ARDS is characterized by a diffuse, bilateral, extensive alveolar-interstitial infiltrate related to damage to the alveolar membrane. Studies of lung morphology assessed by CT scan have shown 2 types of ARDS. When the loss of ventilation is posterior and caudal, and the pulmonary parenchyma is otherwise healthy, ARDS is said to be focal. When the loss of ventilation is diffuse, associated with excess tissue affecting the entire lung parenchyma, ARDS is said to be non-focal or diffuse. Only one recent multicenter study has evaluated whether individualized PEEP adjustment according to lung morphology (focal vs. non-focal) could improve ARDS survival outcomes compared with standard of care. Results for the primary endpoint, 90-day mortality, were negative. But lung morphology was misclassified on CT occurred 85 (21%) of 400 patients. In addition, per-protocol analysis in which misclassified patients were excluded showed that mortality was significantly lower in the personalized ventilation group than in the control group. Lung ultrasound may be the ideal bedside imaging technique. The loss of pulmonary aeration and its variations induced by therapeutic maneuvers can be quantified by lung ultrasound. Lung ultrasound is a non-invasive, non-ionizing radiation method that can be easily used at the bedside. APRV ventilation coupled with spontaneous breathing activity prevents atelectasis formation, recruits less ventilated areas, prevents mechanical ventilation-induced lung injury and improves lung compliance. The hypothesis is that a LUVS ventilatory strategy composed of a bundle of measures all aiming at reopening the atelectatic pulmonary zones in focal ARDS (synergistic effect of spontaneous ventilation in APRV mode, early prone sessions, and adjustment of PEEP according to the patient's BMI), could decrease the number of days under mechanical ventilation, compared to protective ventilation with low tidal volume according to the current international recommendations