A general framework for adaptive processing of data structures.

A structured organization of information is typically required by symbolic processing. On the other hand, most connectionist models assume that data are organized according to relatively poor structures, like arrays or sequences. The framework described in this paper is an attempt to unify adaptive models like artificial neural nets and belief nets for the problem of processing structured information.

On the implementation of frontier-to-root tree automata in recursive neural networks.

In this paper we explore the node complexity of recursive neural network implementations of frontier-to-root tree automata (FRA). Specifically, we show that an FRAO (Mealy version) with m states, l input-output labels, and maximum rank N can be implemented by a recursive neural network with O(radical(log l+log m)lm(N)/log l+N log m) units and four computational layers, i.e.

Inductive inference from noisy examples using the hybrid finite state filter.

Recurrent neural networks processing symbolic strings can be regarded as adaptive neural parsers. Given a set of positive and negative examples, picked up from a given language, adaptive neural parsers can effectively be trained to infer the language grammar. In this paper we use adaptive neural parsers to face the problem of inferring grammars from examples that are corrupted by a kind of noise that simply changes their membership.

Processing directed acyclic graphs with recursive neural networks.

Recursive neural networks are conceived for processing graphs and extend the well-known recurrent model for processing sequences. In Frasconi et al. (1998), recursive neural networks can deal only with directed ordered acyclic graphs (DOAGs), in which the children of any given node are ordered.

Theoretical properties of recursive neural networks with linear neurons.

Recursive neural networks are a powerful tool for processing structured data, thus filling the gap between connectionism, which is usually related to poorly organized data, and a great variety of real-world problems, where the information is naturally encoded in the relationships among the basic entities. In this paper, some theoretical results about linear recursive neural networks are presented that allow one to establish conditions on their dynamical properties and their capability to encode and classify structured information. A lot of the limitations of the linear model, intrinsically related to recursive processing, are inherited by the general model, thus establishing their computational capabilities and range of applicability.

Robust combination of neural networks and hidden Markov models for speech recognition.

Acoustic modeling in state-of-the-art speech recognition systems usually relies on hidden Markov models (HMMs) with Gaussian emission densities. HMMs suffer from intrinsic limitations, mainly due to their arbitrary parametric assumption. Artificial neural networks (ANNs) appear to be a promising alternative in this respect, but they historically failed as a general solution to the acoustic modeling problem.

Development and testing of an algorithm for efficient resource positioning in pre-hospital emergency care.

Response times for pre-hospital emergency care may be improved with the use of algorithms that analyzes historical patterns in incident location and suggests optimal places for pre-positioning of emergency response units. We will develop such an algorithm based on cluster analysis and test whether it leads to significant improvement in mileage when compared to actual historical data of dispatching based on fixed stations.

A study design for comparing electronic patient care report (ePCR) with paper PCR in pre-hospital care.

Data gathering processes in pre-hospital care have been poorly studied. This study aims to compare an electronic PCR with the current paper standard. EMTs will hear a case narrative and enter this information in paper PCR, ePCR or ePCR pre-filled with historical patient data pulled from a database.

[Cardiovascular medicine unit: preliminary results of an in-hospital multidisciplinary intervention in heart failure].

Background: In Italy the majority of patients with heart failure are admitted to Internal Medicine Divisions, where diagnostic and therapeutic procedures, suggested by guidelines, are less utilized than in Cardiology Divisions. In order to provide a better assistance to patients with heart failure, a Cardiovascular Medicine Unit was founded with interdepartmental valence (Cardiovascular and Internal Medicine) in the "Ospedali Riuniti" Hospital of Bergamo. The purpose of this study was to evaluate the preliminary results, concerning management of heart failure, in the first 6 months of activities carried out at the Cardiovascular Medicine Unit and to compare them with those of other structures that usually manage this disease.

Powerful motion illusion caused by temporal asymmetries in ON and OFF visual pathways.

Successive presentations of Glass patterns (randomly positioned pairs of dots oriented in a coherent pattern) create a strong sense of global motion along the orientation of the pattern, but ambiguous in direction. Here we report that dynamic "anti-Glass" patterns, created by successive pairs of globally structured pairs of opposite polarity, create an even more powerful motion illusion that is unambiguous in direction: the dark dots always move toward the light. The motion can be cancelled and reversed by introducing a real delay in the presentation of the light dots, suggesting that the effective stimulation of the light is about 3 ms faster than the dark dots.

Recursive processing of cyclic graphs.

Recursive neural networks are a powerful tool for processing structured data. According to the recursive learning paradigm, the input information consists of directed positional acyclic graphs (DPAGs). In fact, recursive networks are fed following the partial order defined by the links of the graph.

The loading problem for recursive neural networks.

The present work deals with one of the major and not yet completely understood topics of supervised connectionist models. Namely, it investigates the relationships between the difficulty of a given learning task and the chosen neural network architecture. These relationships have been investigated and nicely established for some interesting problems in the case of neural networks used for processing vectors and sequences, but only a few studies have dealt with loading problems involving graphical inputs.

Tako-Tsubo cardiomyopathy: new insights into the possible underlying pathophysiology.

