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 1,711 records · Page 95Linked to original sources
PubMed2005

An intelligent adaptive control scheme for postsurgical blood pressure regulation.

This paper presents an adaptive modeling and control scheme for drug delivery systems based on a generalized fuzzy neural network (G-FNN). The proposed G-FNN is a novel intelligent modeling tool, which can model unknown nonlinearities of complex drug delivery systems and adapt to changes and uncertainties in these systems online. It offers salient features, such as dynamic fuzzy neural topology, fast online learning ability and adaptability. System approximation formulated by the G-FNN is employed in the adaptive controller design for drug infusion in intensive care environment. In particular, this paper investigates automated regulation of mean arterial pressure (MAP) through intravenous infusion of sodium nitroprusside (SNP), which is one attractive application in automation of drug delivery. Simulation studies demonstrate the capability of the proposed approach in estimating the drug's effect and regulating blood pressure at a prescribed level.

Artificial Intelligence↗
PubMed1993

European research efforts in medical knowledge-based systems.

This article describes the major projects going on in Europe in the field of Artificial Intelligence in Medicine. The important role of the Commission of the European Communities in providing the needed resources is stressed throughout the paper. Particular attention is given to the methodological and technological issues addressed by the European research teams, since the results which these teams accomplish are fundamental for a more extensive diffusion of knowledge-based systems in real medical settings. The variety of medical problems tackled shows that there is no field of medicine where the potential of advanced informatics technologies has not yet been assessed.

Artificial Intelligence↗
PubMed1996

Decision aids in radiology.

Computer systems can help radiologists decide which tests to perform and which diagnoses to consider. Repositories of clinical data and general medical knowledge can provide information on demand for decision-making tasks; many of these information resources are available remotely via the Internet. Decision support systems can incorporate the techniques of artificial intelligence to apply their general knowledge to the features of a particular patient. Successful use of these technologies requires careful attention to design, implementation, and rigorous evaluation. Computer-based decision aids can improve the cost effectiveness and diagnostic accuracy of radiologic practice and are poised to play an important role in the future of radiology.

Decision Support Techniques↗
PubMed1993

Exploratory data analysis of hyperlipidemia on the Macintosh: software tools for analysis of biochemical, clinical, and genetic variables in 1677 consecutive lipid clinic patients.

Exploratory data analysis (EDA) software facilitates unstructured, iterative open exploration of complex datasets with the aid of multiple linked graphical displays. We are investigating relationships between plasma lipoproteins and coronary artery disease by retrospective analysis of 1677 consecutive UCSF Lipid Clinic patients. Our preliminary experience is with Data Deck 3.0 although several additional software programs (JMP 2.0, Systat 5.1, Minitab 8.0, StatView 4.0) are mentioned. Lipid diagnosis (751 women and 925 men) was 22% primary hypercholesterolemia, 19% combined hyperlipidemia, 3% dysbetalipoproteinemia, 15% endogenous lipemia, 4% mixed lipemia, 5% elevated Lp(a) and 32% with no major lipid abnormality. We found the Macintosh platform (68030) to be flexible and powerful for analysis of moderate size (less than 1 Mb) clinical datasets. High resolution color monitors (1024 x 768 pixels), fast hard disks (< 18 msec) and moderate amounts of system memory (8 + Mb) facilitate exploratory analysis.

Artificial Intelligence↗
PubMed1996

Prism: a new approach to radiotherapy planning software.

