Breathing and temperature detection using thermal images in infants

Abstract

Sudden Infant Death Syndrome (SIDS) is one of the leading causes of death in children under one year of age. SIDS is defined as the death of a child under one year of age, without apparent symptoms or causes during sleep that lead to his death. Therefore, monitoring of childhood sleep has become vital for parents, health care providers and health educators during the perinatal period. Temperature and respiration are the parameters most used during neonatal and postnatal consultations as an indicator of events related to the SIDS. This document proposes a method of non-invasive detection of temperature and breathing, based on the automatic detection of areas of the forehead, nose and mouth of the infant, calculating statistical measures and pixel-temperature conversion of thermal images. The proposed method used 7.189 thermal images of 6 infants during sleep, obtaining a general accuracy of 88,26% in temperature and breathing detection into three different classes.