Acute respiratory failure (ARF) is a common condition and a common reason for urgent medical consultation. Assessing the extent of respiratory impairment is important to improve the management of patients with ARF. When Acute respiratory failure is caused by pathology of the pulmonary parenchyma, quantification of pulmonary radiographic involvement may be a component of the initial assessment of severity. This radiographic quantification would only be usable in clinical routine if it can be automated and provide a real-time result.

The objective of this work is to assess the feasibility of an automated technique for quantifying radiological lung damage in situations of known or potential ARF.

Inclusion Criteria:

• Patients over 18 years of age
• Presenting a situation at risk of acute respiratory failure
• Requiring thorax imaging when they were treated.

Exclusion Criteria:

• Patients with anatomically incomplete thoracic imaging (e.g., technical acquisition error).

Patients presenting a situation at risk of acute respiratory failure