Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Feature selection”

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 1,027 records · Page 57Linked to original sources
PubMed2000

Occipital sawtooth: a physiological EEG pattern in very premature infants.

OBJECTIVE: To evaluate EEG maturational features in preterm infants below 27 weeks postmenstrual age. METHODS: EEGs recorded from 5 preterm infants (postmenstrual age 24-26 weeks) were examined and selected maturational features were scored and quantified. The five infants also had serial cranial ultrasound scans (US) and magnetic resonance images of the brain within the first weeks after birth. RESULTS: Background activity was markedly discontinuous in all patients and very variable. Temporal sawtooths occurred but less frequently than in older preterm infants. All 5 infants also showed a particular novel feature, characterized by rhythmic, regularly shaped, medium-high amplitude 4-7 Hz activities, lasting 0.5-3 s and located in the occipital regions. This pattern was symmetrical but sometimes asynchronous. CONCLUSIONS: Occipital sawtooth, so called because it shares shape and frequency with temporal sawtooth but has an occipital localisation, constitutes a physiological EEG pattern characteristic of premature infants between 24 and 26 weeks of postmenstrual age.

Electroencephalography↗
PubMed2004

Computerized characterization of breast masses on three-dimensional ultrasound volumes.

We are developing computer vision techniques for the characterization of breast masses as malignant or benign on radiologic examinations. In this study, we investigated the computerized characterization of breast masses on three-dimensional (3-D) ultrasound (US) volumetric images. We developed 2-D and 3-D active contour models for automated segmentation of the mass volumes. The effect of the initialization method of the active contour on the robustness of the iterative segmentation method was studied by varying the contour used for its initialization. For a given segmentation, texture and morphological features were automatically extracted from the segmented masses and their margins. Stepwise discriminant analysis with the leave-one-out method was used to select effective features for the classification task and to combine these features into a malignancy score. The classification accuracy was evaluated using the area Az under the receiver operating characteristic (ROC) curve, as well as the partial area index Az(0.9), defined as the relative area under the ROC curve above a sensitivity threshold of 0.9. For the purpose of comparison with the computer classifier, four experienced breast radiologists provided malignancy ratings for the 3-D US masses. Our dataset consisted of 3-D US volumes of 102 biopsied masses (46 benign, 56 malignant). The classifiers based on 2-D and 3-D segmentation methods achieved test Az values of 0.87+/-0.03 and 0.92+/-0.03, respectively. The difference in the Az values of the two computer classifiers did not achieve statistical significance. The Az values of the four radiologists ranged between 0.84 and 0.92. The difference between the computer's Az value and that of any of the four radiologists did not achieve statistical significance either. However, the computer's Az(0.9) value was significantly higher than that of three of the four radiologists. Our results indicate that an automated and effective computer classifier can be designed for differentiating malignant and benign breast masses on 3-D US volumes. The accuracy of the classifier designed in this study was similar to that of experienced breast radiologists.

Algorithms↗
PubMed2001

Texture analysis of fluorescence microscopic images of colonic tissue sections.

The aim of this study was to assess the potential of texture analysis for the characterization of fluorescence images from colonic tissue sections stained with a novel and selective fluoroprobe, Rhodamine B-phenylboronic acid. Fluorescence microscopy images of colonic healthy mucosa (n = 35) and adenocarcinomas (n = 35) were digitally captured and subjected to image texture analysis. Textural features derived from the grey level co-occurrence matrix were calculated. A modified version of the multiple discriminant analysis criterion was used to choose an appropriate subset of features. A minimum Mahalanobis distance, linear discriminant classifier and a simple evaluation 'score' method were used to classify image feature data into the two categories. A subset of four textural features was selected and used for the description and classification of each image field. They were found appropriate to correctly classify 95% of the images into the two classes, using two different classifiers. These features contained information about local homogeneity and grey level linear dependencies of the image. This study demonstrated that texture analysis techniques could provide valuable diagnostic decision support in a complex domain such as colorectal tissue.

Adenocarcinoma↗
PubMed2002

Pharmacophore mapping of a series of 2,4-diamino-5-deazapteridine inhibitors of Mycobacterium avium complex dihydrofolate reductase.

