Integrating Data, Models, and Reasoning in Critical Care

描述（由申请人提供）：

DESCRIPTION (provided by applicant): The objective of this Bioengineering Research Partnership is to focus the resources of a powerful interdisciplinary team from academia (MIT), industry (Philips Medical Systems), and clinical medicine (Beth Israel Deaconess Medical Center, BIDMC) to develop and evaluate advanced ICU patient monitoring systems that will substantially improve the efficiency, accuracy, and timeliness of clinical decision making in intensive care. Modern intensive care units employ an impressive array of technologically sophisticated instrumentation to provide detailed measurements of the pathophysiological state of each patient. In the long term, we plan to build monitoring systems that not only report these measurements to human users but also form pathophysiological hypotheses that best explain the rich and complex volume of relevant data from clinical observations, bedside monitors, mechanical ventilators and a wide variety of laboratory tests and imaging studies. Such systems should reduce the ever-growing problem of information overload, and provide much more accurate and timely alarms than today's unintegrated limit alarms. By helping to focus the practitioner's attention on the most significant events and changes in the patient's state and by suggesting likely physiological interpretations of that state, such systems will eventually permit early detection of even complex problems and provide useful guidance on therapeutic interventions; thus their use should lead to improved patient outcomes. To achieve these long-term goals, we propose a step-wise approach. First, we will create a research database of 500 data-rich ICU cases that we will de-identify and thoroughly annotate so that we can make it available as a resource for ourselves and other researchers. Second, we will develop an array of sophisticated model-based and reasoning methods and corresponding software to analyze the data we collect and to create the technical means of abstracting from detailed data to pathophysiological hypotheses. Third, we will evaluate the utility of our newly developed tools in the laboratory utilizing the new database. Finally, we will deploy the most successful of our new techniques into clinical practice in the BIDMC ICUs to compare their safety and efficacy with existing monitoring systems.