Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Intelligent Systems”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 577 records · Page 32Linked to original sources
PubMed1996

Analyzing components of work samples to evaluate performance.

A method is described in which work sample data were analyzed to reveal how aircraft maintenance technicians differed in troubleshooting performance after a training intervention. During the training phase of the study, approximately half the technicians received troubleshooting instruction from an intelligent tutoring system, and the remaining technicians received informal on-the-job training. Data from work samples, in which technicians verbally isolated simulated equipment failures, were collected from both groups before and after the training. The work sample data were then examined by subject matter experts who scored performance for overall proficiency and noted specific violations of good troubleshooting practice. These violations provided the basis for the components of performance examined in the work sample data. The results of this componential analysis allow more specific conclusions to be drawn regarding the effectiveness of the training intervention by revealing how the tutoring system improved technicians' troubleshooting (e.g., tutored technicians were more likely than untutored technicians to test components thoroughly to ensure they were faulty before replacing them).

Aerospace Medicine↗
PubMed1987

Autosomal recessive juvenile cataract in Hutterites.

Autosomal recessive inheritance of juvenile cataract is described amongst several related sibships of Lehrerleut Hutterites. The main features of the cataract include onset between three and seven years of age; rapid progression to maturity within one to three months; normal intelligence; no systemic associations, and no urinary reducing substances and normal erythrocyte galactokinase activity. Genetic analysis demonstrates the close relationship between parents of affected sibships with a coefficient of inbreeding of affected sibships of 0.0512. Estimates of heterozygote frequency within Lehrerleut Hutterites at 0.128 indicate that if current inbreeding practice continues additional cases can be expected.

Alberta↗
PubMed2006

Neuronal networks and textural descriptors for automated tissue classification in endoscopy.

This study examines the potential of neuronal networks and textural feature extraction for recognising suspicious regions in endoscopy under variable perceptual conditions and systematic or random noise in the data. Second-order statistics and discrete wavelet transform-based methodologies are examined in terms of their discrimination abilities, and several neuronal network learning algorithms are compared in terms of success. The results provide numerical evidence that neuronal networks are capable of classifying offline and online tissue samples extracted from standard images and VHS videotape recordings of colonoscopy procedures with satisfactory success rates. This type of technology could prove to be useful for developing intelligent adaptive systems that will assist medical experts in real-time to automate minimally invasive diagnostic procedures.

Artifacts↗
PubMed1999

The Internet--approaching a ubiquitous tool for dental education.

The Internet holds promise of addressing some of dentists' and dental students' needs caused by the rapid expansion of knowledge. The Internet is used primarily for communication and for courses, but the quality of Internet courses is uneven. The Internet holds potential in formal dental education for supporting information and instructional databases and as an educational administrative tool. The use of the Internet for educational purposes raises issues such as locating resources on line, ensuring information quality, and incorporation of emerging computer architecture. Some projections are offered for the future: increased instructional quality, replacement of textbooks, creating "courses" out of modules of instructional material, faculty becoming coaches, increased use of trained instructional technologists, intelligent tutoring systems, and the merging of information, instruction, and practice systems. Numerous web sites are given.

Databases as Topic↗
PubMed2001

Development of visual diagnostic expertise in pathology.

In this study, we use methods from information-processing to explore the visual diagnostic processes of novice, intermediate, and expert pathologists. Subjects were asked to examine and interpret a set of slides while we collected think-aloud verbal protocols and captured on digital video the actual visual data they examined from the microscope. We performed an in-depth combined video and protocol-based analysis of processes and errors occurring as the task was performed. Additionally, we collected measures of accuracy, certainty, and difficulty for all cases. Our preliminary analysis identified significant differences between groups in all three major aspects of this task: searching skills, perceptual skills and cognitive/reasoning skills. We describe the implications of our preliminary cognitive task analysis on the design of a developing intelligent educational system in Pathology.

Breast↗
PubMed2002

[Applications of informatics in anesthesiology: anesthesia graphics].

