Objectives Sudden death is common in patients with hypoplastic left heart syndrome and comparable lesions with parallel systemic and pulmonary circulation from a common ventricular chamber. It is hypothesized that unforeseen acute deterioration is preceded by subtle changes in physiologic dynamics prior to overt clinical extremis. Our objective is to develop a computer algorithm to automatically recognize precursors to deterioration in real-time, providing an early warning to care staff. Methods Continuous high-resolution physiologic recordings were obtained from 25 children with parallel systemic and pulmonary circulation who were admitted to the CVICU of Texas Children’s Hospital between their early neonatal palliation and stage 2 surgical palliation. Instances of cardiorespiratory deterioration (defined as the need for CPR or endotracheal intubation) were found via a chart review. A classification algorithm was applied to both primary and derived parameters that were significantly associated with deterioration. The algorithm was optimized to discriminate pre-deterioration physiology from stable physiology. Results Twenty cardiorespiratory deterioration events were identified in 13 of the 25 infants. The resulting algorithm was both sensitive and specific for detecting impending events, one to two hours in advance of overt extremis (ROC Area = 0.91, 95% CI = 0.88–0.94). Conclusion Automated, intelligent analysis of standard physiologic data in real time can detect signs of clinical deterioration too subtle for the clinician to observe without the aid of a computer. This metric may serve as an early warning indicator of critical deterioration in patients with parallel systemic and pulmonary circulation.

Our algorithms provide accurate and timely predictions of intracranial hypertension and tissue hypoxia crises in patients with severe traumatic brain injury. Almost all of the information needed to predict the onset of these events is contained within the signal of interest and the time since last crisis.

Summary Background The US FDA has been collecting information on medical devices involved in significant adverse advents since 1984. These reports have been used by researchers to advise clinicians on potential risks and complications of using these devices. Objective Research adverse events related to the use of Clinical Information Systems (CIS) as reported in FDA databases. Methods Three large, national, adverse event medical device databases were examined for reports pertaining to CIS. Results One hundred and twenty unique reports (from over 1.4 million reports) were found, representing 32 manufacturers. The manifestations of these adverse events included: missing or incorrect data, data displayed for the wrong patient, chaos during system downtime and system unavailable for use. Analysis of these reports illustrated events associated with system design, implementation, use, and support. Conclusion The identified causes can be used by manufacturers to improve their products and by clinical facilities and providers to adjust their workflow and implementation processes appropriately. The small number of reports found indicates a need to raise awareness regarding publicly available tools for documenting problems with CIS and for additional reporting and dialog between manufacturers, organizations, and users.

Proposal and execution of clinical trials, computation of quality measures and discovery of correlation between medical phenomena are all applications where an accurate count of patients is needed. However, existing sources of this type of patient information, including Clinical Data Warehouses (CDW) may be incomplete or inaccurate. This research explores applying probabilistic techniques, supported by the MayBMS probabilistic database, to obtain accurate patient counts from a clinical data warehouse containing synthetic patient data. We present a synthetic clinical data warehouse (CDW), and populate it with simulated data using a custom patient data generation engine. We then implement, evaluate and compare different techniques for obtaining patients counts. We model billing as a test for the presence of a condition. We compute billing’s sensitivity and specificity both by conducting a “Simulated Expert Review” where a representative sample of records are reviewed and labeled by experts, and by obtaining the ground truth for every record. We compute the posterior probability of a patient having a condition through a “Bayesian Chain”, using Bayes’ Theorem to calculate the probability of a patient having a condition after each visit. The second method is a “one-shot” approach that computes the probability of a patient having a condition based on whether the patient is ever billed for the condition Our results demonstrate the utility of probabilistic approaches, which improve on the accuracy of raw counts. In particular, the simulated review paired with a single application of Bayes’ Theorem produces the best results, with an average error rate of 2.1% compared to 43.7% for the straightforward billing counts. Overall, this research demonstrates that Bayesian probabilistic approaches improve patient counts on simulated patient populations. We believe that total patient counts based on billing data are one of the many possible applications of our Bayesian framework. Use of these probabilistic techniques will enable more accurate patient counts and better results for applications requiring this metric.