Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “artificial intelligence”

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 847 records · Page 47Linked to original sources
PubMed1996

Intelligent fetal heart rate computer systems in intrapartum surveillance.

The intrapartum cardiotocogram has had a disappointing impact on clinical practice. Misinterpretation of the cardiotocogram not only causes an increase in unnecessary intervention but is also implicated in a large proportion of patients with birth asphyxia and avoidable perinatal morbidity. Over the past 10-20 years, groups have attempted to develop computer systems capable of analysing the cardiotocogram but with limited success. The likely explanation is that these conventional systems analyse the cardiotocogram in isolation from clinical factors. More recently, research has focused on the use of artificial intelligence techniques which can assess the whole clinical picture to support clinical decision making during labour. The current literature is reviewed and a system that has been validated by comparison of its performance with 17 experts is described.

Cardiotocography↗
PubMed2000

The Turing Test: the first 50 years.

The Turing Test, originally proposed as a simple operational definition of intelligence, has now been with us for exactly half a century. It is safe to say that no other single article in computer science, and few other articles in science in general, have generated so much discussion. The present article chronicles the comments and controversy surrounding Turing's classic article from its publication to the present. The changing perception of the Turing Test over the last 50 years has paralleled the changing attitudes in the scientific community towards artificial intelligence: from the unbridled optimism of 1960s to the current realization of the immense difficulties that still lie ahead. I conclude with the prediction that the Turing Test will remain important, not only as a landmark in the history of the development of intelligent machines, but also with real relevance to future generations of people living in a world in which the cognitive capacities of machines will be vastly greater than they are now.

Journal Article↗
PubMed1998

Expert systems in anesthesiology.

There are only a limited number of computer-based systems designed to support anesthesiologists in the operating room. This is evident from the very small number of publications on this topic. These systems may be classified by the functions they perform and include: intelligent anesthesia workstations (with current data acquisition, conditioning and analysis subsystems), systems to detect critical conditions in patients (including expert systems in smart alarm capacity), anesthesia management systems (for planning and management) and drug administration systems. Drug administration systems may be subdivided into open-loop and closed-loop systems. The techniques applied for design of such systems vary extensively. There are traditional rule-based expert systems and probability-based systems, and more recently developed artificial intelligence methods, such as neural networks and fuzzy logic. Computers are valuable tools that have the potential to assist anesthesiologists in carrying out cumbersome and monotonous processes. Future efforts may result in the development of sophisticated systems capable of assuming more responsibilities and of reducing human workload and stress.

Journal Article↗
PubMedDate not supplied

ESP: a method to predict toxicity and pharmacological properties of chemicals using multiple MCASE databases.

We describe here the development of a computer program which uses a new method called Expert System Prediction (ESP), to predict toxic end points and pharmacological properties of chemicals based on multiple modules created by the MCASE artificial intelligence system. The modules are generally based on different biological models measuring related end points. The purpose is to improve the decision making process regarding the overall activity or inactivity of the chemicals and also to enable rapid in silico screening. ESP evaluates the significance of the biophores from a different viewpoint and uses this information for predicting the activity of new chemicals. We have used a unique encoding system to represent relevant features of a chemical in the form of a pattern vector and a genetic artificial neural network (GA-ANN) to gain knowledge of the relationship between these patterns and the overall pharmacological property. The effectiveness of ESP is illustrated in the prediction of general carcinogenicity of a diverse set of chemicals using MCASE male/female rats and mice carcinogenicity modules.

Algorithms↗
PubMed2001

Impact of different variables on the outcome of patients with clinically confined prostate carcinoma: prediction of pathologic stage and biochemical failure using an artificial neural network.

BACKGROUND: The advent of advanced computing techniques has provided the opportunity to analyze clinical data using artificial intelligence techniques. This study was designed to determine whether a neural network could be developed using preoperative prognostic indicators to predict the pathologic stage and time of biochemical failure for patients who undergo radical prostatectomy. METHODS: The preoperative information included TNM stage, prostate size, prostate specific antigen (PSA) level, biopsy results (Gleason score and percentage of positive biopsy), as well as patient age. All 309 patients underwent radical prostatectomy at the University of Colorado Health Sciences Center. The data from all patients were used to train a multilayer perceptron artificial neural network. The failure rate was defined as a rise in the PSA level > 0.2 ng/mL. The biochemical failure rate in the data base used was 14.2%. Univariate and multivariate analyses were performed to validate the results. RESULTS: The neural network statistics for the validation set showed a sensitivity and specificity of 79% and 81%, respectively, for the prediction of pathologic stage with an overall accuracy of 80% compared with an overall accuracy of 67% using the multivariate regression analysis. The sensitivity and specificity for the prediction of failure were 67% and 85%, respectively, demonstrating a high confidence in predicting failure. The overall accuracy rates for the artificial neural network and the multivariate analysis were similar. CONCLUSIONS: Neural networks can offer a convenient vehicle for clinicians to assess the preoperative risk of disease progression for patients who are about to undergo radical prostatectomy. Continued investigation of this approach with larger data sets seems warranted.