Computerization has brought radical changes to anesthesiology. Quality of care, management, cost control, training, research, safety and privacy have all improved. The anesthesiologist has been freed from repetitive clerical tasks and is able to make better use of time. A graphic display of anesthesia is only one of the many computer applications available as a consequence of links created among monitoring, continuous infusion and intelligent alarm systems, automatic data collection, network monitoring and the availability of bibliographic information (through Internet connection). The computer graphic display of anesthesia is more precise, legible, complete and reliable (during critical events, in substitutions of anesthesiologists or for research) than the traditional graph. One of the greatest problems of computer graphing today--besides start-up costs--is that of inserting comments on monitoring artifacts, given that the graph is a legally valid medical document.

Anesthesiology↗
PubMed2002

Detecting relationships between physiological variables using graphical models.

In intensive care physiological variables of the critically ill are measured and recorded in short time intervals. The proper extraction and interpretation of the information contained in this flood of information can hardly be done by experience alone. Intelligent alarm systems are needed to provide suitable bedside decision support. So far there is no commonly accepted standard for detecting the actual clinical state from the patient record. We use the statistical methodology of graphical models based on partial correlations for detecting time-varying relationships between physiological variables. Graphical models provide information on the relationships among physiological variables that is helpful e.g. for variable selection. Separate analyses for different pathophysiological states show that distinct clinical states are characterized by distinct partial correlation structures. Hence, this technique can provide new insights into physiological mechanisms.

Critical Care↗
PubMed1992

Monitoring and management of acute circulatory problems: the expanded role of the physiologically oriented critical care nurse.

The intensive care unit is set apart from other hospital patient care areas by (1) physiological instrumentation that permits better assessment and more effective therapy and (2) more intensive nursing. These capabilities allow nurse and physician intensivists to evaluate tissue perfusion and tissue oxygenation by the temporal patterns of oxygen delivery and oxygen consumption, as well as hemodynamics. Such evaluation provides important insight into the functional impairment of hemorrhagic, traumatic, septic, and postoperative shock. Using artificial intelligence-based systems, complex clinical algorithms--tailored to specific patient conditions--have been developed and are described. These algorithms are based on both invasive and noninvasive monitoring systems and on clinical experience with a large series of high-risk surgical patients.

Algorithms↗
PubMed2000

Multiparameter trend monitoring and intelligent displays using wavelet analysis.

An Intelligent Patient Monitoring (IPM) framework is defined for the analysis and display of multiparameter trends from ICU patients. Wavelet analysis was utilized for detection of physiological events and artifacts in long-term trends. A group of 58 patients from the MIMIC database were identified in which the heart rate (HR), arterial blood pressure (ABP), and pulmonary artery pressure (PAP) were monitored. An estimated cardiac output (CO) signal, using HR and ABP, was shown to correlate strongly (r=.67) with actual CO measurements. Using wavelet analysis, automated artifact and physiological event detection algorithms were developed to monitor left ventricular hemodynamic function. Finally, an intelligent display system is presented for presentation of the data in ICU monitors.

Algorithms↗
PubMed1992

Concept graphics: a language for medical knowledge.

This paper makes a case for Concept Graphics, a novel form of medical knowledge representation. Concept Graphics are assemblies of icons each of which has a precise meaning. Individual icons are metaphors of the object or process they represent. Concept Graphics are analogs of pathological situations, their symptoms, signs and other relevant components necessary for a diagnosis. We propose three principal areas of application for Concept Graphics: medical education, medical records management and research based on medical records. Our earlier work in medical education showed a clear advantage in using Concept Graphics in parallel with text of equivalent information content, over text alone. A Concept Graphics based intelligent tutoring systems (ITS) is being developed. In the area of medical records management we are developing a system for the rapid identification of relevant records based on rapid visual screening. The Concept Graphics based system can reveal properties common to specific groups of records. As such the graphics are a research tool.

Computer Graphics↗
PubMed1992

The effect of two intraoperative heat-conserving methods on orthopedic patients receiving regional and general anesthesia.

Postoperative hypothermia is a serious complication of surgery. The best way to prevent postoperative hypothermia is to prevent perioperative heat loss. The purpose of this study was to compare the efficacy of Thermadrape (O.R. Concepts, Inc., Roanoke, TX), a heat-retaining surgical covering, with cotton blankets in maintaining core body temperature for patients receiving general or regional anesthesia. One hundred seventeen patients having hand, wrist, or elbow surgery were covered with either cotton blankets or Thermadrape covering throughout the perioperative period. Core body temperature was measured using a tympanic membrane device (FirstTemp, Intelligent Medical Systems, Carlsbad, CA) preoperatively, on PACU admission, and at discharge. Thermadrape covering and cotton blankets were equally effective in conserving body heat. Patients receiving general anesthesia were cooler on PACU admission than those receiving regional anesthesia, regardless of the type of covering. Hospital decisions regarding choice of perioperative heat conservation interventions will be guided by the cost of the intervention as well as efficacy. If overall hospital cost of cotton blankets is greater than that associated with Thermadrape covering, a decision to use Thermadrape covering may be warranted. Important factors for consideration in future research are presented and discussed.