PURPOSE: We describe the capabilities and performance of Prism, an innovative new radiotherapy planning system with unusual features and design. The design and implementation strategies are intended to assure high quality and clinical acceptability. The features include Artificial Intelligence tools and special support for multileaf collimator (MLC) systems. The design provides unusual flexibility of operation and ease of expansion. METHODS AND MATERIALS: We have implemented Prism, a three-dimensional (3D) radiotherapy treatment-planning system on standard commercial workstations with the widely available X window system. The design and implementation use ideas taken from recent software engineering research, for example, the use of behavioral entity-relationship modeling and the "Mediator Method" instead of ad-hoc programming. The Prism system includes the usual features of a 3D planning system, including Beam's Eye View and the ability to simulate any treatment geometry possible with any standard radiotherapy accelerator. It includes a rule-based expert system for automated generation of the planning target volume as defined in ICRU Report 50. In addition, it provides special support for planning treatments with a multileaf collimator (MLC). We also implemented a Radiotherapy Treatment Planning Tools Foundation for Prism, so that we are able to use software tools form other institutions without any source code modification. RESULTS: The Prism system has been in clinical operation at the University of Washington since July 1994 and has been installed at several other clinics. The system is run simultaneously by several users, each with their own workstation operating from a common networked database and software. In addition to the dosimetrists, the system is used by radiation oncologists to define tumor and target volumes and by radiation therapists to select treatment setups to load into a computer controlled accelerator. CONCLUSIONS: Experience with the installation and operation has shown the design to be effective as both a clinical and research tool. Integration of software tools has eased the development and significantly enhanced the clinical usability of the system. The design has been shown to be a sound basis for further innovation in radiation treatment planning software and for research in the treatment planning process.

Computer Communication Networks↗
PubMed1996

Consistency among speech parameter vectors: application to predicting speech intelligibility.

Previous researchers interested in physical assessment of speech intelligibility have largely based their predictions on preservation of spectral shape. A new approach is presented in which intelligibility is predicted to be preserved only if a transformation modifies relevant speech parameters in a consistent manner. In particular, the parameters from each short-time interval are described by one of a finite number of symbols formed by quantizing the output of an auditory model, and preservation of intelligibility is modeled as the extent to which a one-to-one correspondence exists between the symbols of the input to the transformation, and those of the output. In this paper, a consistency-measurement system is designed and applied to prediction of intelligibility of linearly filtered speech and speech degraded by additive noise. Results were obtained for two parameter sets: one consisting of band-energy values, and the other based on the ensemble interval histogram (EIH) model. Predictions within a class of transformation varied monotonically with the amount of degradation. Across classes of transformation, the predicted effect of additive-noise transformations was more severe than typical perceptual effects. With respect to the goal of achieving predictions that varied monotonically with human speech-perception scores, performance was slightly better with the EIH parameter set.

Auditory Threshold↗
PubMed2002

On application of adaptive decorrelation filtering to assistive listening.

This paper describes an application of the multichannel signal processing technique of adaptive decorrelation filtering to the design of an assistive listening system. A simulated "dinner table" scenario was studied. The speech signal of a desired talker was corrupted by three simultaneous speech jammers and by a speech-shaped diffusive noise. The technique of adaptive decorrelation filtering processing was used to extract the desired speech from the interference speech and noise. The effectiveness of the assistive listening system was evaluated by observing improvements in A-weighted signal-to-noise ratio (SNR) and in sentence intelligibility, where the latter was evaluated in a listening test with eight normal hearing subjects and three subjects with hearing impairments. Significant improvements in SNR and sentence intelligibility were achieved with the use of the assistive listening system. For subjects with normal hearing, the speech reception threshold was improved by 3 to 5 dBA, and for subjects with hearing impairments, the threshold was improved by 4 to 8 dBA.

Acoustic Stimulation↗
PubMed2026

Privacy, security, and reliability risks of artificial intelligence in healthcare: a systematic review of empirical evidence.