Adult↗
PubMed1998

Weaning from artificial ventilation.

Every intubated and mechanically-ventilated patient should be clinically evaluated, at least on a daily basis, by a skilled team in order to speed up the weaning process as much as possible. Again, it should be emphasized that the adoption of an active clinical strategy when faced with "difficult" to wean patients is of paramount importance. In one study, performed in Spain, analysing the prevalence of mechanical ventilation in intensive care units [3], reported the mean number of days that patients spent on mechanical ventilation was 27. In a more recent intervention study, in which a specific protocol was followed each day [2], the mean number of days on mechanical ventilation was only 12. These data have been confirmed by several authors [4, 40], and it has also been reported that a protocol-directed weaning strategy leads not only to a significant reduction in the duration of mechanical ventilation but also to a significant decrease in the number of complications and cost [4]. However, even following a protocol-directed weaning strategy, it is possible that weaning duration can be further reduced. In a prospective study performed in our institution [41] during 32 months, we reported that, following an episode of unplanned extubation, the only independent variables associated with the need for reintubation were the number of days of mechanical ventilation and the type of ventilatory support at the time of autoextubation. Indeed, when patients were in the weaning period only 16% (5 out of 32) needed reintubation, whereas reintubation was needed in 82% (22 out of 27) of patients who had an unplanned extubation during full mechanical ventilatory support. These data suggest that there are still some patients being on mechanical ventilation for a longer than necessary period of time. Finally, very recent advances in technological areas such as artificial intelligence, are proving to be useful in the management of the weaning process. When such systems are applied to modern microprocessor-controlled mechanical ventilators they can significantly help in the process of weaning [42] by automatically reducing the ventilatory assistance and by indicating the optimal time to withdraw the patient from the ventilator and proceed with extubation.

Humans↗
PubMed1984

An expert system for the evaluation of liver functional assessment.

The paper describes an expert system for the assessment of the liver function. Since the system must act as an intelligent assistant for a general physician, a major emphasis has been laid upon its interactive capabilities. In particular, the user can ask the system how a given conclusion has been reached (explanation facilities) and can alter the normal operation flow. In the design of the system, the different significance and availability of the clinical data and the laboratory tests have been taken into account: The investigations about a given patient start from the clinical data, and only when there is some evidence of hepatopathy are the results of some laboratory tests requested. The final result of the investigations consists of the assessment of the liver function in terms of four aspects (biosynthesis, cholestasis, cytolysis, reactivity), each of which is assigned a linguistic value describing its impairment degree. The techniques adopted in the system are based on Artificial Intelligence methodologies augmented with linguistic terms handled according to the fuzzy set theory.

Diagnosis, Computer-Assisted↗
PubMed2000

A fuzzy-neural system for identification of species-specific alarm calls of Gunnison's prairie dogs.

In this study we describe the design and application of an automated classification system that utilizes artificial intelligence to corroborate the finding that Gunnison's prairie dogs have different alarm calls for different species of predators. This corroboration is strong because it utilizes an entirely different analysis technique than that used in the original research by Slobodchikoff et al. [Slobodchikoff, C.N., Fischer, C., Shapiro, J., 1986. Predator-specific alarm calls of prairie dogs. Am. Zool. 26, 557] or in subsequent study done by Slobodchikoff et al. [Slobodchikoff, C.N., Kiriazis, J., Fischer, C., Creef, E., 1991. Semantic information distinguishing individual predators in the alarm calls of Gunnison's prairie dogs. Anim. Behav. 42, 713-719]. The study described here also is more completely automated than earlier study in this area. This automation allowed a large volume of field data to be processed where all measurements of relevant parameters were performed through software control. Previous study processed a smaller data set and utilized manual measurement techniques. The new classification system, which combines fuzzy logic and an artificial neural network, classified alarm calls correctly according to the eliciting predator species, achieving accuracy levels ranging from 78.6 to 96.3% on raw field data digitized with low quality audio equipment.

