Search

In Intensive Care Units (ICU), the abundance of multivariate time series presents an opportunity for machine learning (ML) to enhance patient phenotyping. In contrast to previous research focused on electronic health records (EHR), here we propose an ML approach for phenotyping using routinely collected physiological time series data. Our new algorithm integrates Long Short-Term Memory (LSTM) networks with collaborative filtering concepts to identify common physiological states across patients. Tested on real-world ICU clinical data for intracranial hypertension (IH) detection in patients with brain injury, our method achieved an area under the curve (AUC) of 0.889 and average precision (AP) of 0.725. Moreover, our algorithm outperforms autoencoders in learning more structured latent representations of the physiological signals. These findings highlight the promise of our methodology for patient phenotyping, leveraging routinely collected multivariate time series to improve clinical care practices., Comment: Published in 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)

Artifacts are a common problem in physiological time series collected from intensive care units (ICU) and other settings. They affect the quality and reliability of clinical research and patient care. Manual annotation of artifacts is costly and time-consuming, rendering it impractical. Automated methods are desired. Here, we propose a novel fully unsupervised approach to detect artifacts in clinical-standard, minute-by-minute resolution ICU data without any prior labeling or signal-specific knowledge. Our approach combines a variational autoencoder (VAE) and an isolation forest (IF) into a hybrid model to learn features and identify anomalies in different types of vital signs, such as blood pressure, heart rate, and intracranial pressure. We evaluate our approach on a real-world ICU dataset and compare it with supervised benchmark models based on long short-term memory (LSTM) and XGBoost and statistical methods such as ARIMA. We show that our unsupervised approach achieves comparable sensitivity to fully supervised methods and generalizes well to an external dataset. We also visualize the latent space learned by the VAE and demonstrate its ability to disentangle clean and noisy samples. Our approach offers a promising solution for cleaning ICU data in clinical research and practice without the need for any labels whatsoever.

Despite prejudicial treatment at work and school, members of low-status groups such as urban Appalachians consciously use stigmatized nonstandard dialects to foster group distinctiveness and preserve cultural identity. Research indicates that varieties of Appalachian English show no sign of total assimilation. Implications for high dropout rates among urban Appalachian students in Dayton (Ohio) are discussed. Contains 21 references. (SAS)

Supported by the Bureau of Indian Affairs, Family and Child Education (FACE) provides home-based or center-based early childhood education to children ages 0-5 years, teaches parents about child development and the promotion of active learning in their children, and addresses parents' own educational needs. In 1994-95, 22 culturally specific FACE programs served 951 families. (SV)

The design of interactive systems, including levels of automation, displays, and controls, is usually based on design guidelines and iterative empirical prototyping. A complementary approach is to use computational human performance models to evaluate designs. An integrated strategy of model-based and empirical test and evaluation activities is particularly attractive as a methodology for verification and validation of human-rated systems for commercial space. This talk will review several computational human performance modeling approaches and their applicability to design of display and control requirements.

Human Systems Integration principles and methods can be used to help integrate people, technology, and organizations in an effective and efficient manner. Over the past decade, a wide range of tools, techniques, and technologies have been developed by federal agencies to achieve significant cost and performance benefits. In this discussion, we will explore trends in military human systems integration and learn about the critical role being played by human performance and effectiveness research. We will also examine case studies on the planning and design of future human space flight vehicles, the national air space system and the first nuclear reactors to be built in the United States in over 30 years. And with an eye toward sustaining the discipline s principles and methods, we ll take a look at educating and training the next generation of human systems integration practitioners.

The exploration of space is one of the most fascinating domains to study from a human factors perspective. Like other complex work domains such as aviation (Pritchett and Kim, 2008), air traffic management (Durso and Manning, 2008), health care (Morrow, North, and Wickens, 2006), homeland security (Cooke and Winner, 2008), and vehicle control (Lee, 2006), space exploration is a large-scale sociotechnical work domain characterized by complexity, dynamism, uncertainty, and risk in real-time operational contexts (Perrow, 1999; Woods et ai, 1994). Nearly the entire gamut of human factors issues - for example, human-automation interaction (Sheridan and Parasuraman, 2006), telerobotics, display and control design (Smith, Bennett, and Stone, 2006), usability, anthropometry (Chaffin, 2008), biomechanics (Marras and Radwin, 2006), safety engineering, emergency operations, maintenance human factors, situation awareness (Tenney and Pew, 2006), crew resource management (Salas et aI., 2006), methods for cognitive work analysis (Bisantz and Roth, 2008) and the like -- are applicable to astronauts, mission control, operational medicine, Space Shuttle manufacturing and assembly operations, and space suit designers as they are in other work domains (e.g., Bloomberg, 2003; Bos et al, 2006; Brooks and Ince, 1992; Casler and Cook, 1999; Jones, 1994; McCurdy et ai, 2006; Neerincx et aI., 2006; Olofinboba and Dorneich, 2005; Patterson, Watts-Perotti and Woods, 1999; Patterson and Woods, 2001; Seagull et ai, 2007; Sierhuis, Clancey and Sims, 2002). The human exploration of space also has unique challenges of particular interest to human factors research and practice. This chapter provides an overview of those issues and reports on sorne of the latest research results as well as the latest challenges still facing the field.