BACKGROUND: Artificial intelligence (AI) is increasingly integrated into healthcare information systems, supporting clinical decision-making, imaging analysis, and predictive modeling. While these applications offer operational and clinical benefits, they also introduce emerging risks to patient privacy, data security, and system reliability. OBJECTIVE: To systematically review empirical evidence on privacy breaches, security vulnerabilities, and misuse associated with AI applications in healthcare settings. METHODS: PubMed, Embase, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library were searched for empirical studies published between January 2015 and November 2025 that evaluated AI use or misuse in clinical diagnosis, treatment, or decision-making. Two reviewers independently screened studies and extracted data using a standardized form. Findings were synthesized narratively due to heterogeneity in study designs, AI methods, and reported outcomes. RESULTS: Of 7,285 records identified through database searches and 205 through citation screening, 22 empirical studies met the inclusion criteria, spanning multiple clinical domains and data modalities, predominantly medical imaging applications. Five recurring threat categories were identified: patient re-identification, membership inference, unauthorized access and adversarial exploitation, input manipulation, and misuse or overinterpretation of AI outputs. Across studies, AI models were shown to encode latent biometric signals across diverse data types, limiting the effectiveness of traditional anonymization and synthetic data approaches. Adversarial attacks and input manipulation were also shown to compromise diagnostic performance and system integrity. CONCLUSION: This systematic review provides empirical evidence suggesting that contemporary AI systems in healthcare introduce privacy and security risks that may challenge traditional assumptions about data protection. These findings underscore the need for privacy- and security-by-design approaches and governance frameworks that address risks across the AI lifecycle.

Humans↗
PubMed1987

Chaos in coupled optimizers.

Autonomous optimizers, a subclass of dynamical systems, arise on the intersection between artificial intelligence and deductive biology. Two simple optimizers coupled symmetrically à la Rashevsky give rise to chaotic attractors and (possibly) cloud attractors. More complicated autonomous optimizers (endowed with internal simulation) will be subject to more complicated interactional function changes. Humanistic ideas of Mead and Bateson may come within scientific reach. Conversely, the physical measurement problem may turn out to belong partly to the humanities.

Adaptation, Physiological↗
PubMed1994

Clinical course of cognitive dysfunction in systemic lupus erythematosus.

OBJECTIVE: To prospectively evaluate changes in cognitive function in a cohort of unselected patients with systemic lupus erythematosus (SLE) and controls over a 12 month period. METHODS: Seventy female patients with SLE, 25 patients with rheumatoid arthritis (RA) and 23 healthy subjects (age and sex matched) were evaluated using the Wechsler Adult Intelligence Scale-Revised (WAIS-R), the Wechsler Memory Scale-Revised (WMS-R), the California Verbal Learning Test (CVLT) and the National Adult Reading Test-Revised to identify impairment in 8 areas of cognitive function. Cumulative disease manifestations and current medications were documented, and disease activity was expressed using the SLE disease activity index (SLEDAI). Decision rules were determined for overall cognitive impairment. RESULTS: At baseline, 21% (15/70) of patients with SLE were impaired compared to 4% (1/25) of patients with RA and 4% (1/23) of healthy subjects (p = 0.042). After a mean interval of 12.8 months (range: 11-17) 84% (59/70) of patients with SLE, 44% (11/25) of patients with RA and 80% (17/23) of healthy subjects were reassessed. This included all subjects who were impaired at the initial assessment. Using the same decision rules as at baseline, 12% (7/59) of patients with SLE were impaired at followup compared to none of the patients with RA and healthy subjects. Over the period of study cognitive impairment persisted in 3 patients with SLE, resolved in 12 and evolved in 4 others. There was no apparent association between changes in cognitive function and concurrent changes in generalized disease activity, overt neuropsychiatric disease or corticosteroid medication. CONCLUSION: Our results suggest that cognitive dysfunction in patients with SLE is evanescent, does not necessarily lead to irreversible neurologic compromise and changes independently of other clinical variables.

Adult↗
PubMed2001

What we can really expect from telemedicine in intensive diabetes treatment: results from 3-year study on type 1 pregnant diabetic women.