Pharmacophore hypotheses were developed for a series of 2,4-diamino-5-deazapteridine inhibitors of Mycobacterium avium complex (MAC) and human dihydrofolate reductase (hDHFR). Training sets consisting of 20 inhibitors were selected in each case on the basis of the information content of the structures and activity data as required by the HypoGen program in the Catalyst software. In the case of MAC DHFR inhibitors, the best pharmacophore in terms of statistics and predictive value consisted of four features: two hydrogen bond acceptors (HA), one hydrophobic (HY) feature, and one ring aromatic (RA) feature. The selected pharmacophore hypothesis yielded an rms deviation of 0.730 and a correlation coefficient of 0.967 with a cost difference (null cost minus total cost) of approximately 52. The pharmacophore was validated on a large set of test inhibitors. For the test series, a classification scheme was used to distinguish highly active from moderately active and inactive compounds on the basis of activity ranges. This classification scheme is more practical than actual estimated values because these values have no meaning for compounds yet to be tested except that they indicate whether the compounds will be active or inactive in a biological assay. For the training set, the success rate for predicting active and inactive compounds was 100%. For the test set, the success rate in predicting active compounds was greater than 92% while about 7% of the inactive compounds were predicted to be active. This successful prediction was further validated on three structurally diverse compounds active against MAC DHFR. Two compounds mapped well onto three of the four features of the pharmacophore. The third compound was mapped to all four features of the pharmacophore. This validation study provided confidence for the usefulness of the selected pharmacophore model to identify compounds with diverse structures from a database search. Comparison of pharmacophores for inhibitors of human and MAC DHFR is expected to reveal fundamental differences between these two pharmacophores that may be effectively exploited to identify and design compounds with high selectivity for MAC DHFR.

Folic Acid Antagonists↗
PubMed2005

Development of 3D chromatin texture analysis using confocal laser scanning microscopy.

INTRODUCTION: Analysis of nuclear texture features as a measure of nuclear chromatin changes has been proven to be useful when measured on thin (5-6 microm) tissue sections using conventional 2D bright field microscopy. The drawback of this approach is that most nuclei are not intact because of those thin sections. Confocal laser scanning microscopy (CLSM) allows measurements of texture in 3D reconstructed nuclei. The aim of this study was to develop 3D texture features that quantitatively describe changes in chromatin architecture associated with malignancy using CLSM images. METHODS: Thirty-five features thoughtfully chosen from 4 categories of 3D texture features (discrete texture features, Markovian features, fractal features, grey value distribution features) were selected and tested for invariance properties (rotation and scaling) using artificial images with a known grey value distribution. The discriminative power of the 3D texture features was tested on artificially constructed benign and malignant 3D nuclei with increasing nucleolar size and advancing chromatin margination towards the periphery of the nucleus. As a clinical proof of principle, the discriminative power of the texture features was assessed on 10 benign and 10 malignant human prostate nuclei, evaluating also whether there was more texture information in 3D whole nuclei compared to a single 2D plane from the middle of the nucleus. RESULTS: All texture features showed the expected invariance properties. Almost all features were sensitive to variations in the nucleolar size and to the degree of margination of chromatin. Fourteen texture features from different categories had high discriminative power for separating the benign and malignant nuclei. The discrete texture features performed less than expected. There was more information on nuclear texture in 3D than in 2D. CONCLUSION: A set of 35 3D nuclear texture features was used successfully to assess nuclear chromatin patterns in 3D images obtained by confocal laser scanning microscopy, and as a proof of principle we showed that these features may be clinically useful for analysis of prostate neoplasia.

Cell Nucleus↗
PubMed1998

Prognosis of intravesical bacillus Calmette-Guerin therapy for superficial bladder cancer by immunological urinary measurements: statistically weighted syndrome analysis.

