PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Decision Making, Computer-Assisted”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 91 records · Page 5Linked to original sources
PubMed1975

Computerization of the surgical intensive care unit: improvement of patient care via education.

For the past 18 months we have been evaluating and developing a computerized patient-monitoring system in our surgical intensive care unit. Despite the enormous potential for use of such systems, we have been impressed with its underutilization and its failure to yield objective improvement in patient care at our institution and many others. The output of the system was ignored because the decision maker was unable or unwilling to integrate the more sophisticated data presented to him. The computer was relegated to the position of "redundant secretary". In an attempt to reverse this situation, we have developed a multilevel, multigoal educational system employing the computer. We have implemented brief educational programs for use by all unit personnel to explain deviant monitored variables. Given a physiologic subsystem and a particular variable, personnel can: (1) inquire whether or not the variable is deviant; (2) obtain a list of probable causes for the deviation; (3) obtain an explanation of the pathophysiology of particular deviants as well as instruction on how to identify a most probable cause; and (4) inquire how to correct specific deviants. When we monitored the system utilization after implementation of the educational programs, we found all of the system had improved utilization. As a result we have a better educated staff who communicate more effectively, deal with more sophisticated information, and make better decisions with resultant improved patient care. Additionally, the staff is eager to help improve the system. We believe the full potential of such systems can be obtained only through education.

California↗
PubMed1992

Evaluation of the incremental diagnostic value and impact on patient treatment of thallium scintigraphy.

The incremental diagnostic yield of exercise 201Tl scintigraphy with visual and quantitative analysis was determined in 191 patients with known or suspected coronary artery disease (CAD). The coronary arteriogram was used as the gold standard. After pre-test clinical and exercise electrocardiographic data were taken into consideration, scintigraphy was found to have additional diagnostic value both in the diagnosis of CAD and of multivessel disease, with quantitative analysis being superior to visual analysis. The impact of 201Tl scintigraphy on the patient's treatment--conservative treatment versus revascularization--was also evaluated. The impact was relatively low, as the decision for revascularization was based primarily on the angiographic result and the severity of the anginal pain. This result reflects only the decision making process used in our clinic and permits no conclusion to be made concerning the possible value of 201Tl scintigraphy in this type of medical decision making process.

Coronary Angiography↗
PubMedDate not supplied

A computerized analysis system for vigilance studies.

A digital signal analysis system for vigilance studies is presented. The analysis is based on adaptive segmentation, band pass filtering, nonlinear eye movement detection and rule-based decision making. A preliminary evaluation of seven subjects falling asleep showed that the system is able to detect small vigilance fluctuations reliably.

Artifacts↗
PubMed1979

Quality assessment and the art of medicine: the anatomy of laceration care.

Assuring high quality medical care has remained an elusive goal because of several problems which have hampered development of effective medical audit programs: inadequate patient data, unreasonable evaluative criteria and insensitive audit procedures. The present study demonstrates the use of a clinical algorithm to help overcome these problems. An examination of medical record data from a series of 703 laceration patients treated in an emergency service yielded only 27 cases (4 per cent) with medical records sufficiently complete to use for auditing physician compliance with algorithmic criteria. Substituting a structured checklist for the handwritten note increased this rate to 86 per cent. A computer-assisted branching audit of 1,400 laceration cases demonstrated that 1) physician compliance with an algorithmic instruction varied significantly (p less than .001) according to the specific instruction, and 2) compliance with a given instruction varied significantly (p less than .001) across different providers. These results underscore the need for medical audit with educational feedback which is provider specific.

Connecticut↗
PubMed1992

Bayesian belief networks in quantitative histopathology.

Bayesian belief networks have a dynamic range and numeric response characteristics that make them uniquely suitable for descriptive classification schemes. Features showing considerable overlap of tolerance regions may be used, in a cumulative manner, to derive unequivocal classification decisions. The numeric response characteristics of Bayesian belief networks are analyzed, and their application as control modules in automated scene segmentation in histopathology is demonstrated.

Bayes Theorem↗
PubMed1992

On-line measurement in biotechnology: exploitation, objectives and benefits.

