SEARCH · PubMed Health

Results for “Algorithm”

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 307 records · Page 17Linked to original sources

PubMed2000

Comparing the performance of mammographic enhancement algorithms: a preference study.

OBJECTIVE: The objective of this study was to compare the performance of four image enhancement algorithms on secondarily digitized (i.e., digitized from film) mammograms containing masses and microcalcifications of known pathology in a clinical soft-copy display setting. MATERIALS AND METHODS: Four different image processing algorithms (adaptive unsharp masking, contrast-limited adaptive histogram equalization, adaptive neighborhood contrast enhancement, and wavelet-based enhancement) were applied to one image of secondarily digitized mammograms of forty cases (10 each of benign and malignant masses and 10 each of benign and malignant microcalcifications). The four enhanced images and the one unenhanced image were displayed randomly across three high-resolution monitors. Four expert mammographers ranked the unenhanced and the four enhanced images from 1 (best) to 5 (worst). RESULTS: For microcalcifications, the adaptive neighborhood contrast enhancement algorithm was the most preferred in 49% of the interpretations, the wavelet-based enhancement in 28%, and the unenhanced image in 13%. For masses, the unenhanced image was the most preferred in 58% of cases, followed by the unsharp masking algorithm (28%). CONCLUSION: Appropriate image enhancement improves the visibility of microcalcifications. Among the different algorithms, the adaptive neighborhood contrast enhancement algorithm was preferred most often. For masses, no significant improvement was observed with any of these image processing approaches compared with the unenhanced image. Different image processing approaches may need to be used, depending on the type of lesion. This study has implications for the practice of digital mammography.

Aged↗

PubMedDate not supplied

Evaluation of shading algorithms for surface display: depth information, surface orientation, colour and transparency.

When interpreting a shaded surface image the observer must infer a three-dimensional shape from a two-dimensional image. The principal monocular cues used in this process are surface orientation and apparent depth. Four shading algorithms were compared, ranging from pure depth to pure surface orientation with two intermediates. Eight observers assessed these algorithms using 40 sets of images showing test objects derived from emission computed axial tomograms, X-ray computed tomograms and simulated data. The results demonstrated a highly significant preference (p less than 0.01) for surface orientation over depth information for all observers and both imaging modalities. The coefficient of concordance showed that the observers were in good agreement as to the rank order of the algorithms, with significant agreement (p less than 0.05) for 37 of the 40 sets of images. The overall preferred algorithm was based on a local polynomial fitting procedure and contained primarily surface information. This shading algorithm was extended to include colour, which was used both as an arbitrary surface property to identify parts of a complex object and as a means of conveying temporal information. Further extension of the algorithm to the display of transparent surfaces was facilitated by an illumination model based on purely isotropic light. This enabled even irregular surfaces to be displayed as transparent objects, and was combined with opaque shading for displaying nested surfaces in nuclear magnetic resonance data.

Algorithms↗

PubMed1998

Evaluation of syndromic patient management algorithm for urethral discharge.