PURPOSE: The goal of this research was to discover new biological indicators in urine which could be used for short-term prognosis of local Bacillus Calmette-Guerin (BCG) therapy outcome in patients with superficial bladder cancer. PATIENTS AND METHODS: We measured and statistically evaluated soluble immunological molecules in urine from bladder cancer patients (n = 34) receiving BCG intravesically. Urine was collected following each of 6 weekly treatments, processed and assayed. The data base included measurements of interleukin-1 (IL-2, IL-4, IL-6, IL-10, IL-12, soluble intercellular adhesion molecule-1 (sICAM-1), tumour necrosis factor-alpha (TNF alpha), soluble CD14 (sCD14), interferon-gamma (IFN gamma), GM-CSF, volume of urine and its pH. The clinical response was evaluated by urine histology and random quadrant biopsy 3 months after the start of therapy. Patients were divided into 2 groups, with good and poor therapeutic effect. The initial complete response rate was 62% (21/34). The data base was analyzed using traditional multivariate statistical methods and a pattern recognition method which deals with combinatorial-statistical analysis (statistically weighted syndromes (SWS) method) of the gradated features. The SWS method is capable of identifying robust patterns in small "fuzzy" sets with high dimensional objects and some missing values. RESULTS: Only one parameter gave significant differences at p < 0.05, GM-CSF at instillation 6. Repeated measurement analysis of variance, backward stepwise multiple logistic regression and linear discriminant analysis failed to show any significance. However, significant differences in the structure of correlation between features in the groups with and without therapeutic effect were observed and four highly informative variables (the masses of sICAM-1, TNF alpha, sCD14 and pH) relating to 5th-6th installations were selected by SWS. These features provided accurate individual prediction of therapeutic outcome for all our patients. Cross-validation analysis and computer simulation showed the statistically significant stability of the prediction. CONCLUSION: We have selected a set of urinary variables that could be considered as a perspective combination of indicators (syndromes) of outcome of pre-operation BCG therapy of patients with superficial bladder cancer. A larger patient database will provide testing and evaluation of the biological and clinical significance of selected features. The computational syndrome-disease approach should be applicable for the solution of decision-making problems for management of cancer.

Adjuvants, Immunologic↗
PubMed1975

The central nervous system in childhood leukemia. II. Subacute leukoencephalopathy.

A study was performed to evaluate the clinical and histopathologic characteristics of a distinctive degenerative lesion within the central nervous system (CNS) of children with acute lymphocytic leukemia (ALL). Of the 231 patients in this study, 13 were found to have specific degenerative changes in telencephalic white matter. Morphologically, the principal changes consisted of diffuse reactive astrocytosis and multiple, noninflammatory necrotic foci, often containing varying amount of mineralized cellular debris. Clinical features common to all patients with this leukoencephalopathy were: 1) cranial irradiation of 2000 rads or more and 2) methotrexate administered systemically after irradiation. Comparison of selected clinical features of patients with and without leukoencephalopathy showed that methotrexate administered intravenously after a cumulative dose of CNS irradiation of 2000 rads or more can result in degeneration of CNS white matter in patients with ALL. Age at time of irradiation, bacterial infections, nutrition, and CNS leukemia were not causally related to the development of this disease. This study suggests that chemotherapeutic agents may diffuse through the blood-brain barrier following CNS irradiation of 2000 rads or more.

Adolescent↗
PubMed1988

Enhancement and simplification of macromolecular images.

Computer graphics programs have been devised to display selected atomic features and to simplify images of complex macromolecular structures. By using boundary outlines, adjustment of size and shape of the molecular components, color coding, shading, and selective omission of obscuring detail, attention can be focused on specific interactions which determine higher levels of organization. A balanced color table has been constructed in which different hues have equal steps in brightness; this table has facilitated distinction of atom types and sequence coding together with representation of an optimum range of depth cueing and surface shading. The graphics system has been used with the atomic coordinates of the tobacco mosaic virus structure to simplify images of the protein subunit, to illustrate intermolecular interactions, and to relate subunit packing arrangements in different assemblies to the underlying atomic structure. The system has also been used to construct a schematic representation of the polyomavirus capsid, based on low resolution data. Application of artistic methods contributes to the effective presentation and interpretation of detailed scientific information about complex macromolecular structures.

Computer Graphics↗
PubMed2001

Selective adsorption of L- and D-amino acids on calcite: Implications for biochemical homochirality.

The emergence of biochemical homochirality was a key step in the origin of life, yet prebiotic mechanisms for chiral separation are not well constrained. Here we demonstrate a geochemically plausible scenario for chiral separation of amino acids by adsorption on mineral surfaces. Crystals of the common rock-forming mineral calcite (CaCO(3)), when immersed in a racemic aspartic acid solution, display significant adsorption and chiral selectivity of d- and l-enantiomers on pairs of mirror-related crystal-growth surfaces. This selective adsorption is greater on crystals with terraced surface textures, which indicates that d- and l-aspartic acid concentrate along step-like linear growth features. Thus, selective adsorption of linear arrays of d- and l-amino acids on calcite, with subsequent condensation polymerization, represents a plausible geochemical mechanism for the production of homochiral polypeptides on the prebiotic Earth.