Sound data biologically relevant are prerequisites when developing high-performance bioprocesses. Understanding of physiological regulation as well as sophisticated control strategies are highly dependent on the observability of the culture, i.e. the generation and exploitation of suited signals even under complex environmental measurement conditions. Against this background, the increasing number of analytical systems is very supportive and, accordingly, an appropriate handling of sensors and measured data is of decisive importance. This article reports on practical experience with routines for maintenance, service and calibration of hardware sensors which improve the quality of measurements significantly. Verification and validation of signals is outlined in order to make the value of data exploitation tools obvious. A method for the characterization of information is introduced by practical examples of Saccharomyces cerevisiae cultures when explaining the specific properties of extracting biological information from raw data. Finally, examples for advantageous exploitation of on-line data are given.

Biosensing Techniques↗
PubMed1976

[User-oriented programs for semi-automated evaluation of radio in vitro tests].

Most of the programmes for the evaluation of radio in vitro tests proceed from the anticipation that the best method should give an approximation of standard values by a curve as perfect as possible. According to our experiences this demand, however, is not decisive for a good standard curve, as in principle all standard values can be incorrect. The application of relatively simple linearising transformations and an additional curve (recovery of a normal serum) guarantees the necessary precision in the programme described. After a short description of the contents of the system, the programme for the assessment of CPBA - methods is shown, by help of which the sample changer - calculator-system determines the absolute concentration of the substance to be measured. This makes the starting point for the more complex RIA programme which methodologically takes into consideration the special problems of these tests.

Diagnosis, Computer-Assisted↗
PubMed1992

PADS (Patient Archiving and Documentation System): a computerized patient record with educational aspects.

Rapid acquisition and analysis of information in an Intensive Care Unit (ICU) setting is essential, even more so the documentation of the decision making process which has vital consequences for the lives of ICU patients. We describe an Ethernet based local area network (LAN) with clinical workstations (Macintosh fx, ci). Our Patient Archiving and Documentation System (PADS) represents a computerized patient record presently used in a university hospitals' ICU. Taking full advantage of the Macintosh based graphical user interface (GUI) our system enables nurses and doctors to perform the following tasks: admission, medical history taking, physical examination, generation of problem lists and follow up notes, access to laboratory data and reports, semiautomatic generation of a discharge summary including full word processor capabilities. Furthermore, the system offers rapid, consistent and complete automatic encoding of diagnoses following the International Classification of Disease (ICD; WHO, [1]). For educational purposes the user can also view disease entities or complications related to the diagnoses she/he encoded. The system has links to other educational programs such as cardiac auscultation. A MEDLINE literature search through a CD-ROM based system can be performed without exiting the system; also, CD-ROM based medical textbooks can be accessed as well. Commercially available Macintosh programs can be integrated in the system without existing the main program thus enabling users to customize their working environment. Additional options include automatic background monitoring of users learning behavior, analyses and graphical display of numerous epidemiological and health care related problems. Furthermore, we are in the process of integrating sound and digital video in our system. This system represents one in a line of modular departmental models which will eventually be integrated to form a decentralized Hospital Information System (HIS).

Computer User Training↗
PubMed1975

[A system to aid decision-making. Application to automatic interpretation of vectorcardiograms].

This paper presents a complete and autonomous system for the automated diagnosis (heuristic approach). The system was worked out by means of a small computer. A flexible, evolutive, quasi-universal system is achieved through original procedures. A specialised language enables the users to describe diagnoses and their criteria in symbolic form. A suitable compiler translates this symbolic writing into an interpretable object program. This, with the Interpreter program, constitutes the 'Automated Diagnosis Program'. Our data are vectorcardiograms recorded according to the Frank orthogonal system. After an interactive pre-processing process, 180 parameters--mostly spatial--are computed. The data, the computed parameters and additional information are stored in a data bank. Finally, the medical interpretation is automatically selected from 125 possibilities. The user could also utilise a data bank interrogation language.

