THE PROJECT AIMS TO SET UP AN ARTIFICIAL INTELLIGENCE-BASED STUDY SYSTEM (AI) ABLE TO ANALYSE ECHOCARDIOGRAPHIC IMAGES, PULMONARY ULTRASOUND AND X-RAY IMAGES OF ADULT PATIENTS (COVID-19 NEGATIVE, COVID-19 POSITIVE AND ARDS-COVID-19 PATIENTS HOSPITALISED IN THE PARTNERSHIP HOSPITALS), TOGETHER WITH THE REPORTING INFORMATION, SO AS TO BUILD A DATASET USEFUL FOR THE TRAINING OF THE MACHINE LEARNING MODEL UNTIL SELECTING THE MODEL CAPABLE OF PROVIDING THE MOST ACCURATE PREDICTION OF DIAGNOSIS. THIS MODEL WILL PRODUCE A BINARY CLASSIFICATION OF THE PROBABILITY OF DIAGNOSIS OF NON-PATHOLOGY FROM COVID-19 OR COVID-19 PATHOLOGY TO SUPPORT CLINICIANS IN DIAGNOSIS — IN TERMS OF EARLYNESS, LESS DIFFICULTY IN DIFFERENTIAL DIAGNOSIS, RISK STRATIFICATION AND EARLY INITIATION OF OTTIMAL THERAPY- AND FOLLOW-UP (IN TERMS OF EARLY PREDICTORS OF ADVERSE CLINICAL COURSE