ICU Clinical Decision Support Case Study

Abstract:The modern ICU generates large volumes of complex and multimodal data. Interpreting and utilizing this information is challenging for the ICU Physician. By enhancing the ICU Clinical Decision Support System its outcomes in critical patients will be improved by providing real-time decision support, decreasing medical errors, and minimizing life-threatening events caused by delayed or uninformed medical decisions. In addition to basic patient demographics, pre-existing comorbidity data, and medication usage, it also emulate more event categories such as nurse-verified chart events, laboratory tests, and fluid balance records. For making critical decisions, these events based categories with event duration are more helpful. For handling

Average mortality rates ranging from 10% to 29% , which are the highest rates of all the units in a hospital. Compared to clinical settings, the ICU has some of the highest rates of medical errors. With the extensive hemodynamic monitoring and use of multiple measurement technologies, the modern ICU generates large volumes of complex and multimodal data. Interpreting and utilizing this information is challenging for the ICU physician.

We have designed and developed an ICU clinical decision support system (CDSS) to improve outcomes in critically ill patients by providing real-time decision support, decreasing medical errors, and minimizing life-threatening events caused by delayed or uninformed medical decisions. CDSSs are computer-aided ``active knowledge systems which use two or more items of patient data to generate case-specific advice'' and it can improve a physician's decision making performance for providing an evidence strongly . For optimal medical decision making, the CDSS needs to be data-driven, rapid, and

In our proposed system, besides basic patient demographics, pre-existing comorbidity data, and medication usage, we will emulate more event categories such as nurse-verified chart events, laboratory tests, and fluid balance records etc. to better assist ICU clinicians in making critical decisions. We are using these event based categories from the MIMIC-II data. Generally, event based categories are also includes the event duration. Since, for making critical decisions these events based categories with event duration are more helpful. For handling this event duration, we are proposing the temporal association rule mining in our proposed system. In our proposed system, we have introduced time in the problem of association rules discovery, given place to what we call Temporal Association Rules. To generate association rules with frequent itemsets. It is different to generate association rules without time, because it adds time information on frequent itemsets. So here the association rules are temporal ones. Each item and rule has now an associated lifespan, which comes from the explicitly defined time in database transactions. One of the problems related to the discovery of association rules that is often mentioned, is the great number of rules that can be generated. A solution is that the user may say which dates are old enough, so the rules with lifespan