Assessment of the Use of Patient Vital Sign Data for Preventing Misidentification and Medical Errors.

Patient misidentification is a preventable issue that contributes to medical errors. When patients are confused with each other, they can be given the wrong medication or unneeded surgeries. Unconscious, juvenile, and mentally impaired patients represent particular areas of concern, due to their potential inability to confirm their identity or the possibility that they may inadvertently respond to an incorrect patient name (in the case of juveniles and the mentally impaired). This paper evaluates the use of patient vital sign data, within an enabling artificial intelligence (AI) framework, for the purposes of patient identification. The AI technique utilized is both explainable (meaning that its decision-making process is human understandable) and defensible (meaning that its decision-making pathways cannot be altered, just optimized). It is used to identify patients based on standard vital sign data. Analysis is presented on the efficacy of doing this, for the purposes of catching misidentification and preventing error.

Americans' Perspectives on Online Media Warning Labels.

Americans are pervasively exposed to social media, news, and online content. Some of this content is designed to be deliberately deceptive and manipulative. However, it is interspersed amongst other content from friends and family, advertising, and legitimate news. Filtering content violates key societal values of freedom of expression and inquiry. Taking no action, though, leaves users at the mercy of individuals and groups who seek to use both single articles and complex patterns of content to manipulate how Americans consume, act, work, and even think. Warning labels, which do not block content but instead aid the user in making informed consumption decisions, have been proposed as a potential solution to this dilemma. Ideally, they would respect the autonomy of users to determine what media they consume while combating intentional deception and manipulation through its identification to the user. This paper considers the perception of Americans regarding the use of warning labels to alert users to potentially deceptive content. It presents the results of a population representative national study and analysis of perceptions in terms of key demographics.

A Pathfinding Algorithm for Lowering Infection Exposure of Healthcare Personnel Working in Makeshift Hospitals.

Due to the recent COVID-19 outbreak, makeshift (MS) hospitals have become an important feature in healthcare systems worldwide. Healthcare personnel (HCP) need to be able to navigate quickly, effectively, and safely to help patients, while still maintaining their own well-being. In this study, a pathfinding algorithm to help HCP navigate through a hospital safely and effectively is developed and verified. Tests are run using a discretized 2D grid as a representation of an MS hospital plan, and total distance traveled and total exposure to disease are measured. The influence of the size of the 2D grid units, the shape of these units, and degrees of freedom in the potential movement of the HCP are investigated. The algorithms developed are designed to be used in MS hospitals where airborne illness is prevalent and could greatly reduce the risk of illness in HCP. In this study, it was found that the quantum-based algorithm would generate paths that accrued 50-66% less total disease quantum than the shortest path algorithm with also about a 33-50% increase in total distance traveled. It was also found that the mixed path algorithm-generated paths accrued 33-50% less quantum, but only increased total distance traveled by 10-20%.

Deceptive Online Content Detection Using Only Message Characteristics and a Machine Learning Trained Expert System.

This paper considers the use of a post metadata-based approach to identifying intentionally deceptive online content. It presents the use of an inherently explainable artificial intelligence technique, which utilizes machine learning to train an expert system, for this purpose. It considers the role of three factors (textual context, speaker background, and emotion) in fake news detection analysis and evaluates the efficacy of using key factors, but not the inherently subjective processing of post text itself, to identify deceptive online content. This paper presents initial work on a potential deceptive content detection tool and also, through the networks that it presents for this purpose, considers the interrelationships of factors that can be used to determine whether a post is deceptive content or not and their comparative importance.

Machine learning performance validation and training using a 'perfect' expert system.

A method is proposed for generating application domain agnostic data for training and evaluating machine learning systems. The proposed method randomly generates an expert system network based upon user specified parameters. This expert system serves as a generic model of an unspecified phenomena. The expert system is run to determine the ideal output from a set of random inputs. These inputs and ideal output are used for training and testing a machine learning system. This allows a machine learning technology to be developed and tested without requiring compatible test data to be collected or before data collection as a proof-of-concept validation of system operations. It also allows systems to be tested without data error noise or with known levels of noise and with other perturbations, to facilitate analysis. It may also facilitate testing system security, adversarial attacks and conducting other types of research into machine learning systems. •Provides an application domain agnostic way to test machine learning technologies and facilitates the generalization of results.•Allows technologies to be tested with data with different characteristics without having to locate datasets that have these characteristics.•Utilizes randomly generated network to represent non-specific phenomena which can be used for training and testing machine learning techniques.

Cyber-mitigation: Cybersecurity emergency management.

Cybersecurity is within the realm of emergency management, as cyber-attacks can generate both virtual and real world issues that emergency responders may be called upon to deal with. However, it has a skillset and other characteristics that are distinct from the types of emergency management that most practitioners commonly-and are prepared-to deal with. This paper compares the two disciplines, discusses areas where cybersecurity professionals and researchers can learn from the emergency management discipline and proposes new research directions within the emergency management domain.

Human face images from multiple perspectives with lighting from multiple directions with no occlusion, glasses and hat.

Facial and other human recognition techniques are being used for a growing number of applications, ranging from device security to surveillance video identification to forensics. Data sets are required to test recognitions algorithms. This data set facilitates the evaluation of the impact of multiple factors on algorithm performance. The data set includes images taken under five different lighting levels (which vary in light brightness and temperature), seven different lighting positions and five different subject positions. The data set includes data collected for all combinations of the foregoing three collection variables, for a total of 175 images per subject. In addition, sets of data under three different occlusion conditions (no occlusion, glasses and hat) have been collected. Each data set includes images taken under all lighting level, lighting position and subject position combinations, for a total of 525 images of each subject. The images are all taken in the same location with the same background and camera equipment.

Correlated lip motion and voice audio data.

This data set is comprised of correlated audio and lip movement data in multiple videos of multiple subjects reading the same text. It was collected to facilitate the development and validation of algorithms used to train and test a compound biometric system that consists of lip-motion and voice recognition. The data set is a collection of videos of volunteers reciting a fixed script that is intended to be used to train software to recognize voice and lip-motion patterns. A second video is included of the individual reciting a shorter phrase, which is designed to be used to test the recognition functionality of the system. The recordings were collected in a controlled, indoor setting with a 4K professional-grade camcorder and adjustable, LED lights.