Journal Article↗
PubMed2006

Implementing tutoring strategies into a patient simulator for clinical reasoning learning.

OBJECTIVE: This paper describes an approach for developing intelligent tutoring systems (ITS) for teaching clinical reasoning. MATERIALS AND METHODS: Our approach to ITS for clinical reasoning uses a novel hybrid knowledge representation for the pedagogic model, combining finite state machines to model different phases in the diagnostic process, production rules to model triggering conditions for feedback in different phases, temporal logic to express triggering conditions based upon past states of the student's problem solving trace, and finite state machines to model feedback dialogues between the student and TeachMed. The expert model is represented by an influence diagram capturing the relationship between evidence and hypotheses related to a clinical case. RESULTS: This approach is implemented into TeachMed, a patient simulator we are developing to support clinical reasoning learning for a problem-based learning medical curriculum at our institution; we demonstrate some scenarios of tutoring feedback generated using this approach. CONCLUSION: Each of the knowledge representation formalisms that we use has already been proven successful in different applications of artificial intelligence and software engineering, but their integration into a coherent pedagogic model as we propose is unique. The examples we discuss illustrate the effectiveness of this approach, making it promising for the development of complex ITS, not only for clinical reasoning learning, but potentially for other domains as well.

Computer Simulation↗
PubMed1998

Detection of lung injury with conventional and neural network-based analysis of continuous data.

OBJECTIVE: To test if analysis of pressure and flow waveform patterns with an artificial intelligence neural network could distinguish between normal and injured lungs. METHODS: Acute lung injury was induced in ten healthy anesthetized, mechanically ventilated dogs with repeated injections of oleic acid, until arterial blood oxyhemoglobin saturation reached 85% breathing room air. Airway pressure, esophageal pressure, airway flow, and arterial and mixed venous saturation signals were stored at 2 min intervals. Hemodynamic and blood gas data were collected every 10 min. Back-propagation neural networks were trained with normalized airway pressure and flow waveforms from normal and fully injured lungs. RESULTS: The networks scored lung injury on a continuous scale from +1 (normal) to -1 (injured). Network scores unequivocally distinguished between normal and fully injured lungs and suggested a gradual transition from normal to injury pattern. However, the response of the network was slow compared to compliance, resistance and venous admixture. CONCLUSIONS: Normal and fully injured lungs display distinct flow and pressure waveform patterns which are independent of changes in calculated pulmonary mechanics variables. These patterns can be recognized by a neural network. Further research is needed to determine the full potential of automated pattern recognition for lung monitoring.

Animals↗
PubMed2025

Cost-Effectiveness Analysis of 3D Total-Body Photography for People at High Risk of Melanoma.

IMPORTANCE: Greater use of novel digital technologies could be associated with improved health outcomes and save health care costs by detecting smaller melanomas earlier (needing less treatment) or benign tumors (needing no treatment). OBJECTIVE: To compare costs and health effects of 3-dimensional (3D) total-body photography (TBP) and sequential digital dermoscopy imaging (SDDI) vs usual care for early detection of melanoma. DESIGN, SETTING, AND PARTICIPANTS: This prespecified cost-effectiveness analysis using randomized clinical trial (n = 309) data with 2 years of follow-up was conducted at a research hospital in Brisbane, Australia, and took a health system perspective. It included adults 18 years or older at high risk of developing a primary or subsequent melanoma. INTERVENTION: The intervention group received usual care plus clinical skin examinations by junior clinicians at baseline and 6, 12, 18, and 24 months with 3D TBP-SDDI reviewed by a teledermatologist. The control group continued to receive usual care and completed online surveys every 6 months. MAIN OUTCOMES AND MEASURES: Government health care costs, patient out-of-pocket costs, numbers of benign and malignant skin tumor excisions, and quality-adjusted life-years. Skin biopsy, excisions, pathology, and their costs were collected using administrative claims data. Quality of life was collected using the EuroQol-5D-5L. RESULTS: The trial included 314 participants (mean [SD] age, 51.6 [12.8] years; 194 female individuals [62%]) who completed all of the study procedures (158 in the intervention and 156 in the control groups). Compared with controls, intervention group participants had fewer melanoma excisions, more keratinocyte carcinomas and benign excisions, and more biopsy specimens. Over 24 months, mean per-person costs (analyzed in Australian dollars and converted to US$) for the intervention group were $1708 (95% CI, $1455-$1961) vs $763 (95% CI, $655-$870) for controls, an incremental cost of $945 (95% CI, $738-$1157) to provide the intervention. Total quality-adjusted life-years per person were similar for the intervention (1.84; 95% CI, 1.82-1.86) and control groups (1.84; 95% CI, 1.83-1.86). The incremental cost per additional malignant skin tumor excised was $40 (95% CI, $34-$48). CONCLUSIONS AND RELEVANCE: Over 2 years of the trial, the 3D TBP-SDDI model by junior clinicians and teledermatologist review generated higher costs and detected similar numbers of malignant tumors than usual care in a high-risk melanoma cohort. Cost-effectiveness is a necessary but not sufficient consideration for implementation. Other benefits of 3D TBP-SDDI may arise once artificial intelligence clinician support systems are integrated, and more research is needed to understand factors associated with costs and whether there are other benefits of 3D TBP-SDDI.

