PubMed Health⌕ Search

Biomedical subjects

F Vivarelli

Publications and source records attributed to F Vivarelli.

At least 19 recordsLinked to original sources
PubMed2001

Comparing Bayesian neural network algorithms for classifying segmented outdoor images.

In this paper we investigate the Bayesian training of neural networks for region labelling of segmented outdoor scenes; the data are drawn from the Sowerby Image Database of British Aerospace. Neural networks are trained with two Bayesian methods, (i) the evidence framework of MacKay (1992a,b) and (ii) a Markov Chain Monte Carlo method due to Neal (1996). The performance of the two methods is compared to evaluating the empirical learning curves of neural networks trained with the two methods. We also investigate the use of the Automatic Relevance Determination method for input feature selection.

Algorithms↗
PubMed1995

LGANN: a parallel system combining a local genetic algorithm and neural networks for the prediction of secondary structure of proteins.

In this work we describe a parallel system consisting of feed-forward neural networks supervised by a local genetic algorithm. The system is implemented in a transputer architecture and is used to predict the secondary structures of globular proteins. This method allows a wide search in the parameter space of the neural networks and the determination of their optimal topology for the predictive task. Different neural network topologies are selected by the genetic algorithm on the basis of minimal values of mean square errors on the testing set. When the alpha-helix, beta-strand and random coil motifs of secondary structures are discriminated, the maximal efficiency obtained is 0.62, with correlation coefficients of 0.35, 0.31 and 0.37 respectively. This level of accuracy is similar to that previously attained by means of neural networks without hidden layers and using single protein sequences as input. The results validate the neural network topologies used for the prediction of protein secondary structures and highlight the relevance of the input information in determining the limit of their performance.

Algorithms↗
PubMed1995

Predicting free energy contributions to the conformational stability of folded proteins from the residue sequence with radial basis function networks.

Radial basis function neural networks are trained on a data base comprising 38 globular proteins of well resolved crystallographic structure and the corresponding free energy contributions to the overall protein stability (as computed partially from chrystallographic analysis and partially with multiple regression from experimental thermodynamic data by Ponnuswamy and Gromiha (1994)). Starting from the residue sequence and using as input code the percentage of each residue and the total residue number of the protein, it is found with a cross-validation method that neural networks can optimally predict the free energy contributions due to hydrogen bonds, hydrophobic interactions and the unfolded state. Terms due to electrostatic and disulfide bonding free energies are poorly predicted. This is so also when other input codes, including the percentage of secondary structure type of the protein and/or residue-pair information are used. Furthermore, trained on the computed and/or experimental delta G values of the data base, neural networks predict a conformational stability ranging from about 10 to 20 kcal mol-1 rather independently of the residue sequence, with an average error per protein of about 9 kcal mol-1.

Calorimetry↗
PubMed1986

Childhood non-Hodgkin's lymphoma and "leukemia-lymphoma syndrome": long-term results with the modified LSA2-L2 protocol.

From June 1976 to December 1984, 48 previously untreated children with non-Hodgkin's lymphoma (NHL) were treated according to the LSA2-L2 protocol, modified by inclusion of cranial irradiation for patients in stage III and stage IV disease. According to the staging system proposed by Wollner, 4 patients were in stage I, 8 in stage II, 11 in stage III, 8 in stage IVA (less than or equal to 25% blasts in the bone marrow), 15 in stage IVB (greater than 25% blasts in the bone marrow), and 2 in stage IV central nervous system disease. The complete remission rate was 95.8%. The relapse-free survival (RFS) rate of 46 complete responders was 76% after a median observation time of 47+ months. Only 1 of 35 high-risk responder patients developed CNS relapse after prophylactic treatment. Clinical stages were related to the RFS: 100% in stage I-II vs. 69% in stage III-IV. All 8 patients in stage IV were alive without evidence of disease with a median observation time of 59+ months. Fifteen patients in stage IVB who had leukemia-lymphoma syndrome attained 59% RFS with a median observation time of 39+ months. After a median observation time of 38+ months, 29 of 37 patients are off therapy. The results emphasize the value of both the histologic and immunologic features and the stage of disease in predicting the outcome of NHL in children. The LSA2-L2 regimen appears to be a very effective protocol for children with lymphoblastic lymphoma, although it may be less efficacious for patients with large bone marrow involvement.

Antineoplastic Combined Chemotherapy Protocols↗
PubMed1982

Neuroblastoma in a pair of identical twins.

Hereditary neuroblastoma in a pair of identical twins is reported. In one of the twins the tumor occurred in fetal life and caused death of the twin before birth; in the other one, the tumor occurred at two months of life as IV-S neuroblastoma in the liver with primary tumor undetected. The unusual clinical course of the latter case is described: twice there was tumor regression, but in spite of treatment, fatal uncontrolled growth occurred. The family history, characterized by neoplasms of dissimilar cell types but without additional cases of neuroblastoma, is also detailed.

Adrenal Gland Neoplasms↗