Algorithm to quantify the abdominal commitment in newborns

Abstract

Introduction Necrotizing enterocolitis (NEC) is a intestinal disorders involving systemic inflammatory responses in premature infants. The disease affects 1–3% of neonatal intensive care unit admissions. The NEC diagnosis and management must be based on both clinical signs, symptoms and findings on abdominal radiographs. These findings may help to predict or detect the presence of complications that require surgery. Determination of the clinical severity of NEC and possibly early prediction of its course are desirable objectives and perhaps prerequisites of successful therapy. Purpose The purpose of this investigation was to evaluate features extracted of abdominal radiographs, through image processing, as indicators of disease severity in patients with NEC. Materials and methods This study was a retrospective evaluation of 20 newborns diagnosed with NEC. A semiautomatic algorithm was developed in Matlab for the objective evaluation of the abdominal radiographs. The abdominal involvement region was manually segmented by a radiologist and assessed objectively through texture features (skewness and kurtosis). The results were compared between two groups: neonates treated medically and treated surgically. Results The results of both groups were compared using t -student test. The analyses showed significant statistic differences ( p Conclusion The proposed method can semi-automatically analyze the abdominal involvement in NEC patients. These indicators may help guide clinical management and surgery decision by adding new information to the exam. Disclosure The authors declare that there is no conflict of interest.