Filters
Results 1 - 10 of 60
Results 1 - 10 of 60.
Search took: 0.037 seconds
Sort by: date | relevance |
AbstractAbstract
[en] An original and entirely automatic algorithm is proposed to select regions of interest (ROIs) on dynamic scintigrams. This algorithm is based on factor analysis and on cluster analysis. It consists of first extracting the orthogonal factor images of the series using factor analysis of correspondence. These factor images are then automatically segmented in ROIs using a hierarchical ascendant classification procedure. The distance used for the classification is the 'minimum added intra-class variance' distance. This algorithm has been implemented on a fast computer dedicated to nuclear medicine (Nodecrest Micas V system). The time of calculation on 1000 pixels from 40 images is less than 5 min when three factor images are used. This algorithm is validated using a numerical phantom and is illustrated using renal (99Tcm DTPA) and cardiac (equilibrium gated angiography) dynamic scintigraphies. The results show that the algorithm is able to recognize the bladder, the renal cavities and the renal parenchyma on the renal series, and the ventricules and the atria on the cardiac series. (Author)
Primary Subject
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] Published in summary form only
Original Title
Correction de la diffusion Compton par analyse factorielle des structures dynamiques
Primary Subject
Source
28. French Colloquium on nuclear Medicine; 28. Colloque de Medecine Nucleaire de langue francaise; Paris (France); 8-10 Dec 1988
Record Type
Journal Article
Literature Type
Conference
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
Itti, R.; Liehn, J.C.; Valeyre, J.
20. French language symposium on nuclear medicine. Bordeaux, 20-22 September 19791979
20. French language symposium on nuclear medicine. Bordeaux, 20-22 September 19791979
AbstractAbstract
[en] A brief history enables the place taken by computers in nuclear medicine to be situated. Three stages are involved: at first, the quantitative functional studies developed (radioisotopic uptake curves) then the morphological studies with the research into increasingly efficient pictures systems and the introduction of data processing to improve the quality of the pictures and, finally, the digital analysis of the pictures leading to the build-up of regional physiological studies. The part played by computer systems in a Nuclear Medicine Department is summarized. The main advantages are emphasized and present possibilities are indicated. The components which make it possible to design a future acquisition and data processing system that may be used not only for routine practice examinations but also for axial scanning studies by emission are presented. The application of hyper-specialized systems using microprocessors is also under consideration
[fr]
Un bref historique permet de situer la place que prennent les ordinateurs dans la medecine nucleaire. En trois etapes se sont developpees, au debut, les etudes fonctionnelles quantitatives (courbes de fixation radioisotopiques) puis, les etudes morphologiques avec la recherche de systemes d'imaginerie de plus en plus performants et l'introduction de l'informatique pour ameliorer la qualite des images et, enfin, l'analyse numerique des images qui conduit a la realisation d'etudes physiologiques regionales. On resume le role joue par les systemes informatiques dans un service de medecine nucleaire. Les principaux avantages sont soulignes et les possibilites actuelles sont indiquees. Les elements qui permettent de concevoir un future systeme d'acquisition et de traitement des donnees utilisable tant pour les examens de pratique courante que pour les examens tomographiques axiaux par emission sont presentes. On envisage en outre l'application de systemes hyper-specialises utilisant la technique des microprocesseursOriginal Title
L'equipement informatique d'un Service de Medecine Nucleaire en 1980
Secondary Subject
Source
Blanquet, P.; Ducassou, D. (eds.); p. 2.a-2.q; 1979; p. 2.a-2.q; Universite de Bordeaux II; Bordeaux, France; 20. French language symposium on nuclear medicine; Bordeaux, France; 20 - 22 Sep 1979; Published in summary form only.