OBJECTIVE: To determine feasibility, validity, and cost effectiveness of the syndromic approach to male patients with urethral discharge in Bandung, Indonesia. METHODS: The WHO algorithm on urethral discharge with no microscopy available was evaluated. Patients presented with a complaint of urethral discharge and if discharge was confirmed the algorithm was applied. Treatment covered gonococcal and chlamydial infection (ciprofloxacin 500 mg single oral dose plus doxycycline 100 mg, twice daily orally for 7 days). The gold standard for validation was gonococcal culture and chlamydia antigen detection. RESULTS: 140 male patients with a complaint of urethral discharge were enrolled; 119 had confirmed discharge and entered the decision tree: 107 were followed and 104 (97%) were clinically cured. Of the three patients with persistent discharge, one had a purulent urethral discharge, diagnosed as gonococcal urethritis and he was probably reinfected; two patients had a serous discharge and microbiological tests were negative. Overall, 106 out of 107 patients (99%) were microbiologically cured. Sensitivity of the algorithm is 100% and its positive predictive value (PPV) is 75% or 97% if validated against gold standard microbiological tests or Gram stain, respectively. Cost per patient is rupiah (Rp)5.894 ($US2.56) for the algorithm compared with Rp43.024 ($18.70) for full microbiological diagnosis. The cost estimate for an algorithm of urethral discharge with microscopy available is Rp6.432 ($2.80) CONCLUSION: The "symptom and sign" algorithm is fully adapted to the prevailing situation in primary healthcare settings, is acceptable to healthcare workers and patients (who are effectively treated at their first visit), is highly cost effective, is 100% sensitive (no false negatives, which is not the case with microbiological diagnosis), and has a high PPV, between 75% and 97%. It is an excellent patient management tool and a sound basis for partner notification so that it should have a major impact on STD/HIV control and prevention in both men and women.

Administration, Oral↗

PubMed1999

Do management algorithms improve chest pain triage?

OBJECTIVE: To audit the use of management algorithms for chest pain in an emergency department. DESIGN AND SETTING: Prospective study of all patients with chest pain presenting to the emergency department of an urban teaching hospital between 12 January and 4 May 1997. Staff were asked to complete a standardised admission form that incorporated the risk stratification algorithms for managing patients with suspected acute coronary syndrome. MAIN OUTCOME MEASURES: Compliance with the use of management algorithms; concordance with a cardiologist's review of the triage grouping and admission/discharge decision; and major cardiovascular events over four months. RESULTS: Emergency department staff documented the triage group in 223 of 503 cases (45%). Concordance with the group assigned by a cardiologist was 70% (kappa = 0.73; SE kappa = 0.04). When the management algorithm was applied correctly, 92% of triage decisions were correct (95% confidence interval [CI], 87%-96%). The triage decision was less often correct when risk stratification was not done (78% [73%-83%], P < 0.001), overestimated (77% [66%-88%], P < 0.01), or underestimated (50% [18%-82%], P < 0.001). The proportion of patients free of major cardiovascular events at four-month follow-up was 50% for those with myocardial infarction with ST-segment elevation, 47% for those with a high short-term risk of an adverse cardiac event, 82% for those with intermediate risk, and 99% for those with a low risk or non-coronary chest pain (P < 0.001). CONCLUSIONS: Use of management algorithms by emergency staff was poor. When used, triage decisions were more likely to be correct. Subsequent outcome confirms that the NHMRC risk stratification algorithms are useful for prognostic stratification of patients with suspected acute coronary syndrome.

Adult↗

PubMed2000

Evaluation of the WHO/UNICEF algorithm for integrated management of childhood illness between the ages of one week to two months.

OBJECTIVE: To evaluate the utility of the WHO/UNICEF algoritham for integrated management of childhood illness (IMCI) between the ages of 1 week to 2 months. DESIGN: Prospective observational. SETTING: The Outpatient Department and Emergency Room of a medical college hospital. METHODS: 129 infants presenting to Outpatient Department (n=70) or Emergency Room (n=59) were assessed and classified as per 'IMCI' algorithm and treatment required was identified. A detailed evaluation with all relevant investigations was also done for these subjects. The final diagnoses made and therapies instituted on this basis served as 'gold standard'. The diagnostic and therapeutic agreement between 'gold standard and the 'IMCI' was computed. RESULTS: More than one illness was present in 97(75.2%) of subjects as per 'gold standard' (mean 2.1). Subjects having any referral criteria as per 'IMCI' algorithm had a greater (p=0.002) co-existence of illnesses (mean 2.3 vs. 1.8 illnesses per child, respectively. IMCI algorithm covered majority (81-84%) of the recorded diagnoses either partly (40-41%) or fully (40-44%). The referral criteria proved quite sensitive (86-87%) in predicting hospitalization but had a lower specificity (53-58%). a total agreement with IMCI was found in 60-66% cases. The mismatch (34-40%) was more commonly of overdiagnosis (21-23%) rather than underdiagnosis (15-21%). The sensitivity of the algorithm to identify serious bacterial infection was high (96.1-96.5%) while the specificity was relatively low (51. 8-59.7%). Upper respiratory infection (URI)emerged as an important cause resulting in unnecessary referrals (13 out of 21 cases). Of the 43 cases identified as diarrhea by the algorithm, 6 had breast fed stools, which do not require any therapy. The 'IMCI' algorithm had a provision for preventive services of immunization and breastfeeding counseling (18% possibility of availing missed opportunities in both). CONCLUSION: There is a sound scientific basis for adopting IMCI approach even in young infants as there is a need to improve the specificity of referral criteria. Two important conditions identified for possible refinement are URI and breast fed stools