Adsorption↗
PubMed1997

Potential for use of a House of Quality Matrix technique in rehabilitation engineering.

This article describes a concept of using the House of Quality matrix tool to aid the process of designing customized seating. The work of a cross-function team in two seating clinics was videotaped. At a later date the information derived from the videotapes was analyzed using a House of Quality matrix. We were interested in the capacity of House of Quality to draw out customer requirements in a seating device and relate engineering requirements for manufacturing the seating. The scores for engineering features derived from the matrix were compared with the features the cross-function team treated with high priority and also the features which the team experienced most difficulty in satisfying as revealed in the video viewed. The matrix scores demonstrated the House of Quality's capacity to select key features 50% of the time. House of Quality was time-consuming to execute. Its application in everyday clinical settings is limited by this fact.

Activities of Daily Living↗
PubMed1991

Identification of first acute Q wave and non-Q wave myocardial infarction by multivariate analysis of body surface potential maps.

BACKGROUND: Patients with acute non-Q wave myocardial infarction (NQMI) appear to have more jeopardized residual myocardium at high risk for subsequent angina, reinfarction, or malignant arrhythmias than patients with acute Q wave myocardial infarction (QMI). Unfortunately, conventional electrocardiographic (ECG) criteria have limited utility in recognizing NQMI. METHODS AND RESULTS: The present study combines the increased information content of body surface potential maps (BSPM) over the 12-lead ECG with the power of multivariate statistical procedures to identify a practical subset of leads that would allow improved diagnosis of NQMI. Discriminant analysis was performed on 120-lead data recorded simultaneously in 159 normal subjects and 308 patients with various types of myocardial infarction (MI) by using instantaneous voltages on time-normalized P, PR, QRS, and ST-T waveforms as well as the duration of these waveforms as features. Leads and features for optimal separation of 159 normals from 183 patients with recent or old QMI (group A) were selected. A total of six features from six torso sites accounted for a specificity of 96% and a sensitivity of 94%. All lead positions were outside the conventional electrode sites and selected features were voltages at mid-P, early and mid-QRS, and before and after the peak of the T wave. The discriminant function was then tested on 57 patients with acute NQMI (group B) and 68 patients with acute QMI (group C): Rates of correct classification were 91% and 93%, respectively. Because of the possible deterioration of the results caused by ST-T abnormalities also present in other clinical entities, a second classification model including an independent group of 116 patients with left ventricular hypertrophy (LVH) but without MI was developed. Two additional measurements were required, namely, P wave duration and a mid-QRS voltage on a lead located 10 cm below V1. Testing the model on both acute MI groups produced correct classification rates of 88% for acute NQMI and 93% for acute QMI. Group mean BSPM were plotted for the three MI groups at successive instants throughout the PQRST waveform. Typical patterns for each MI group were identified during PQRST by removing the corresponding normal variability at each electrode site from sequential MI maps. These standardized maps or discriminant maps provided information on the capability of each measurement at each electrode site and at each instant to separate each class of MI from the normal group (N). Striking similarities were observed between the three MI groups, particularly at mid-QRS and throughout ST-T. The closest resemblance was between acute NQMI and old QMI. Discriminant analysis was also performed on the 12-lead ECG: The first classification model (N versus MI) produced correct classification rates of 85% for acute QMI and 70% for NQMI. With the second model (MI versus N or LVH), correct rates were 81% and 65%, respectively. CONCLUSIONS: Diagnosis of acute NQMI and QMI (also in the presence of LVH) can be improved substantially by appropriate selection of ECG leads and features. Comparison of discriminant maps from groups A, B, and C does not support the concept of acute NQMI as a distinct ECG entity but rather as a group with infarcts of smaller size. However, pathophysiological and clinical differences between acute NQMI and acute QMI influence long-term risks and may define different therapeutic approaches.

Adult↗
PubMed2002

Conformational memory in photodissociation of formic acid.

