Quantification de la respiration en unité de soins intensifs pédiatrique à partir de séquences d’images de vidéo

Abstract

Children may suffer from respiratory failure due to lung infections. This can be difficult to predict by clinicians, whose analysis includes quantitative measurements (respiration rate) and qualitative observations (inhaled volume and chest indrawing). However, qualitative observation is difficult and imprecise when the volume is small, especially in infants. It is with that perspective that we have developed new objective approaches for respiratory failure assessment, using 3D imaging. The main objective is to design and develop a computerized clinical decision support system (CCDSS) to detect respiratory failure in critically ill children. Two types of approaches were mainly proposed to quantify four respiratory parameters (respiratory rate, tidal volume, minute ventilation and thoracoabdominal asynchrony), primarily in the paediatric population. Our approaches suggest different techniques to estimate the respiratory parameters. The first approach suggests performing a 3D reconstruction of the surface to calculate values of respiratory rate, tidal volume and minute ventilation. The system consists of two KINECT cameras recording 3D input data. Both sensors are placed in opposite sides of the patient bed. The aim of using two KINECT cameras was to provide a high spatial coverage of the patient’s thoraco-abdominal zone as well as its lateral sides. The space organization, including cable configurations and camera positions, has been optimized to facilitate the system effective deployment in intensive care rooms. Our system is fitting patient beds and can be easily incorporated into hospital rooms. The second approach is using 3D motion reconstruction to characterize the thoraco-abdominal asynchrony. The aim is to detect, extract and visualize the global motion of the surface regions and to quantify the amplitude differences between the thorax and abdomen compartments, when the seesaw motion occurs. An extensive theoretical study of human assessment monitoring systems has been proposed. The study includes both data acquisition process and data processing techniques. The systems performance has been also discussed in terms of reliability and validity, considering the needs and expectations of users, whether they are experts, patients, or the general public. To verify the applicability of the proposed CCDSS system in the clinical environment, a number of experiments have been carried by comparing the CCDSS quantitative measures to those given by the mechanical ventilator (gold standard). The experimental study investigated the robustness of our approaches in clinical environment. The proposed system is an innovative and promising support measurement tool intended to assist clinicians to monitor respiratory failure in clinical environment, especially when healthcare resources such as pediatric expert are limited.

Date	11 Dec 2019
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Rita Noumeir (Supervisor) & Philippe Jouvet (Co-supervisor)