A boosting inspired personalized threshold method for sepsis screening

Sepsis is one of the biggest risks to patient safety, with a natural mortality rate between 25% and 50%. It is difficult to diagnose, and no validated standard for diagnosis currently exists. A commonly used scoring criteria is the quick sequential organ failure assessment (qSOFA). It demonstrates very low specificity in ICU populations, however. We develop a method to personalize thresholds in qSOFA that incorporates easily to measure patient baseline characteristics. We compare the personalized threshold method to qSOFA, five previously published methods that obtain an optimal constant threshold for a single biomarker, and to the machine learning algorithms based on logistic regression and AdaBoosting using patient data in the MIMIC-III database. The personalized threshold method achieves higher accuracy than qSOFA and the five published methods and has comparable performance to machine learning methods. Personalized thresholds, however, are much easier to adopt in real-life monitoring than machine learning methods as they are computed once for a patient and used in the same way as qSOFA, whereas the machine learning methods are hard to implement and interpret.     

Enterprise risk management for automation in correctional facilities with pandemic and other stressors

Real-time tracking of tool and equipment inventories is a critical function of many organizations and sectors. For prisons and correctional facilities, tracking and monitoring of assets such as cookware, hardware, keys, janitorial equipment, vocational/technical specialty tools, etc., is essential for safety, security, trust, efficiency, education, etc. The performance of automated systems for this purpose can be diminished by a variety of emergent and future sociotechnical factors alone and in combination. This article introduces a methodology for contractor evaluation and selection in acquisition of innovative asset management systems, with an emphasis on evolving system requirements under uncertainty. The methodology features a scenario-based preferences analysis of emergent and future conditions that are disruptive to the performance of the asset-control system. The conditions are across technologies, operating environments, regulations, workforce behaviors, offender behaviors, prices and markets, organizations, cyber threats, etc. The methodology addresses the influence and interaction of the conditions to disrupt system priorities. Examples include: (i) infectious disease disrupting priorities among requirements and (ii)  radio-frequency identification (RFID) and wireless-technology innovations disrupting priorities among stakeholders. The combinations of conditions that most and least matter for the system acquisition are characterized. The methodology constitutes a risk register for monitoring sources of risk to project performance, schedule, and cost throughout the system lifecycle. The results will be of interest to both practitioners and scholars engaged in systems acquisition as the pandemic interacts with other factors to affect risk, uncertainty, and resilience of organizational missions and operations.