Adult↗
PubMed1997

Discrimination of fibroblast subtypes by multivariate analysis of gene expression.

Fibroblasts and myofibroblasts from normal, fibrotic or tumoral breast tissues present multiple quantitative differences in gene expression even when grown in isolation. We were therefore prompted to investigate whether one could recognize various subtypes by their constitutive-gene expression profile. Quantitative autoradiographic data for 34 constitutively expressed transcripts were submitted to multivariate analysis of variance, followed by discriminant analysis and single linkage cluster analysis. Models assuming up to 8 putative fibroblast subtypes (among fibroblasts or myofibroblasts from breast skin, normal mammary stroma, tumor-adjacent "normal" stroma, post-radiation fibrosis lesions and benign or malignant tumors) and an epithelial-cell group used as an internal control resulted in 100% correct classification. Myofibroblasts from various origins clustered close to, although distinctly apart from, their corresponding alpha-smooth-muscle-actin-negative counterparts. Malignant tumor fibroblasts were phenotypically more distant from normal cells compared with other pathological types. Our results support the hypothesis of co-adaptive transformation of stromal and epithelial tissues during breast tumoral development and suggest that different types of fibroblasts give rise to different types of myofibroblasts. Discriminant analysis of quantitative molecular variation may be considered for the development of a powerful artificial-intelligence method for cell typing and should be particularly useful when no reliable discrete molecular markers are available.

Breast↗
PubMed2000

Classification of human ovarian tumors using multivariate data analysis of polypeptide expression patterns.

Large amounts of data on quantitative gene expression are generated by procedures such as 2-DE analysis of proteins or cDNA microarrays. Quantitative molecular variation may potentially be used for the development of methods for the classification of tumors. We used here the statistical concepts of principal components analysis (PCA) and partial least square analysis (PLS) in an attempt to type ovarian tumors. Using a set of 170 polypeptides, 22 tumors were used to establish a model ("learning set") for classification into 3 groups (benign/borderline/malignant). Eighteen tumors were then used to test the model. Six of 8 carcinomas and 3 of 4 borderline tumors were correctly classified. Two of 6 benign lesions were correctly classified, 3 were classified as borderline and 1 as carcinoma. We conclude that it may be possible to classify tumors according to their constitutive protein expression profile using multivariate analysis, thus making classification by artificial intelligence a future possibility.

Breast Neoplasms↗
PubMed1998

Flow cytometry in diagnostic cytology.

Flow cytometry (FCM) is a useful adjunct to cytologic examination, because the quantitative biochemical information it provides complements the morphologic information gained during visual examination. It aids in the interpretation of bladder washings, and is particularly useful for the assessment of lymphoid lesions, whether they originate from fine-needle aspiration, cerebrospinal fluid, or effusions. Optimal use of FCM frequently requires assessment of more than one parameter; simultaneous use of cell differentiation markers and nuclear DNA quantitation is often significantly more useful than either alone. Despite the utility of FCM, however, the potential for future development appears to be limited. Improvements in image cytometry allow reasonable assessment of ploidy and S-fraction to be made from specimens prepared on glass slides. Multiparameter measurements may also be accomplished with imaging techniques, which allow the further advantage of visual identification of cells with equivocal morphologic changes. The development of artificial intelligence methods for use with imaging technology has also significantly exceeded that of FCM. Finally, image cytometry is often more useful for samples with few cells. Other challenges are posed by immunocytochemical methods which compete with flow cytometry as tools for assessment of proliferation. Given the relatively high cost of FCM instrumentation, survival of FCM as an ancillary technique in cytopathology will require further technical refinements to offset the advantages currently associated with image cytometry and immunocytochemistry.