Record Type
Book
Literature Type
Conference
Country of publication
Reference NumberReference Number
Related RecordRelated Record
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] Published in summary form only
Original Title
Analyse factorielle des structures dynamiques: combien de facteurs faut-il extraire
Primary Subject
Source
28. French Colloquium on nuclear Medicine; 28. Colloque de Medecine Nucleaire de langue francaise; Paris (France); 8-10 Dec 1988
Record Type
Journal Article
Literature Type
Conference
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] This article is a review of the Factor Analysis methods used in Nuclear Medicine. These methods consist of processing temporal or energetic dynamic scintigraphic series with Descriptive Multivariate Analysis methods. The principle for using these methods is the following: the dynamic series is regarded as a table in which each row is a dixel and each column is an image. The three methods used in Nuclear Medicine are Principal Components Analysis and Factor Correspondence Analysis which are Orthogonal Analysis methods and Factor Analysis which is an Oblique Analysis method. The results of Orthogonal Analysis consist of a limited number of non correlated factor curves and images which can be used for generating parametric images, for nosologic classification, for smoothing or for data compression. The software of Oblique Analysis used in Nuclear Medicine is called factor Analysis of Dynamic Structures (FADS). It generates a limited number of correlated oblique factor images and curves which are estimates of Dynamic Structures and of kinetics in these structures
[fr]
Cet article est une revue des techniques d'analyse des sequences d'images par les methodes dites d'Analyse Factorielle. Ces methodes consistent a appliquer les methodes statistiques d'Analyse Multivariee Descriptive aux series dynamiques, temporelles ou energetiques, d'images scintigraphiques. Leur principe est de considerer la serie dynamique comme un tableau de chiffres dont chaque ligne est un dixel et chaque colonne une image de la serie. Les trois principales methodes d'Analyse Multivariee utilisees en Medecine Nucleaire sont d'une part l'Analyse en Composantes Principales et l'Analyse Factorielle des Correspondances qui sont des methodes d'Analyse Orthogonale et d'autre part l'Analyse Factorielle proprement dite qui est une methode d'Analyse Oblique. L'Analyse Orthogonale permet de representer la serie dynamique par un nombre limite de courbes et d'images factorielles qui peuvent etre utilisees pour generer des images parametriques, pour la classification nosologique, ou pour le lissage et la compression des images. Les applications de l'Analyse Oblique en Medecine Nucleaire sont basees sur l'Algorithme de l'Analyse Factorielle des Structures Dynamiques (AFSD). Cet algorithme permet de generer un nombre limite d'images et de courbes factorielles qui sont correlees entre elles et qui sont les estimations respectivement des Structures Dynamiques et des cinetiques dans ces structuresOriginal Title
L'analyse des sequences d'images en Medecine Nucleaire: 2 - Application des methodes dites d'Analyse Factorielle
Primary Subject
Source
94 refs.
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] The value of new ultrasensible and rapid immunoradiometric assay of thyroid stimulating hormone (TSH) for the diagnosis of hyperthyroidism was assessed in 130 patients with suspected hyperthyroidism and in 330 controls. The diagnosis was established by the clinical evaluation, thyroid scintigraphy and serum concentrations of thyroid hormones. Using the ROC (Receiver Operating Characteristic) curve methodology which allows the optimization of sensitivity and specificity, the physician can choose the Cut-off value between hyperthyroidism and euthyroidism. Two points of the curve seem to be interesting: using the cut-off value of 0.1 mUI/1, sensitivity is 0.98 and specificity is 0.98; using the cut-off value of 0.3 mUI/1, sensitivity is 1.00 and specificity is 0.92. Using the association TSH and FT4 (Free Thyroxin), sensitivity is 0.94 and specificity is 0.99. Sixty four per cent of euthyroid patients with TSH under 0.3 mUI/1 have one or several hot nodules and only two have no thyroid disease. A TRH (Thyrotrophin Releasing Hormone) test was carried out in 63 patients with suspected thyrotoxicosis: basal and TRH stimulated TSH levels were under 0.1 mUI/1. This immunoradiometric assay for TSH may simplify the approach to thyroid function testing in patients with suspected thyrotoxicosis: a basal TSH under 0.3 mUI/1 is sufficient to confirm a clinical suspicion of thyrotoxicosis without TRH test within four hours. In a department devoted to testing thyroid function, this new method provides a great benefit in cost and work
[fr]
Une nouvelle methode radioimmunometrique, ultrasensible et rapide de dosage de la thyreostimuline (TSH) plasmatique a ete evaluee chez 130 patients suspects d'hyperthyroidie et 330 temoins. Le diagnostic a ete etabli par l'examen clinique, la scintigraphie thyroidienne et les concentrations plasmatiques d'hormones thyroidiennes. La valeur seuil entre les hyperthyroidiens et les temoins est choisie en utilisant la methodologie des courbes ROC (Receiver Operating Characteristic) afin d'optimiser le couple sensibilite-specificite en fonction des objectifs du clinicien; 2 points de la courbe semblent interessants: une valeur seuil fixee a 0,1 mUI/1 confirme 98 pour cent des suspicions d'hyperthyroidies, une valeur seuil a 0,3 mUI/1 confirme 100 pour cent des hyperthyroidies mais 8 pour cent des temoins sont des faux positifs (FP). L'association TSH-FT4 (Free Thyroxin) ramene ce taux de FP a 1 pour cent. Parmi les euthyroidiens qui ont une TSH inferieure a 0,3 mUI/1, 64 pour cent ont un ou plusieurs nodules chauds et 2 seulement n'ont aucune pathologie thyroidienne. Un test au TRH a ete realise chez 63 hyperthyroidiens: leur TSH basale et apres stimulation est restee inferieure a 0,1 mUI/1. L'apparition de cette methode de dosage de la TSH simplifie l'evaluation de la fonction thyroidienne puisqu'une valeur basale inferieure a 0,3 mUI/1 est suffisante pour confirmer en 4 heures une suspicion clinique d'hyperthyroidie sans avoir recours au test au TRH. Il en resulte un allegement de la charge de travail et une reduction du cout dans un service d'exploration fonctionnelle thyroidienneOriginal Title