Anesthesia, Conduction↗
PubMed2004

[Measuring the effectiveness of health promotion policy: what can be learnt from Australian success stories?].

In 1987, the Australian state of Victoria has created VicHealth, the Victorian Health Promotion Foundation, which has since been quite successful in implementing effective health promotion policies. This paper, after presenting briefly three of these success stories (tobacco control, road trauma decrease and skin cancer prevention), goes on to derive from them the essential steps of successful health promotion interventions. It concludes by reflecting on the nature of the intelligence gathering systems required to anchor good health promotion in good science, and on the barriers to proper measurement that are still pervasive.

Accidents, Traffic↗
PubMed2005

[The compensatory light for car plate recognition].

In the intelligent traffic system, the automatic recognition system of the car plate is an important development field. This system must satisfy the requirement of all-day long inspection. In the course of inspection, the lack of natural light(such as at night, in rainy day...) could be a problem, that is to say there could be backlighting problem, and this problem can be solved by using compensatory light. By now, the high-power floodlight has been adopted. Because of its own characteristics of the floodlight, the satisfied result cannot be obtained. In the present thesis, how to select the right light is analyzed, including car lights' spectra and visual response, and so on. The results of the practices are very satisfied.

Automobiles↗
PubMed2006

Pulse!!--A virtual learning space project.

Game-based technology structured within an epistemic framework is revolutionizing learning. Intersecting military simulation technologies with modern pedagogical education practices, Pulse!! offers an epistemic framework for optimizing cognitive and psychomotor skills in clinical practices. Its intelligent authoring system is an integrated, fully immersive, interactive virtual learning simulation for civilian and military health care practitioners, providing opportunities to make mistakes and repeat actions.

Computer Simulation↗
PubMed2005

A method for automated detection of usability problems from client user interface events.

Think-aloud usability analysis provides extremely useful data but is very time-consuming and expensive to perform because of the extensive manual video analysis that is required. We describe a simple method for automated detection of usability problems from client user interface events for a developing medical intelligent tutoring system. The method incorporates (1) an agent-based method for communication that funnels all interface events and system responses to a centralized database, (2) a simple schema for representing interface events and higher order subgoals, and (3) an algorithm that reproduces the criteria used for manual coding of usability problems. A correction factor was empirically determining to account for the slower task performance of users when thinking aloud. We tested the validity of the method by simultaneously identifying usability problems using TAU and manually computing them from stored interface event data using the proposed algorithm. All usability problems that did not rely on verbal utterances were detectable with the proposed method.

Algorithms↗
PubMed2006

Intelligent mobile voice information centre system: Taking the emergency triage support system as an example.

According to research by the International Telecommunications Union (ITU), Taiwan has the highest percentage of people holding mobile (cell) phones, and every one has one or two cell phones. We usually use mobile phones to make contact with hospitals or dial 119 in an emergency. We have developed a new call flow style and combine traditional passive call flow. The system is designed with a voice menu and users can command services actively. The telephone connects to a computer server to enter the call flow module with ASR and TTS. The recognition term comes from a voice term database. The voice management center is an ER triage decision system. We have tested three scenarios which are DOA, a large number of patients and critical trauma. We measured the completion time and recognition rate for all the scenarios. The recognition rate in all three cases was above 90%. The average completion time for DOA is 38.6 seconds, for the large number of patients is 20 seconds and for the critical trauma case, the average is 252.7 seconds. It is too slow to input one patient's data but the communication transfer of the message is fast and convenient; the DOA notice only takes 38 seconds to inform medical staff, which is very efficient and medical staff believe that this can help their communication. We will continue to extend our system and further apply it to other domains. Future studies may combine the Internet platform with the WiFi phone and soft phone (Skype) and broaden the scope of the application in Taiwan.

Emergency Medical Services↗