Algorithms↗

PubMed1999

Algorithm for assessment of the oxygen, acid-base and electrolyte status (ALBOA BEACH).

To achieve accurate interpretation of blood gas analysis data in intensive care units, the oxygen, acid-base and electrolyte-metabolite profile of arterial blood should be comprehensively and adequately monitored and assessed. A number of diagnostic software programmes have been developed to assist clinicians in this and to help improve instruction in this field. In the present study we describe an algorithm and a programme for assessment of the oxygen, acid-base and electrolyte status of the arterial blood. The algorithm, as well as the software programme, is named ALBOA BEACH, which is an acronym of the programme's full name--ALgorithm for Blood Oxygen, Acid-Base, Electrolyte And respective CHarts. The algorithm takes account of the latest achievements in blood-gas analysis; it can be used both in on-line and off-line modes and is useful in the laboratory and pathophysiological assessment of blood-gas disorders. The algorithm is based on a rigorous pathophysiological analysis and the interpretation achieved as a result is much more accurate than usual. Thus it provides a most reliable basis for taking the right clinical decision and making the proper prognosis of the possible critical conditions. The algorithm also provides additional bases for interpretations of data in the borderline zones, with due warnings in cases of absence of physiological coherence between the various parameters and modules. The programme is designed to be used with the modern multifunctional blood-gas analysers; it can be of assistance in making diagnoses, in the training and research activity in the intensive care units and specialised blood gas laboratories.

Acid-Base Equilibrium↗

PubMed1992

An annotated algorithm approach to clinical guideline development.

The Urinary Incontinence in Adults Guideline Panel facilitated the ready elucidation of its guideline's management recommendations through the use of an annotated algorithm approach. The algorithms created as part of this guideline differ from previous algorithms in two ways: (1) they employ systematic annotation to link explicitly the algorithms' recommendations to the literature, and (2) they contain patient counseling and decision nodes to depict the major preference-dependent decision or branch points in the algorithm. We believe that these two innovations can help ensure the clinical validity of guidelines' algorithms while preserving appropriate clinical flexibility and respecting patient preferences.

Algorithms↗

PubMed1991

Exploring a new best information algorithm for Iliad.

Iliad is a diagnostic expert system for internal medicine. One important feature that Iliad offers is the ability to analyze a particular patient case and to determine the most cost-effective method for pursuing the work-up. Iliad's current "best information" algorithm has not been previously validated and compared to other potential algorithms. Therefore, this paper presents a comparison of four new algorithms to the current algorithm. The basis for this comparison was eighteen "vignette" cases derived from real patient cases from the University of Utah Medical Center. The results indicated that the current algorithm can be significantly improved. More promising algorithms are suggested for future investigation.

Algorithms↗

PubMed1991

Stochastic simulation algorithms for query networks.