Computers↗
PubMed2026

A multi-scale fusion model based on multi-phase contrast-enhanced CT for predicting pancreatic cancer resectability.

Purpose.Develop a multi-scale fusion model (MSFM) based on multi-phase contrast-enhanced computed tomography (CECT) to predict pancreatic cancer (PC) resectability, thereby assisting expert decision-making.Methods.This retrospective study enrolled 280 patients with PC from four institutions, which were randomly divided into a training cohort (202 patients) and an independent test cohort (78 patients). Three-phase CECT images (arterial, venous, and delayed phases) were used for modeling. The MSFM comprises two sub-networks: (1) a multi-phase fusion network for extracting cross-phase shared fusion features, (2) a phase-specific branch network for capturing phase-specific features; and a post-fusion strategy to generate the final predictive score by integrating the shared fusion features and three groups of phase-specific features. Additionally, a human-machine fusion deep learning model (HMfDL) was constructed by fusing the predictive score of the MSFM with expert assessments.Results.In the independent test, the MSFM achieved an AUC (area under the receiver operating characteristic curve) of 0.8385 (95% CI: 0.7521-0.9249), accuracy of 84.62%, sensitivity of 72.00%, and specificity of 90.57%. This performance outperformed single-phase models (AUC range: 0.7638-0.7781), two-phase models (AUC range: 0.7826-0.7864), and ten states-of-the-art classifiers (AUC range: 0.7404-0.7796). The HMfDL further improved the performance, reaching an AUC of 0.8626 (95% CI: 0.7853-0.9400), accuracy of 91.03%, sensitivity of 80.00%, and specificity of 96.23%. Notably, the HMfDL corrected 58.82% of misdiagnosis made by experts.Conclusions. The MSFM effectively fuses multi-phase CECT to enable highly accurate predictions of PC resectability, and provides valuable support for expert decision-making through HMfDL.

Humans↗
PubMedDate not supplied

Towards computer analysis of pulmonary infiltration.

A feasibility study is described to provide quantitative texture measures to distinguish between normal lung, alveolar infiltrates and interstitial infiltrates. Advanced computer imaging technology and decision making processes were applied to distinguish between these textural patterns. The results, based on computer extracted quantitative measures, show an excellent separation of the three classes considered with 95% accuracy in the training phase and 90% accuracy in the testing phase.

Diagnosis, Computer-Assisted↗
PubMed1979

An approach to quality and performance control in a computer-assisted clinical chemistry laboratory.

A locally developed, computer-based clinical chemistry laboratory system has been in operation since 1970. This utilises a Digital Equipment Co Ltd PDP 12 and an interconnected PDP 8/F computer. Details are presented of the performance and quality control techniques incorporated into the system. Laboratory performance is assessed through analysis of results from fixed-level control sera as well as from cumulative sum methods. At a simple level the presentation may be considered purely indicative, while at a more sophisticated level statistical concepts have been introduced to aid the laboratory controller in decision-making processes.

Chemistry, Clinical↗
PubMed1975

On the evolution of the physiological model.

Most of us who have concerned ourselves with models can perceive outlines like those above to catalog the future evolution of the expository function of models. In the context of a single class of computerized mathematical models of respiratory physiology, we can observe at once the burgeoning interest among scientists, and the similarities between model activity and the general organization of scientific information for use. Although physiological models have become quite advanced in their subject control, there is relatively little coordinated activity in the mechanization of the purposes and philosophical potential of automata. The outlines, however, are visible. An assiduous pursuit of the notion of "explanation" by machine is a major evolutionary step next to occur. It appears to us that various diagrams similar to Figures 5 or 6 can be created and investigated in terms of their relation to the human mind and in terms of formalizing rules for traversing from one plane to the next. The evolution of models will require program-making programs which can decide when and how to aggregate for deductive inference, and how far to penetrate top-down for explanation. The rules for identifying "second order" effects must be established. The decision to ignore or use these rules will be crucial. These are the means whereby the systems are traversed from plane to plane. In a word, models need to synthesize the means to ignore, "forget," and gloss over; only then will we have useful tools for taking informed action in physiology, diagnosis in medicine, or the writing of "scholarly" reviews.