Human factors is a critical discipline for human spaceflight. Nearly every human factors research area is relevant to space exploration -- from the ergonomics of hand tools used by astronauts, to the displays and controls of a spacecraft cockpit or mission control workstation, to levels of automation designed into rovers on Mars, to organizational issues of communication between crew and ground. This chapter focuses more on the ways in which the space environment (especially altered gravity and the isolated and confined nature of long-duration spaceflight) affects crew performance, and thus has specific novel implications for human factors research and practice. We focus on four aspects of human performance: neurovestibular integration, motor control and musculo-skeletal effects, cognitive effects, and behavioral health. We also provide a sampler of recent human factors studies from NASA.

Airplane upset accidents are a leading factor in hull losses and fatalities. This study compared five types of airplane-upset training. Each group was composed of eight, non-military pilots flying in their probationary year for airlines operating in the United States. The first group, 'No aero / no upset,' was made up of pilots without any airplane upset training or aerobatic flight experience; the second group, 'Aero/no upset,' of pilots without any airplane-upset training but with aerobatic experience; the third group, 'No aero/upset,' of pilots who had received airplane-upset training in both ground school and in the simulator; the fourth group, 'Aero/upset,' received the same training as Group Three but in addition had aerobatic flight experience; and the fifth group, 'In-flight' received in-flight airplane upset training using an instrumented in-flight simulator. Recovery performance indicated that clearly training works - specifically, all 40 pilots recovered from the windshear upset. However few pilots were trained or understood the use of bank to change the direction of the lift vector to recover from nose high upsets. Further, very few thought of, or used differential thrust to recover from rudder or aileron induced roll upsets. In addition, recovery from icing-induced stalls was inadequate.

An earlier study compared NASA/NSO Spectromagnetograph (SPM) data with spacecraft measurements of total solar irradiance (TSI) variations over a 1.5 year period in the declining phase of solar cycle 22. This paper extends the analysis to an eight-year period which also spans the rising and early maximum phases of cycle 23. The conclusions of the earlier work appear to be robust: three factors (sunspots, strong unipolar regions, and strong mixed polarity regions) describe most of the variation in the SPM record, but only the first two are associated with TSI. Additionally, the residuals of a linear multiple regression of TSI against SPM observations over the entire eight-year period show an unexplained, increasing, linear time variation with a rate of about 0.05 W m(exp -2) per year. Separate regressions for the periods before and after 1996 January 01 show no unexplained trends but differ substantially in regression parameters. This behavior may reflect a solar source of TSI variations beyond sunspots and faculae but more plausibly results from uncompensated non-solar effects in one or both of the TSI and SPM data sets.

This panel addresses the topic of human-centered automation in, however, a very different context-manned missions to deep space in which predominantly autonomous systems must control the crew's life-support systems most of the time. The Mars spacecraft and its inhabitants are totally dependent on the proper operation of thousands of control loops and pieces of equipment to maintain the delicate, unbuffered homeostasis of their remote existence. Mission success hangs on a thread that depends on a successful partnership, or symbiosis. between crew and machines, which exceeds anything ever required of ground systems. The need for a new approach to human-centered automation is not a "frill" in future Mars mission, but rather a sine qua non of survival. Thus the issue is design: Can we design systems intended to operate in a predominantly autonomous mode. without a human backup, that are sufficiently safe such that human travel into deep space is possible?

The Global Oscillation Network Group (GONG) Project is planning to place a set of instruments around the world to observe solar oscillations as continuously as possible for at least three years. The Project has now chosen the sites that will comprise the network. This paper describes the methods of data collection and analysis that were used to make this decision. Solar irradiance data were collected with a one-minute cadence at fifteen sites around the world and analyzed to produce statistics of cloud cover, atmospheric extinction, and transparency power spectra at the individual sites. Nearly 200 reasonable six-site networks were assembled from the individual stations, and a set of statistical measures of the performance of the networks was analyzed using a principal component analysis. An accompanying paper presents the results of the survey.

