Keynote: “Re-engineering acute care”

Patients in the intensive care unit (ICU) are exposed to multiple potentially critical complications such as delirium, acute lung injury, acute renal failure, and sepsis. Those who suffered critical complications and survived often require costly long term care, e.g., post-hospital treatment attributed to delirium alone adds $143-152 billion per year to U.S. healthcare costs. Delirium in hospitalized patients could be reduced by 30-40% if first rate care is widely available. However, “mass production” of low cost expert physicians is impossible. In 2010, ICUs already comprised 20% of hospitals´ budgets in U.S. As the population ages, ICU usages will intensify. The magnitude of challenges faced in the ICU is enormous, and complicated by the fact that critical complications in ICU mentioned above are NOT single diseases. They are syndromes with a very wide range of precipitating factors, including aging, surgery, trauma, anesthesia, compromised immunity, systemic inflammation, infection or drug side effects. For example, with disparate mechanisms of onset, patients with sepsis have highly variable presentations: one may have a fever and high white blood cell count while another has a low body temperature and low white blood cell count. This is also a complex big data challenge. A critically ill patient can generate up to a million data points per hour from monitoring devices that need to be analyzed together with medical records, genetic profile and physicians´ insight. To address these challenges, it is important to create an accurate and robust model of the pathophysiological organ interactions and the effect and side effects of treatments, using physician supervised machine learning that integrates the conventional clinical data, biomarkers and genetic profile. Such a quantitative model will provide key description of characteristic patterns in critical complications, leading to early warning and timely preventive intervention. The success will o- en a new scientific frontier, cyber - medical systems: ushering synchronized advancement of medicine, nano-biosensors, genetics, machine learning and safety critical system integration architecture and development process. This will improve almost all aspects of medicine, creating a new industry segment, and greatly reduce costs of health care.