Biopsy, Needle↗
PubMed1998

PAPNET computer-aided rescreening for detection of benign and malignant glandular elements in cervicovaginal smears: a review of 61 cases.

An increasing proportion of the recent cytodiagnostic literature has focused on automation of the Pap smear screening process in hopes of finding a feasible system to aid in the reduction of the number of reported false-negative cases. In a sense, these systems can be thought of as computer-driven sensitivity enhancers for better detection of abnormalities in smeared cervicovaginal specimens. The PAPNET system (Neuromedical Systems, Inc., Suffern, NY) relies on a neural network of artificial-intelligence technology to recognize the complex cellular arrays present in Pap smears, and was originally intended to aid in the identification of morphologically abnormal cells of squamous origin. Herein, we present the results of 61 smears containing a mixture of known diagnostically important benign, dysplastic, and malignant glandular cellular abnormalities which were reviewed by the PAPNET technology. The PAPNET system detected the diagnostic glandular material in 44 of the 45 benign cases reviewed (98% detection rate). In addition, the PAPNET technology identified abnormal cellular material in 15 of the 16 studied smears from patients with malignant/dysplastic morphology (94% detection rate). These data indicate that the PAPNET neural networks are capable of detecting cells with aberrant glandular cytomorphology. In both cases missed by the PAPNET system, the number of abnormal cells per slide was very low, indicating that as with human screeners, the capabilities of this semiautomated method may be exceeded when an extreme paucity of diagnostic cellular material is present in a given slide. Further and larger reviews of glandular abnormalities by automated technologies are needed to assess these systems for their true efficacy at diminishing false-negative cases.

Adult↗
PubMed1994

The development, validation, and applications of a computerized measurement of cognitive impairment from the content analysis of verbal behavior.

This is a report of the development, validation, and computerized applications of a method of measuring cognitive and intellectual impairment through the content analysis of verbal behavior. The content analysis procedure utilized is based on the method developed by Gottschalk and Gleser for the measurement of the magnitude of many other psychological states and traits, in addition to cognitive dysfunction. Verbal behavior studies are reviewed that examine the cognitive effects of age, certain psychoactive drugs, alcohol, total body irradiation, sensory overload, and dementia. Finally, the availability of a recently developed artificial intelligence software program is reported that will reliably, rapidly, and objectively score speech samples (on Gottschalk-Gleser Scales) transcribed according to specific directions from IBM-compatible computer diskettes.

Adolescent↗
PubMed1991

The structure of expert diagnostic knowledge in occupational medicine.

Development of an artificial intelligence expert system for diagnosing occupational lung disease requires explicit specification of the structure of knowledge necessary in clinical occupational medicine independent of the process by which the knowledge is utilized. Furthermore, explicit recognition of sources of uncertainty is necessary. Seven categories of knowledge define the diagnostic knowledge base in occupational pulmonary medicine. These include four objects (jobs, industries, exposures, and diseases) and three relationships between pairs of objects. This analysis demonstrates some of the unique aspects of occupational medicine expertise.

Algorithms↗
PubMed2002

Anatomical informatics: Millennial perspectives on a newer frontier.

One of the most ancient of sciences, anatomy has evolved over many centuries. Its methods have progressively encompassed dissection instruments, manual illustration, stains, microscopes, cameras and photography, and digital imaging systems. Like many other more modern scientific disciplines in the late 20th century, anatomy has also benefited from the revolutionary development of digital computers and their automated information management and analytical capabilities. By using newer methods of computer and information sciences, anatomists have made outstanding contributions to science, medicine, and education. In that regard, there is a strong rationale for recognizing anatomical informatics as a proper subdiscipline of anatomy. A high-level survey of the field reveals important anatomical applications of computer sciences methods in imaging, image processing and visualization, virtual reality, modeling and simulation, structural database processing, networking, and artificial intelligence. Within this framework, computational anatomy is a developing field focusing on data-driven mathematical models of bodily structures. Mastering such computer sciences and informatics methods is crucial for new anatomists, who will shape the future in research, clinical knowledge, and teaching.

Anatomy↗