One of the barriers to using belief networks for medical information retrieval is the computational cost of reasoning as the networks become large. Stochastic simulation algorithms allow one to compute approximations of probability values in a reasonable amount of time. We previously examined the performance of five stochastic simulation algorithms applied to four simple belief networks networks and found that the Self-Importance algorithm performed well. In this paper, we examine how the same five algorithms perform when applied to a belief network derived from the cardiovascular subtree of the Medical Subject Headings (MeSH). Both the Likelihood Weighting and Self-Importance algorithms perform well when applied to the MeSH-derived network, suggesting that stochastic simulation algorithms may provide reasonable performance in medical information retrieval settings.

Algorithms↗

PubMed1991

Use of immunoglobulin heavy-chain and light-chain measurements in a multicenter trial to investigate monoclonal components: II. Classification by use of computer-based algorithms.

We describe a computer algorithm for classifying serum monoclonal proteins (MC) based on serum protein electrophoresis (SPE) and the automated measurement of kappa and lambda light chains and IgG, IgA, and IgM. We developed the algorithm by using a large database of unselected samples containing MC collected in a multicenter study. The performance of the algorithm was optimized by using iterative computational procedures and was tested on both the development database and on an independent set of MC-containing samples. With the development database, the algorithm correctly classified 50% and misassigned 2.5% of the MC. Where the MC were present in concentrations greater than 10 g/L, the rate of successful classification increased to 72% with 3% misclassification. When the algorithm was tested on a group of 101 MC-containing samples from an independent source, 67% were correctly classified and 8% misclassified, half of the latter being unusual IgD myelomas. We discuss the scope for the application of the algorithm in routine laboratory practice involving personal computer software.

Algorithms↗

PubMed1991

Algorithm to predict triple-vessel/left main coronary artery disease in patients without myocardial infarction. An international cross validation.

Logistic regression was applied to the clinical, risk factor, and exercise data of consecutive angiographic referrals without prior myocardial infarction to determine an algorithm predicting the probability of triple-vessel/left main coronary artery disease. These data were obtained from a total of 1,074 such subjects from patient populations at four centers (Cleveland Clinic Foundation, Cleveland, Ohio; Hungarian Institute of Cardiology, Budapest, Hungary; the university hospitals, Zurich and Basel, Switzerland; and the Veterans Administration Medical Center, Long Beach, Calif.) and used to derive four separate probability algorithms. Each algorithm is based on patient data from study samples at three of the four centers and consists of 272 logistic functions, which are related to linear combinations of 13 variables (age, sex, type of chest pain, systolic blood pressure, resting electrocardiogram, serum cholesterol, fasting blood sugar, achieved exercise work load, achieved heart rate, exercise-induced angina and hypotension, heart rate-adjusted resting ST depression, and exercise ST slope). The four algorithms were cross validated by testing them on the populations not involved in their derivation. The resulting probabilities in the four test groups were then compared with the angiographic findings of triple-vessel/left main coronary artery disease. The discriminatory power of all the algorithms was fair to good (area under receiver operating characteristic curve, 0.68, 0.75, 0.82, 0.85) in the test groups. The algorithm did not significantly underestimate or overestimate disease probability except in one center (Long Beach).(ABSTRACT TRUNCATED AT 250 WORDS)

Algorithms↗

PubMed1989

[The use of a computer for the diagnosis of comatose states in diabetics (a differential diagnostic algorithm)].

An algorithm for differential diagnosis of comatose conditions in patients with diabetes mellitus has been devised. The algorithm is intended for the general practitioner and non-specialized department. The algorithm uses the minimum of the crucial signs of comatose conditions, ensuring their diagnosis under the conditions of any hospital. The algorithm can be applied to the recognition of the typical variants of comatose conditions in a "pure" form. The amount of algorithm steps is minimized. The program of differential diagnosis is written in the Fokal language and realized on the computer "Elektronika BK 0010". The program is run in the dialogue mode. The algorithm is used in clinical practice and in the training process, with its efficacy being independent of the professional skills of the user.

Algorithms↗

PubMed1989

Prediction of protein helices with a derivative of the strip-of-helix hydrophobicity algorithm.