Existing standards of the management of the diabetic patients are not efficient enough, and further improvement is needed. The major objective of this paper is to present and discuss the therapeutic effectiveness of an intensive care telematic system designed and applied for intensive treatment of pregnant type 1 diabetic women. The developed system operates automatically, every night transferring all the data recorded during the day in the patient's glucometer memory to a central clinical unit. In order to assess the efficiency of the designed and developed system, a 3-year randomized prospective clinical trial was conducted, using the study group and the control group, each consisting of 15 pregnant type 1 diabetic women. All patients were treated by the same diabetologist. In the presented analysis, two indices calculated weekly were used for the assessment of glycemic control: MBG represents mean blood glucose level, and the universal J-index is sensitive to the glycemic level and glycemic variations. The most important results from the study concern: (a) better glycemic control in the study group in comparison with the control group during the course of treatment, as assessed by the average differences of the MBG and J indices calculated weekly (n = 24) (deltaMBG = -3.2 +/- 4.3 mg/dL, p = 0.0016, deltaJ = -1.4 +/- 2.3, p = 0.0065); (b) much more similar results in glycemic control among members of the study group compared to each other, than among members of the control group compared to each other, as indicated by significantly lower variations of the applied glycemic control indices (SDMBG: 11.9 vs. 18.7 mg/dL, p = 0.0498; SDJ: 6.5 vs. 10.9, p = 0.0318); (c) the observed tendency of a better glycemic control for patients with a lower level of intelligence (IQ < 100) supported by the telematic system in comparison with all other assessed groups of patients. The last result was not statistically significant (p > 0.05). This telematic intensive care system improved the effectiveness of diabetes treatment during pregnancy. It also allows the diabetologist's strategy to be much more precise than if it were conducted without telematic support. This telematic system is inexpensive and simple in use.

Adult↗
PubMed2003

A meta-data model for knowledge in decision support systems.

Clinical decision support such as alerts, reminders and guidance are driven by rules often distributed among a variety of applications in a healthcare information system. Due to the increasing size of rule bases, there is a growing need to manage this dispersed knowledge in an integrated environment. A system for management of executable clinical knowledge such as rules should (1) assist in the development and maintenance of rules throughout the rules' life-cycles, (2) support search and retrieval of rules in the knowledge base (e.g., rules for diabetes, rules created by a particular individual), and (3) facilitate the analyses of rules in the knowledge base (e.g., identify rules not updated in the last year). In order to create such a clinical knowledge management system it is necessary to model the meta-data of rules. There have been efforts to document meta-data about rules within the Arden Syntax Medical Logical Modules' project. However, the maintenance and library categories in that project allow mainly free-text information about a rule. We have created a comprehensive meta-data structure and taxonomy for describing clinical rules that supports the features of a knowledge management system. We also tested this model using a representative set of rules.

Artificial Intelligence↗
PubMed1993

Assessment of cognitive function in patients with systemic lupus erythematosus.

The spectrum of central nervous system manifestations of systemic lupus erythematosus (SLE) is very broad and has been found to include subtle subclinical cognitive dysfunction which may be detected only by the lengthy process of detailed neuropsychological evaluation. This study reports the value of estimating premorbid intelligence as a simple yet effective means of screening for subclinical cognitive dysfunction. Twenty one female patients with clinically quiescent SLE underwent neuropsychological examination at entry to the study. In all patients, this examination included measurement of full-scale intelligence quotient (IQ), verbal and performance IQ as well as verbal and visual memories. In addition, premorbid intelligence was estimated using the Schonell graded word reading test. Nine patients (43%) gave a history of neuropsychiatric (NP) disease. No difference was identified between the results of the neuropsychological evaluation in these 9 patients and in either the other SLE patients or in age and sex matched control patients. Sixteen patients were re-evaluated 1 year later. A comparison of measured full-scale IQ with the estimated premorbid intelligence identified a subgroup of 3 patients who demonstrated a significant reduction in intelligence. Unlike the other 13 patients, these 3 patients had multiple (3 or more) other features of cognitive impairment.

Cognition Disorders↗
PubMed1999

Integration of clinical decision support with on-line encounter documentation for well child care at the point of care.