The Global Oscillation Network Group (GONG) Project will place a network of instruments around the world to observe solar oscillations as continuously as possible for three years. The Project has now chosen the six network sites based on analysis of survey data from fifteen sites around the world. The chosen sites are: Big Bear Solar Observatory, California; Mauna Loa Solar Observatory, Hawaii; Learmonth Solar Observatory, Australia; Udaipur Solar Observatory, India; Observatorio del Teide, Tenerife; and Cerro Tololo Interamerican Observatory, Chile. Total solar intensity at each site yields information on local cloud cover, extinction coefficient, and transparency fluctuations. In addition, the performance of 192 reasonable networks assembled from the individual site records is compared using a statistical principal components analysis. An accompanying paper descibes the analysis methods in detail; here we present the results of both the network and individual site analyses. The selected network has a duty cycle of 93.3%, in good agreement with numerical simulations. The power spectrum of the network observing window shows a first diurnal sidelobe height of 3 x 10(exp -4) with respect to the central component, an improvement of a factor of 1300 over a single site. The background level of the network spectrum is lower by a factor of 50 compared to a single-site spectrum.

This paper describes a cooperative approach to the design of intelligent automation and describes the Mission Operations Cooperative Assistant for NASA Goddard flight operations. The cooperative problem solving approach is being explored currently in the context of providing support for human operator teams and also in the definition of future advanced automation in ground control systems.

The design, implementation, and evaluation of ALLY, a computer-based associate for the human supervisor of a simulated satellite ground control system, is described. ALLY's dynamic intent inferencing abilities and system control properties are described. Experimental evaluation of ALLY compared performance of a two-person team with a person-ALLY team in the control of a simulated satellite ground control system.

The design, implementation, and evaluation of an Operator Function Model intelligent tutoring system (OFMTutor) is presented. OFMTutor is intended to provide intelligent tutoring in the context of complex dynamic systems for which an operator function model (OFM) can be constructed. The human operator's role in such complex, dynamic, and highly automated systems is that of a supervisory controller whose primary responsibilities are routine monitoring and fine-tuning of system parameters and occasional compensation for system abnormalities. The automated systems must support the human operator. One potentially useful form of support is the use of intelligent tutoring systems to teach the operator about the system and how to function within that system. Previous research on intelligent tutoring systems (ITS) is considered. The proposed design for OFMTutor is presented, and an experimental evaluation is described.

A portion of the Operator Function Model Expert System (OFMspert) research project is described. OFMspert is an architecture for an intelligent operator's associate or assistant that can aid the human operator of a complex, dynamic system. Intelligent aiding requires both understanding and control. The understanding (i.e., intent inferencing) ability of the operator's associate is discussed. Understanding or intent inferencing requires a model of the human operator; the usefulness of an intelligent aid depends directly on the fidelity and completeness of its underlying model. The model chosen for this research is the operator function model (OFM). The OFM represents operator functions, subfunctions, tasks, and actions as a heterarchic-hierarchic network of finite state automata, where the arcs in the network are system triggering events. The OFM provides the structure for intent inferencing in that operator functions and subfunctions correspond to likely operator goals and plans. A blackboard system similar to that of Human Associative Processor (HASP) is proposed as the implementation of intent inferencing function. This system postulates operator intentions based on current system state and attempts to interpret observed operator actions in light of these hypothesized intentions.

The authors proposes an architecture for an expert system that can function as an operator's associate in the supervisory control of a complex dynamic system. Called OFMspert (operator function model (OFM) expert system), the architecture uses the operator function modeling methodology as the basis for the design. The authors put emphasis on the understanding capabilities, i.e., the intent referencing property, of an operator's associate. The authors define the generic structure of OFMspert, particularly those features that support intent inferencing. They also describe the implementation and validation of OFMspert in GT-MSOCC (Georgia Tech-Multisatellite Operations Control Center), a laboratory domain designed to support research in human-computer interaction and decision aiding in complex, dynamic systems.

In the supervisory control of a complex, dynamic system, one potential form of aiding for the human operator is a computer-based operator's associate. The design philosophy of the operator's associate is that of 'amplifying' rather than automating human skills. In particular, the associate possesses understanding and control properties. Understanding allows it to infer operator intentions and thus form the basis for context-dependent advice and reminders; control properties allow the human operator to dynamically delegate individual tasks or subfunctions to the associate. This paper focuses on the design, implementation, and validation of the intent inferencing function. Two validation studies are described which empirically demonstrate the viability of the proposed approach to intent inferencing.