The strip-of-helix hydrophobicity algorithm was devised to identify protein sequences which, when coiled as alpha or 3(10) helices, had one axial, hydrophobic strip and otherwise variably hydrophilic residues. The strip-of-helix hydrophobicity algorithm also ranked such sequences according to an index, the mean hydrophobicity of amino acids in the axial strip. This algorithm well predicted T cell-presented fragments of antigenic proteins. A derivative of this algorithm (the structural helices algorithm (SHA] was tested for the prediction of helices in crystallographically defined proteins. For the SHA, eight amino acid sequences, 2 cycles plus one amino acid in an alpha helix, with strip-of-helix hydrophobicity indices greater than 2.5, were selected with overlapping segments joined. These selections were terminated according to simple "capping rules," which took into account the roles of N-terminal Asn or Pro and C-terminal Gly in the stability of helices. In analyses of 35 crystallographically defined proteins with known alpha and 3(10) helices, the predictions with the SHA overlapped (had overlap indices x greater than or equal to 0.5) with 34% of known helices, touched (had overlap indices 0.5 greater than x greater than 0) or overlapped with 66% of known helices, or were neighboring (came within 6 residues) or touched or overlapped with 82% of known helices. At each level of judging the quality of prediction, the SHA was usually less sensitive (correct predictions/total number of known helices) and more efficient (correct predictions/total number of predictions) than the Chou-Fasman and Garnier-Robson methods. It was simpler in design and calculation. The chemical mechanisms underlying these algorithms appear to apply both to protein folding and to selection of T cell-presented antigenic sequences.

Algorithms↗

PubMed1994

[Development of clinical algorithms for quality assurance in management of multiple trauma].

Resuscitation and management of high-risk multiple trauma patients require a systematic and coordinated approach to diagnostic and therapeutic interventions. Clinical algorithms with branch chain decision logic can provide a clear and organized transformation of clinical standards for trauma care. Owing to their capability in formalization and standardization, algorithms define precisely the process of care and serve as a central interface within the system of quality assurance and quality control. The standardized document symbols and conventions for information processing according to ANSI/ISO/CCITT regulations are generally applied to the flowchart design of clinical algorithms. Special starting and ending point symbols make it possible to break down complex processes in several single interrelated algorithms. Inclusion of optional criteria checklists reduces the number of decision nodes and loops and minimizes the extent of a comprehensive algorithm. Clinical algorithms are an excellent tool for converting highly complex concepts of multiple trauma management into a logical, prioritized and systematic process of care.

Algorithms↗

PubMed1996

[Verification of accuracy of several algorithms to quantitate left ventricular regional wall motion: a study using cine MR imaging with myocardial tagging].

To verify the accuracy of several algorithms used to quantitate left ventricular (LV) regional wall motion, five volunteers were examined by cine MR imaging with presaturation myocardial tagging in short-axis and 4-chamber sections. Three algorithms for the wall motion analysis, radial, centerline, and originally developed "modified-Hildreth" methods, were applied to the cine MR images, and dissociation of the end-systolic position of the tags estimated by each algorithm from the true position was examined. The modified-Hildreth method was comparable in accuracy to the other methods for estimating end-systolic tag position. Significantly worse estimation of the tag position by the three algorithms occurred in the 4-chamber section compared with the short-axis section (p < 0.001, 0.005), indicating difficulties in the wall motion analysis of "long-axis" LV images. Among the algorithms, the centerline method showed the highest accuracy of the estimation in the 4-chamber section, and the modified-Hildreth method was the best in the short-axis section. In the 4-chamber section, correction of the position of end-diastolic and end-systolic images around the luminal centroids improved the estimation (p < 0.01, 0.05). Tagging cine MR imaging was proved to be useful for determining the most suitable algorithm for quantitative wall motion analysis of LV images obtained from conventional angiocardiography and other imaging modalities.