The trans and cis forms of formic acid (HCOOH) in solid argon favor essentially different photodissociation (193 nm) products, H2O + CO and H2 + CO2, respectively. The branching ratio of these channels differs between the two conformers by a factor of >10. The observed selective photodissociation features conformational memory when the transition state of a molecule is reached before torsional randomization. These data demonstrate that the photodissociation products can be efficiently steered with selective narrow-band infrared radiation promoting rotational isomerism, which makes a strong case of optically controlled chemical reactions

Journal Article↗
PubMed1992

Structural aspects of recognition motifs contributing to autoimmune responses.

Sequence analysis of autoimmune-associated antibodies has suggested a structural relatedness between genes used to encode autoantibodies and those encoding unrelated antibodies without autoreactive specificities. Subsequently, the basis for cross-reactive idiotypes across germ-line lineages, as well as conserved interspecies cross-reactivities of autoantigens among serologically similar antibodies, may result from evolutionary duplication of particular types of recognition motifs. As a first step toward elucidating structural recognition principles underlying possible cross-reactive epitopes involved in autoimmune pathologies, structural features of selected motifs associated with native ligand binding are examined for their inherent occurrence in antibody and T-cell receptor repertoires. This analysis considers the putative recognition features representative of common motif subsets shared with loop structures in CDR2 and FR3 regions of antibodies such as charge-2x-charge-x-charge or hydrogen bond donor (acceptor)-2x-charge-x-hydrogen bond donor (acceptor) type motifs, where x is any residue that can participate in maintaining a loop conformation. Such tracts encoded in the CRD2 and FR3 regions of heavy chains of antibodies and T-cell receptors (TCRs) associated with autoimmune dysfunction, with non-autoreactive antibodies, and with native host proteins. Such evolutionarily conserved motifs may be targets for complementary interactions involving autoantibodies and receptors.

Amino Acid Sequence↗
PubMed2005

Evidence for recent positive selection at the human AIM1 locus in a European population.

Two missense polymorphisms (E272K and L374F) of the AIM1 locus, encoding a melanocyte differentiation antigen, were shown to have a clear association with human ethnicities. These two nonpathogenic single nucleotide polymorphisms (SNPs) may be associated with human pigmentation variation. In this study, we investigated sequence variation in the coding region and exon-flanking sequence and found low genetic variation only in subjects of European descent. All four statistical tests applied to the 7.55-kb region surrounding the L374F polymorphism detected statistically significant deviations from selective neutrality in Europeans. In addition, haplotype analysis revealed that one haplotype carrying 374F was overrepresented in this population, and the low rate of variation, with some features of selective sweeps, was shown to be statistically significant. These results suggest that positive selection recently has been acting or has acted on at least this region of the melanogenic gene and that an advantageous haplotype spread rapidly in Europe.

Chromatography, High Pressure Liquid↗
PubMed2005

Evolutionary optimization of radial basis function classifiers for data mining applications.

In many data mining applications that address classification problems, feature and model selection are considered as key tasks. That is, appropriate input features of the classifier must be selected from a given (and often large) set of possible features and structure parameters of the classifier must be adapted with respect to these features and a given data set. This paper describes an evolutionary algorithm (EA) that performs feature and model selection simultaneously for radial basis function (RBF) classifiers. In order to reduce the optimization effort, various techniques are integrated that accelerate and improve the EA significantly: hybrid training of RBF networks, lazy evaluation, consideration of soft constraints by means of penalty terms, and temperature-based adaptive control of the EA. The feasibility and the benefits of the approach are demonstrated by means of four data mining problems: intrusion detection in computer networks, biometric signature verification, customer acquisition with direct marketing methods, and optimization of chemical production processes. It is shown that, compared to earlier EA-based RBF optimization techniques, the runtime is reduced by up to 99% while error rates are lowered by up to 86%, depending on the application. The algorithm is independent of specific applications so that many ideas and solutions can be transferred to other classifier paradigms.

Algorithms↗
PubMed1992

Measurement of subvisual changes in cervical squamous metaplastic cells for detecting abnormality.