Unlabelled: Tako-Tsubo cardiomyopathy is characterised by an atypical distribution of left ventricular (LV) dysynergy with apical ballooning and compensatory basal hyperkinesis. Coronary angiography is normal. Several substrates have been put forward to explain the underlying pathophysiology such as raised catecholamine levels (due to physical or emotional stress), multivessel epicardial coronary spasm or diffuse microvascular spasm.

End-systolic pressure/volume relationship during dobutamine stress echo: a prognostically useful non-invasive index of left ventricular contractility.

Aims: Left ventricular end-systolic pressure-volume relationship (PVR) provides a robust, relatively load-insensitive evaluation of contractility and can be assessed non-invasively during exercise echo. Dobutamine might provide an exercise-independent alternative approach to assess inotropic reserve. The feasibility of a non-invasive estimation of PVR during dobutamine stress in the echo lab and its relationship with subsequent clinical events was assessed.

Exact and approximate graph matching using random walks.

In this paper, we propose a general framework for graph matching which is suitable for different problems of pattern recognition. The pattern representation we assume is at the same time highly structured, like for classic syntactic and structural approaches, and of subsymbolic nature with real-valued features, like for connectionist and statistic approaches. We show that random walk based models, inspired by Google's PageRank, give rise to a spectral theory that nicely enhances the graph topological features at node level.

Effects of two different glibenclamide dose-strengths in the fixed combination with metformin in patients with poorly controlled T2DM: a double blind, prospective, randomised, cross-over clinical trial.

A double-blind, prospective, randomised, cross-over clinical trial was performed comparing a glibenclamide (G) 5.0 mg/metformin (M) 400 mg combination with a G 2.5 mg/M 400 mg formulation to evaluate whether a higher dose of glibenclamide was able to improve glycaemia in poorly controlled Type 2 diabetic patients.

[Analysis of nucleotide sequences polymorphism of its regions of ribosomal genes of diploid Aegilops (L.) species].

rDNA sequences of ITS region including full sequences of ITS1, 5.8S and partly sequenced ITS2 of 22 accessions of 5 diploid Aegilops species, were determined. The full alignment length was 524 bp.

Artificial neural networks for document analysis and recognition.

Artificial neural networks have been extensively applied to document analysis and recognition. Most efforts have been devoted to the recognition of isolated handwritten and printed characters with widely recognized successful results. However, many other document processing tasks, like preprocessing, layout analysis, character segmentation, word recognition, and signature verification, have been effectively faced with very promising results.

Encoding nondeterministic fuzzy tree automata into recursive neural networks.

Fuzzy neural systems have been a subject of great interest in the last few years, due to their abilities to facilitate the exchange of information between symbolic and subsymbolic domains. However, the models in the literature are not able to deal with structured organization of information, that is typically required by symbolic processing. In many application domains, the patterns are not only structured, but a fuzziness degree is attached to each subsymbolic pattern primitive.

Clarithromycin in rheumatoid arthritis patients not responsive to disease-modifying antirheumatic drugs: an open, uncontrolled pilot study.

Objective: In 1996 we found by serendipity that 2 patients with rheumatoid arthritis (RA) who were taking clarithromycin (CM) to eradicate Helicobacter pylori experienced a regression of their RA symptoms. Following this observation, we tested the hypothesis that this reduction in symptoms could have been caused by CM administration.

Methods: We performed a 6-month, open, uncontrolled pilot study on 18 patients (14 females and 4 males, mean age 62 yrs.

[3H]Ro 15-1788 binding sites to brain membrane of the saltwater Mugil cephalus.

The equilibrium binding parameters of the benzodiazepine antagonist [3H]Ro 15-1788 (8-fluoro-3-carboethoxy-5,6-dihydro-5-methyl-6-oxo-4H-imidazol-[1,5-a]-1,4 benzodiazepine) were evaluated in brain membranes of the saltwater teleost fish, Mugil cephalus. To test receptor subtype specificity, displacement studies were carried out by competitive binding of [3H]Ro 15-1788 against six benzodiazepine receptor ligands, flunitrazepam [5-(2-fluoro-phenyl)-1,3-dihydro-1-methyl-7-nitro-2H-1,4-benzodiazepin-2-one], alpidem [N,N-dipropyl-6-chloro-2-(4-chlorophenyl)imidazo[1,2-a]pyridine-3-acetamide], zolpidem [N,N-6 trimethyl-2-(4-methyl-phenyl)imidazo[1,2-a]pyridine-3-acetamide hemitartrate], and beta-CCM (methyl beta-carboline-3-carboxylate). Saturation studies showed that [3H]Ro 15-1788 bound saturatably, reversibly and with a high affinity to a single class of binding sites (Kd value of 1.

Thyroid carcinoma usually occurs in patients with familial adenomatous polyposis in the absence of biallelic inactivation of the adenomatous polyposis coli gene.

Papillary thyroid carcinoma (PTC) is a rare extracolonic manifestation of familial adenomatous polyposis, determined by germline mutations of the adenomatous polyposis coli (APC) gene. The aim of this study was to assess the presence of loss of heterozygosity of APC in the thyroid tumoral tissue. Specimens from six female patients, aged 20-36, were analyzed for germline and somatic mutations of the APC gene by restriction enzyme analysis and sequence analysis.