Adult↗

PubMed1996

A modified EM algorithm for estimation in generalized mixed models.

Application of the EM algorithm for estimation in the generalized mixed model has been largely unsuccessful because the E-step cannot be determined in most instances. The E-step computes the conditional expectation of the complete data log-likelihood and when the random effect distribution is normal, this expectation remains an intractable integral. The problem can be approached by numerical or analytic approximations; however, the computational burden imposed by numerical integration methods and the absence of an accurate analytic approximation have limited the use of the EM algorithm. In this paper, Laplace's method is adapted for analytic approximation within the E-step. The proposed algorithm is computationally straightforward and retains much of the conceptual simplicity of the conventional EM algorithm, although the usual convergence properties are not guaranteed. The proposed algorithm accommodates multiple random factors and random effect distributions besides the normal, e.g., the log-gamma distribution. Parameter estimates obtained for several data sets and through simulation show that this modified EM algorithm compares favorably with other generalized mixed model methods.

Algorithms↗

PubMed1997

Evaluation of an algorithm for integrated management of childhood illness in an area of Kenya with high malaria transmission.

In 1993, the World Health Organization completed the development of a draft algorithm for the integrated management of childhood illness (IMCI), which deals with acute respiratory infections, diarrhoea, malaria, measles, ear infections, malnutrition, and immunization status. The present study compares the performance of a minimally trained health worker to make a correct diagnosis using the draft IMCI algorithm with that of a fully trained paediatrician who had laboratory and radiological support. During the 14-month study period, 1795 children aged between 2 months and 5 years were enrolled from the outpatient paediatric clinic of Siaya District Hospital in western Kenya; 48% were female and the median age was 13 months. Fever, cough and diarrhoea were the most common chief complaints presented by 907 (51%), 395 (22%), and 199 (11%) of the children, respectively; 86% of the chief complaints were directly addressed by the IMCI algorithm. A total of 1210 children (67%) had Plasmodium falciparum infection and 1432 (80%) met the WHO definition for anaemia (haemoglobin < 11 g/dl). The sensitivities and specificities for classification of illness by the health worker using the IMCI algorithm compared to diagnosis by the physician were: pneumonia (97% sensitivity, 49% specificity); dehydration in children with diarrhoea (51%, 98%); malaria (100%, 0%); ear problem (98%, 2%); nutritional status (96%, 66%); and need for referral (42%, 94%). Detection of fever by laying a hand on the forehead was both sensitive and specific (91%, 77%). There was substantial clinical overlap between pneumonia and malaria (n = 895), and between malaria and malnutrition (n = 811). Based on the initial analysis of these data, some changes were made in the IMCI algorithm. This study provides important technical validation of the IMCI algorithm, but the performance of health workers should be monitored during the early part of their IMCI training.

Algorithms↗

PubMed1998

Learning algorithms based on linearization.

The aim of this article is to investigate a mechanical description of learning. A framework for local and simple learning algorithms based on interpreting a neural network as a set of configuration constraints is proposed. For any architectural design and learning task, unsupervised and supervised algorithms can be derived, optionally using unconstrained and hidden neurons. Unlike algorithms based on the gradient in weight space, the proposed tangential correlation (TC) algorithms move along the gradient in state space. This results in optimal scaling properties and simple expressions for the weight updates. The number of synapses is much larger than the number of neurons. A constraint for neural states does not impose a unique constraint for synaptic weights. Which weights to assign credit to can be selected from a parametrization of all weight changes equivalently satisfying the state constraints. At the heart of the parametrization are minimal weight changes. Two supervised algorithms (differing by their parametrizations) operating on a three-layer perceptron are compared with standard backpropagation. The successful training of fixed points of recurrent networks is demonstrated. The unsupervised learning of oscillations with variable frequencies is performed on standard and more sophisticated recurrent networks. The results presented here can be useful both for the analysis and for the synthesis of learning algorithms.

Algorithms↗