Electronic medical record systems and clinical practice guideline (CPG) support applications are emerging in the clinical environment to document and support care. Applications which integrate online documentation with CPG are often complex systems bound to a proprietary infrastructure and as such, can be difficult to adapt to changing care guidelines. This paper describes integration of point-of-care clinical documentation to an Internet-based CPG system that was easily modified, utilized available software resources, and separated patient information from CPG. The system combined a text-based encounter documentation tool, Inbox, with a web-based CPG system, SIEGFRIED (System for Interactive Electronic Guidelines with Feedback and Resources for Instructional and Educational Development), which interactively presented care guidelines to providers. Age-specific well child care documentation templates were developed using Inbox for point-of-care documentation. SIEGFRIED contained the knowledge base of child safety education guidelines and executed independent of the program presenting the guidelines. The CPG were accessed from within the documentation template via an Internet hyperlink. Patient chart evaluation indicated that 77% of safety topics were reviewed and 32% of the charts contained documentation indicating all the safety topics were reviewed. Last, routine use of the Inbox-SIEGFRIED system was not realized due to the clinical time constraints and workload of the medical providers, and lack of data entry experience. A user survey indicated time cost (network access and software execution) were negative aspects of the system. However, the system function was highly regarded and the Internet-based patient education materials were described as useful and accurate. In summary, the system was functional, met original development goals, and provided valuable patient education materials; however, routine system use was prevented by time requirements. We recommend further development be oriented towards integrating the identified beneficial components of the system into clinician workflow.

Accident Prevention↗
PubMed1995

A client/server system for remote diagnosis of cardiac arrhythmias.

Health care practitioners are often faced with the task of interpreting complex heart rhythms from electrocardiograms (ECGs) produced by 12-lead ECG machines, ambulatory (Holter) monitoring systems, and intensive-care unit monitors. Usually, the practitioner caring for the patient does not have specialized training in cardiology or in ECG interpretation; and commercial programs that interpret 12-lead ECGs have been well-documented in the medical literature to perform poorly at analyzing cardiac rhythm. We believe that a system capable of providing comprehensive ECG interpretation as well as access to online consultations will be beneficial to the health care system. We hypothesized that we could develop a client-server based telemedicine system capable of providing access to (1) an on-line knowledge-based system for remote diagnosis of cardiac arrhythmias and (2) an on-line cardiologist for real-time interactive consultation using readily available resources on the Internet. Furthermore, we hypothesized that Macintosh and Microsoft Windows-based personal computers running an X server could function as the delivery platform for the developed system. Although we were successful in developing such a system that will run efficiently on a UNIX-based work-station, current personal computer X server software are not capable of running the system efficiently.

Arrhythmias, Cardiac↗
PubMed2000

Intellectual systems for differential diagnostics within groups of hardly distinguished diseases.

A new method of mathematical modeling based on ideas of the artificial intelligence has been developed called as a method of a "mosaic portrait". A description of a universal computer system "Differentiated diagnostics" is given into which it is possible to introduce the "mosaic model" for any group of diseases which are difficult to distinguish. On its base a number of diagnostic medical intellectual systems have been developed.

Artificial Intelligence↗
PubMed1963

SKELETAL AGE IN SUBJECTS WITH MENTAL RETARDATION.

Delayed skeletal maturation has been long accepted as a correlate of mental retardation. Three hundred mentally retarded children were examined for skeletal age and failed to show any overall delay of bone age as compared to chronological age. Further, birth weight was found to be unrelated to bone age maturity. Children with severe levels of retardation displayed significant delay in bone development only when profound physical disability was also present. In eight diagnostic categories of mental retardation, only those with metabolic syndromes showed any significant delay in skeletal system maturation. Conversely, mongoloid children showed a larger-than-chance percentage of advanced bone maturation. Skeletal measures therefore are perhaps more reflective of etiological states than of diagnostic classifications in mental retardation.

Age Determination by Skeleton↗