A non-invasive method of assessing Frank-Starling Curve and basing decision-making on the results.
Can be done before and after interventions such as volume expansion or diuresis, pressors, inotropes, decreasing afterload, pericardiocentesis etc.
Pattern recognition can help clinician acquire images. Unsupervised deep learning can interpret patient's physiologic profile. Data from EMR can be uploaded to cloud using natural language processing and married with focused echocardiography data.
Because of AI and deep learning clinicians at point of care can repeat the study as often as required. Input from multiple centers and big data will cause decision making to asymptotically approach optimal.
The more data a center contributes the lower will be their cost once project gets out of the developmental phase.