TSH ultrasensible: une nouvelle approche diagnostique de l'hyperthyroidie
Primary Subject
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] This is a review of parametric imaging methods on Nuclear Medicine. A Parametric Image is an image in which each pixel value is a function of the value of the same pixel of an image sequence. The Local Model Method is the fitting of each pixel time activity curve by a model which parameter values form the Parametric Images. The Global Model Method is the modelling of the changes between two images. It is applied to image comparison. For both methods, the different models, the identification criterion, the optimization methods and the statistical properties of the images are discussed. The analysis of one or more Parametric Images is performed using 1D or 2D histograms. The statistically significant Parametric Images, (Images of significant Variances, Amplitudes and Differences) are also proposed
[fr]
Cet article est une revue de la methode des Images Parametriques en Medecine Nucleaire. La Methode du Modele Local consiste a ajuster les parametres d'un modele a chacune des courbes decrivant l'evolution du contenu d'un pixel. Les Images Parametriques sont celles des valeurs de ces parametres. La Methode du Modele Global est appliquee a la comparaison de deux images. Elle consiste a rechercher les valeurs des parametres d'un modele qui explique l'evolution d'une image a l'autre. L'image Parametrique est alors celle des residus, c'est-a-dire celle des regions ne pouvant etre modelisees et qui sont le plus souvent les regions pathologiques. Pour chacune de ces methodes, les differents modeles, les differents criteres d'identification, les differentes methodes d'optimisation, ainsi que les proprietes statistiques des estimateurs peuvent etre decrits et discutes. L'analyse d'une ou plusieurs Images Parametriques peut etre faite par la methode des histogrammes a une ou plusieurs dimensions. Ceux-ci peuvent etre eux-memes analyses en tenant compte des erreurs standards associees aux parametres. La synthese d'Images Parametriques statistiquement significatives (images des variances, des amplitudes et des differences significatives) est proposeeOriginal Title
L'analyse des sequences d'images en medecine nucleaire: (1). Application de la methode du modele a l'imagerie parametrique
Primary Subject
Source
149 ref.
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
No abstract available
Original Title
Analyse bi-exponentielle des courbes de lavage pulmonaire lors de l'exploration fonctionnelle au xenon 133
Primary Subject
Source
Valeyre, J.; Deltour, G.; Delisle, M.J. (eds.); p. 5; nd; p. 5; Faculte de Medecine; Reims, France; 18. French language symposium on nuclear medicine; Reims, France; 9 - 12 Jun 1977; Available from Institut Jean Goudinot, 45, rue Cognacq-Jay, 51090 Reims, France; Published in abstract form only.
Record Type
Book
Literature Type
Conference
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] A software which performs the construction and the statistical comparison between ROC curves is proposed. It uses quantitative or qualitative (i.e. obtained by the rating method) data. The statistical test is based on a method proposed by J.A. Hanley which compares the values of the areas under the ROC curves and takes automatically into account the correlation between the data. The software runs on IBM PC compatible computers. It may be of great help for clinical studies
[fr]
On propose un logiciel qui realise la construction et la comparaison statistique de courbes ROC. Il travaille a partir de donnees quantitatives ou qualitatives (obtenues par la methode du rating). La comparaison des courbes ROC est effectuee selon un test propose par J.A. Hanley. Le test statistique est effectue entre les surfaces sous les courbes ROC et tient automatiquement compte de l'appariement des donnees. Ce logiciel qui fonctionne sur les micro-ordinateurs compatibles IBM PC peut rendre de grands services pour la comparaison de methodes diagnostiques en medecine nucleaire et ailleursOriginal Title
La comparaison de methodes diagnostiques: Realisation d'un logiciel effectuant la construction et la comparaison statistique de courbes ROC
Primary Subject
Secondary Subject
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
AbstractAbstract
[en] In order to perform automatically the thresholding of the amplitude image obtained by phase analysis, a filter based on the probability density of the amplitude is proposed. It generates the image of the significant amplitude by setting to zero the amplitude if its square is lower than 4α0 log(1/α)/N where α0 is the mean of the signal, α the risk of error and N the number of frames. The power of the test is discussed and the modification of the filter for pre-processed data is provided. (author)
Primary Subject
Record Type
Journal Article
Journal
Country of publication
Reference NumberReference Number
INIS VolumeINIS Volume
INIS IssueINIS Issue
1 | 2 | 3 | Next |