily integrated on a typical feeding bottle: the measuring system has been experimentally validated in laboratory and then tested collecting data on newborns during the routine clinical practice. ii) Two novel methods for the ecological estimation of nutrient consumption during bottle-feeding have been proposed and two low-cost sensor-based modules have been developed and experimentally validated in laboratory. iii) Given the importance of oral-respiratory coordination during NS, a sensing non-invasive solution capable of monitoring the temporal breathing pattern has been designed and developed with a completely non-invasive approach, in order to be easily embeddable on a common feeding bottle. Its laboratory validation fosters its successful application, and further work is needed to test and validate the system on newborns. The second objective of this thesis has been to develop an automated analysis system capable of quantitatively assessing oral feeding ability, in order to help clinicians in routine clinical monitoring and decision-making. The increasing need for an objective assessment tool indeed makes also increasingly important to establish quantitative and reliable measures for a proper analysis of oral-motor function. A software system to systematically treat and analyze sucking pressure data has been developed: proper algorithms for automatic segmentation and features extraction have been proposed and tested for the analysis of different experimental data. Besides, a series of measures of motor control and variability in coordination has been introduced for the analysis of oral-motor function, and the results from their analysis seemed to indicate their useful contribution to the quantitative assessment of NS skills. Experimental data have been analyzed to investigate the suitability of the assessment system to the evaluation of the newborn's maturational course, as well as to quantify the deviation from a typical behavior. The results obtained are promising and further work is now dedicated to the collection of a larger longitudinal set of data from both high-risk and healthy infants, in order to support the reported findings and also define normative data.