Machine learning assisted point of care mid infrared spectroscopy for neonatal respiratory distress syndrome diagnosis

Abstract

Point of care devices shorten the time required to reach a diagnosis and access to treatment. In emergency care this can have a direct impact on patient prognosis. Neonatal respiratory distress syndrome (nRDS) is a condition affecting neonates born 10—15 weeks early with underdevel-oped lungs deficient in surfactant. This leads to an increased lung sur-face tension and can lead to alveolar collapse. Treatment requires an exogenous replacement of surfactant which is expensive and can also lead to further chronic complications if not required. There are no cur-rent point of care devices that can diagnose nRDS but the ratio of two lung biomarkers, lecithin (L) and sphingomyelin (S) is known to correlate with lung maturity and knowledge of this can help clinicians decide whether the benefits of treatment outweigh its risks. Some clinical studies have found that neonates with L/S ratios below 2.2 require sur-factant replacement treatment. Since these biomarkers have a mid in-frared spectrum, we propose attenuated total reflectance with Fourier transform infrared (ATR-FTIR) spectrometry to measure the L/S ratio

More information

Identifiers

ORCID for Aneesh Vincent Veluthandath: orcid.org/0000-0003-4306-6723

ORCID for James S. Wilkinson: orcid.org/0000-0003-4712-1697

ORCID for Ganapathy Senthil Murugan: orcid.org/0000-0002-2733-3273

Catalogue record

Export record