Quantitative measures of visually normal squamous metaplastic cells exfoliated from the uterine cervix were obtained to test the hypothesis that these cells, like intermediate squamous and endocervical columnar cells, show subvisual evidence of atypia in cases of bonafide squamous intraepithelial neoplasia. The cells identified as squamous metaplastic were obtained from 14 abnormal (dysplastic) and 9 diagnostically negative cases. Although the cell populations so grouped showed no statistically significant differences in overall cell size, nuclear area or nuclear/cytoplasmic ratios, there were significant differences in nuclear and cytoplasmic densitometric features and in nuclear texture features. A combination of three features (nuclear density, texture and cytoplasmic density) permitted 76% of the cells to be categorized correctly as originating in normal versus abnormal slides. It is concluded that selected quantitative features of exfoliated metaplastic cell populations may contribute to improved diagnostic accuracy in automated screening for cervical abnormalities.

Cell Nucleus↗
PubMed2005

Screening mercury levels in fish with a selective fluorescent chemosensor.

Societal concerns over toxic mercury accumulation in humans from fish and other dietary and environmental sources provide motivation to develop new tools and tactics for mercury detection in a wide range of laboratory and field settings. Here we report the synthesis, properties, and application of a selective and sensitive small-molecule chemosensor for fluorescence screening of mercury levels in fish. Mercuryfluor-1 (MF1) is a water-soluble, fluorescein-based reagent that features excellent selectivity for Hg2+ over competing analytes and the largest turn-on fluorescence response to date (>170-fold increase) for reporting this heavy metal ion in aqueous solution. Combining this chemoselective Hg2+ probe with a microwave digestion protocol provides a facile method for assaying mercury levels in fish samples with mercury concentrations spanning 0.1 to 8 ppm, a range well matched with the United States Environmental Protection Agency (U.S. EPA) standard for the maximum safe level of mercury in edible fish (0.55 ppm).

Animals↗
PubMed1993

Clinical pharmacokinetics of selective serotonin reuptake inhibitors.

A feature common to all selective serotonin reuptake inhibitors (SSRIs) is that they are believed to act as antidepressant drugs because of their ability to reversibly block the reuptake of serotonin (5-hydroxytryptamine; 5-HT) in the synaptic cleft. From a chemical perspective, however, they show distinct differences. Consequently, the pharmacokinetic behaviour of of the drugs can be very different, and these pharmacokinetic differences may have a major influence on their clinical profiles of action. All SSRIs have a great affinity for the 5-HT reuptake carrier in the synaptic cleft in the central nervous system, with much less affinity for the noradrenaline (norepinephrine) reuptake carrier, and for alpha- and beta-adrenergic, dopamine, histamine, 5-HT and muscarine receptors. Fluoxetine and citalopram are available as racemic mixtures, the isomers of fluoxetine having almost equal affinity to the 5-HT reuptake carrier, while the reuptake inhibitor properties of citalopram reside almost exclusively in the (+)-isomer. Norfluoxetine, one of the metabolites of fluoxetine, has a selectivity for the 5-HT reuptake carrier comparable with that of fluoxetine. Gastrointestinal absorption of the SSRIs is generally good, with peak plasma concentrations observed after approximately 4 to 6h. Absolute bioavailability of citalopram is almost 100%, whereas it is likely that the other compounds undergo (substantial) first-pass metabolism. Apparent oral clearance values after single doses range from 26 L/h (citalopram) to 167 L/h (paroxetine), while after multiple doses oral clearance is markedly reduced, particularly for fluoxetine and paroxetine. Plasma protein binding of fluoxetine, paroxetine and sertraline is > or = 95%; values for fluvoxamine (77%) and citalopram (50%) are much lower. For all compounds, however, protein binding interactions do not seem to be of great importance. Although many attempts were made, to date no convincing evidence exists of a relationship between plasma concentrations of any of the SSRIs and clinical efficacy. Elimination occurs via metabolism, probably in the liver. Renal excretion of the parent compounds is of minor importance. Metabolites of fluvoxamine and fluoxetine are predominantly excreted in urine; larger quantities of metabolites of paroxetine (36%) and sertraline (44%) are excreted in faeces. The half-lives of fluvoxamine, paroxetine, sertraline and citalopram are approximately 1 day. The half-life of fluoxetine is approximately 2 days (6 days after multiple doses), and that of the active metabolite norfluoxetine is 7 to 15 days. The metabolism of paroxetine, and possibly also of fluoxetine, is under genetic control of the sparteine/debrisoquine type. Available data indicate that metabolism of SSRIs is impaired with reduced liver function.(ABSTRACT TRUNCATED AT 400 WORDS)

Aging↗