Assessing the consciousness of critically ill patients with RGB video acquisition and artificial intelligence
Critically ill patients are continuously and closely monitored in order to quickly intervene in cases where their state of health decreases abruptly. The Pediatric Intensive Care Unit (PICU) at the Centre Hospitalier Universitaire (CHU) Ste-Justine receives patients in critical condition ranging from newborns to 18-year-olds.
Consciousness in this context represents the patient’s wakefulness and awareness of his/her surroundings. It is an important indicator of the patient’s neurological state; it also indicates poor elimination of carbon dioxide from the blood.
Consciousness is defined by: arousal or wakefulness, and awareness of the environment. Arousal is assessed by the presence of eye opening. It ranges from alert waking (spontaneous eye opening) to coma (no eye opening). Awareness, on the other hand, is determined by the patient's capacity to perceive and interact with the external world. This is assessed by observing the patient’s capacity to voluntarily react to various auditory, tactile, visual, or noxious stimuli. A patient needs to be aroused in order for the cognitive processes required for awareness to occur. Although arousal is necessary, it is not a sufficient condition for awareness. The assessment of consciousness and distress in children under 2 years of age presents a particular challenge since they do not have the ability to respond to commands or explain what they are feeling in the same way that older children do. Patients who are capable of eye opening but fail to show signs of awareness are not considered to be in a coma but rather in an unresponsive state. When voluntary movements are observed, patients are described as being in a minimally conscious state. Physicians observe and assess the patient’s consciousness visually, by watching the patient when they are within the proximity of their bed. This method of observation is subjective, qualitative, and is limited to a very short time window.
Our objective consists in using video analysis and machine learning to help in the assessment of patients’ distress in the Intensive Care Unit. More specifically, we intend to focus on assessing patients’ consciousness. To assess the patient’s consciousness, we plan to evaluate his/her responsiveness to a salient stimuli by measuring Eye features such as the eye gaze, blink rate, pupil dilation, and fixations from video acquired with RGB and RGB-D cameras. Artificial intelligence based methods will be used to segment the face region and extract the facial features.
- Une connaissance des algorithmes d’intelligence artificielle;
- Une excellente motivation;
- Un bon dossier en recherche;
- Capacité de travailler en équipe et d’une façon autonome;
- Capacité de bien communiquer par écrit;
- Excellente connaissance de la programmation.