A Deep Learning approach for classification of neonatal respiratory status

Master thesis work

Problem statement & solution

The problem is to assess the presence of RDS syndrome by a quantitative analysis of Lung Ultrasound (QLUS) videos. Convolutional Neural Networks (CNNs) has been used to accomplish an image-classification problem for a prediction at frame-level. Results at frame-level are then aggregated to ouput a result at video-level.