Biomedical subjects

J P Braga

Publications and source records attributed to J P Braga.

6 recordsLinked to original sources

PubMed2006

Diffusion coefficient distribution from NMR-DOSY experiments using Hopfield neural network.

Diffusion ordered spectroscopy (DOSY) is a powerful two-dimensional NMR method to study molecular translation in various systems. The diffusion coefficients are usually retrieved, at each frequency, from a fit procedure on the experimental data, considering a unique coefficient for each molecule or mixture. However, the fit can be improved if one regards the decaying curve as a multiexponential function and the diffusion coefficient as a distribution. This work presents a computer code based on the Hopfield neural network to invert the data. One small-molecule binary mixture with close diffusion coefficients is treated with this approach, demonstrating the effectiveness of the method.

Bridged Bicyclo Compounds↗

PubMed2005

Retrieval of transverse relaxation time distribution from spin-echo data by recurrent neural network.

Inversion of transverse relaxation time decay curve from spin-echo experiments was carried out using Hopfield neural network, to obtain the transverse relaxation time distribution. The performance of this approach was tested against simulated and experimental data. The initial guess, necessary for the integration procedure, was established as the analytical Laplace inversion. Together with errors in the simulated data, inversion was also carried out with errors in this initial guess. The probability density function, calculated by the neural network, is used in multiple sclerosis diagnostics.

Humans↗

PubMed2001

Training SVMs with EDR algorithm.

The aim of this work is to present a new training algorithm for SVMs based on the pattern selection strategy called Error Dependent Repetition (EDR). With EDR, the presentation frequency of a pattern depends on its error: patterns with larger errors are selected more frequently and patterns with smaller error (or learned) are presented with minor frequency. Using a simple iterative process based on gradient ascent, SVM-EDR can solve the dual problem without any assumption about support vectors or the Karush-Kuhn-Tucker (KKT) conditions.

Algorithms↗

PubMed2000

[Clinical correlation of angiographic restenosis, isolated or associated with stress test variables, with a population that had stent implantation].

OBJECTIVE: We analyse the value of the clinical data, isolated or associated with the stress EKG (SEKG) variables, in the detection of angiographic restenosis (AR: lesion > 50%), in a group of patients (pts) submitted to coronary angioplasty (PTCA) with stenting. POPULATION AND METHODS: 105 patients were followed prospectively (male: 87%; age: 56 +/- 10); clinical evaluation was performed in the 1st, 3rd and 6th month, SEKG in the 4th month and recatheterization for angiographic control in the 6th month. The vascular risk factors, cause of admission, medication at discharge, angiographic and procedure characteristics were analysed. RESULTS: The most frequent vascular risk factors were hypertension (42%), smoking (64%), and dyslipidemia (52%); 30% had previous ischemic heart disease. The indication for PTCA was myocardial infarction (direct PTCA or residual ischemia) in 42% and angina in 59%. All patients were discharged with AAS and ticlopidine, 38% with beta-blockers, 59% with calcium channel blockers and 92% with nitrates. Thirty-eight percent had multivessel disease. They were submitted to dilatation with the implantation of 128 stents in 199 vessels. An intraaortic balloon pump was used in 13% and Abciximab in 47% of the procedures. In the recatheterization for angiographic control in the sixth month, 30 patients had angiographic restenosis (AR rate = 25.2%). In the clinical evaluation, 22 patients (20.9%) had angina, and 14 of them had angiographic restenosis (clinical restenosis rate = 13.3%). In the 16 asymptomatic patients with AR, 5 had positive SEKG. CONCLUSION: The presence of angina in a group of patients submitted to coronary angioplasty with stenting has a low sensitivity (54%), a good specificity (88%), a positive predictive value of 64% and negative of 83% in the detection of angiographic restenosis. The association with clinical data improves sensitivity (83%), with a decrease in specificity (61%).

Aged↗

PubMed1999

[Angioplasty with stenting in single vessel disease: non invasive methods for detecting restenosis].

OBJECTIVE: To compare the value of the myocardial perfusion imaging with 99Tc sestamibi (MIBI) with clinical data and stress EKG (SEKG) variables, in the detection of angiographic restenosis (AR: lesion > 50%), in a group of patients (pts) with single vessel disease (angiographic lesion above 70% in one epicardial artery), submitted to coronary angioplasty (PTCA) with stenting. POPULATION AND METHODS: Fifty-nine pts, followed prospectively, (49 males and 10 females aged 57 +/- 10) were submitted to 61 angioplasties, with dilatation of 59 vessels, and with stenting in 63 (1.07 stent/vessel). Clinical evaluation was performed in the first, third and sixth month, SEKG with MIBI in the fourth month and recatheterization for angiographic control in the sixth month. The pts were divided in two groups: R-positive angiographic restenosis (AR)--14 pts: NR--without AR--45 pts; the angiographic restenosis rate was 23.7%. The patients' history, cause of admission, procedure characteristics, medication at discharge, clinical events during follow-up, variables of stress EKG associated with myocardial perfusion and angiographic control were analyzed. RESULTS: The cause of admission was the same in both groups (MI in R group: 28.6%; NR: 46.7%; p = ns). There was no difference in the discharge medication. The R group was more symptomatic (Angina: R = 50.0% vs NR = 11.1%; p < 0.001 clinical restenosis rate = 11.8%), and had more revascularization procedures (PTCA: R = 57.1% vs NR = 2.8%; p < 0.001/Bypass surgery: R = 14.3% vs NR = 0%; p = 0.01). This group had abnormal stress EKG more frequently (42.9 vs 20.0%, p = ns) and ischemia in the myocardial perfusion scintigraphy (57.1 vs 6.7% p < 0.001). [table: see text] CONCLUSION: The clinical and diagnostic tests evaluated in this study showed a limited sensitivity in the detection of angiographic restenosis in a group of patients with single vessel disease submitted to coronary angioplasty and stenting. Myocardial scintigraphy was the best method for the detection of angiographic restenosis, showing a high specificity and a reasonable sensitivity.

Angioplasty↗

PubMedDate not supplied

Inversion of simulated positron annihilation lifetime spectrum using a neural network.

Inversion of positron annihilation lifetime spectroscopy, based on a neural network Hopfield model, is presented in this paper. From a previous reported density function for lysozyme in water a simulated spectrum, without the superposition of statistical fluctuation and spectrometer resolution effects, was generated. These results were taken as the exact results from which the neural network was trained. The precision of the inverted density function was analyzed taking into account the number of neurons and the learning time of the neural network. A fair agreement was obtained when comparing the neural network results with the exact results. For example, the maximum of the density function, with a precision of 0.4% for the percentual relative error, was obtained for 64 neurons.

Journal Article↗