Computers↗
PubMed2026

Automated Classification of Lymphoma Subtypes From Histopathological Images Using a U-Net Deep Learning Model: Comparative Evaluation Study.

BACKGROUND: Accurate classification and grading of lymphoma subtypes are essential for treatment planning. Traditional diagnostic methods face challenges of subjectivity and inefficiency, highlighting the need for automated solutions based on deep learning techniques. OBJECTIVE: This study aimed to investigate the application of deep learning technology, specifically the U-Net model, in classifying and grading lymphoma subtypes to enhance diagnostic precision and efficiency. METHODS: In this study, the U-Net model was used as the primary tool for image segmentation integrated with attention mechanisms and residual networks for feature extraction and classification. A total of 620 high-quality histopathological images representing 3 major lymphoma subtypes were collected from The Cancer Genome Atlas and the Cancer Imaging Archive. All images underwent standardized preprocessing, including Gaussian filtering for noise reduction, histogram equalization, and normalization. Data augmentation techniques such as rotation, flipping, and scaling were applied to improve the model's generalization capability. The dataset was divided into training (70%), validation (15%), and test (15%) subsets. Five-fold cross-validation was used to assess model robustness. Performance was benchmarked against mainstream convolutional neural network architectures, including fully convolutional network, SegNet, and DeepLabv3+. RESULTS: The U-Net model achieved high segmentation accuracy, effectively delineating lesion regions and improving the quality of input for classification and grading. The incorporation of attention mechanisms further improved the model's ability to extract key features, whereas the residual structure of the residual network enhanced classification accuracy for complex images. In the test set (N=1250), the proposed fusion model achieved an accuracy of 92% (1150/1250), a sensitivity of 91.04% (1138/1250), a specificity of 89.04% (1113/1250), and an F1-score of 90% (1125/1250) for the classification of the 3 lymphoma subtypes, with an area under the receiver operating characteristic curve of 0.95 (95% CI 0.93-0.97). The high sensitivity and specificity of the model indicate strong clinical applicability, particularly as an assistive diagnostic tool. CONCLUSIONS: Deep learning techniques based on the U-Net architecture offer considerable advantages in the automated classification and grading of lymphoma subtypes. The proposed model significantly improved diagnostic accuracy and accelerated pathological evaluation, providing efficient and precise support for clinical decision-making. Future work may focus on enhancing model robustness through integration with advanced algorithms and validating performance across multicenter clinical datasets. The model also holds promise for deployment in digital pathology platforms and artificial intelligence-assisted diagnostic workflows, improving screening efficiency and promoting consistency in pathological classification.

Humans↗
PubMedDate not supplied

Characteristics of the software for computer applications in medicine.

The requirements of clinical medicine which have tended to make the design and implementation of software for hospital computer systems more difficult than that elsewhere, are discussed in this paper. Specific constraints on the software for selected computer-assisted activities in a hospital environment are examined in considerable depth. It is shown that since some of these activities have counterparts elsewhere, hospital computing can benefit from the accumulated experience in dealing with similiar problems in business and scientific environments. The argument is put forward that developing countries, with their characteristic problem of acute shortage of skilled manpower in both medicine and computing, should initially concentrate on applying computers to these activities alone. Furthermore, medical education in such countries should incorporate programmes relating to computer technology in general and the software aspects in particular.

Computers↗
PubMedDate not supplied

Computer assisted instruction for preoperative and postoperative patient education in joint replacement surgery.

This article describes a comprehensive system for preoperative and postoperative patient education. The system offers a cost-effective method of instruction which encourages patient interaction and practice with decision making. The system was designed for patients undergoing total joint replacement surgery and includes two preoperative lessons, and a third lesson presented postoperatively at the bedside. The computer lessons were developed using data collected by a patient assessment instrument, and collaboratively with input from a nurse clinical specialist, orthopedic surgeon, physical therapist, and computer programmer. In this project, several advantages for using computer assisted instruction for preoperative and postoperative patient education were identified.

Computer-Assisted Instruction↗