Semeion researchers have developed and used different kinds of Artificial Neural Networks (ANN) in order to process selected, "standard" data coming from drug users and from people who never used drugs before. In the first step a collection of 112 general variables, not traditionally connected to drug user's behavior, were collected from a sample of 545 people (223 heroin addicted and 322 non-users). Different types of ANNs were used to test the capability of the system to classify the drug users and the non-drug users correctly. A special ANN tool, created by Semeion, was also used to prune the number of the independent variables. The ANN selected for this first experiment was a Supervised Feed Forward Network, whose equations were enhanced by Semeion researchers. For the validation of the capability of generalization of the ANN, the Training-Testing protocol was used. This ANN was able, in the Testing phase, to classify approximately 95% of the sample with accuracy. A special sensitivity tool selected only 47 among the 112 independent variables as necessary to train the ANN. In the second step, different types of ANN were tested on the new 47 variables to decide which kind of ANN was better able to classify the sample. This benchmark included the following ANNs: a) Back Propagation with Soft Max; b) Learning Vector Quantization; c) Logicon Projection; d) Radial Basis Function; e) Squash (Semeion Network); f) Fuzzy Art Map; g) Modular Neural Network. In the third step a Constraint Satisfaction Network, specifically created by Semeion, was used to simulate a dynamic fuzzy map of the drug user's world; that is, which fuzzy, or approximate, variables are critical to decide the fuzzy membership of a subject from the fuzzy membership of the drug users to the fuzzy membership of non-users and vice versa.

Artificial Intelligence↗

PubMed1998

Application of artificial neural networks to eating disorders.

An experimental application of Artificial Neural Networks to Eating Disorders is presented. The sample, composed of 172 cases (all women) collected at the Centre for the Diagnosis and Treatment of Eating Disorders of the 1st Medical Division of the St. Eugenio Hospital of Rome, was subdivided, on the basis of the diagnosis made by the specialist of the St. Eugenio, into four classes: Anorexia Nervosa (AN), Nervous Bulimia (NB), Binge Eating Disorders (BED) and Psychogenic Eating Disorders that are Not Otherwise Specified (PED-NOS). The data base was composed of 124 different variables: generic information, alimentary behavior, eventual treatment and hospitalization, substance use, menstrual cycles, weight and height, hematochemical and instrumental examinations, psychodiagnostic tests, etc. The goal of this experiment was to verify the accuracy of the Neural Networks in recognising anorexic and bulimic patients. This article describes 6 experiments, using a Feed Forward Neural Network, each one using different variables. Starting from only the generic variables (life styles, family environment, etc.) and hematoclinical and instrumental examinations, a Neural Networks provided 86.94% of the prediction precision. This work is meant to be a first contribution to creating diagnostic procedures for Eating Disorders, that would be simple and easy-to-use by professionals who are neither psychologists nor psychiatrists nor psychotherapists but who are, however, among the first to meet these patients and who are therefore called upon to give such patients the very first pieces of advice on seeking proper treatment.

Anorexia Nervosa↗

PubMed1998

A neural network investigation of the crucial facets of urban sustainability.

This paper focuses on the concept of a sustainable city and its theoretical implications for the Italian urban system. Urban sustainability is based on positive interactions among three different urban subsystems: social, economic and physical, where social well-being coexists with economic development and environmental quality. This utopian scenario does not appear in the existing cities. The aesthetic quality of natural and man-made environment is often associated with marginality and poverty, labor market variety and urban efficiency coexisting with pollution, criminality and high settlement costs. Moreover, since each city differs institutionally, historically, culturally and economically, few attempts have been implemented to build a comparative synthetic vision of the urban sustainability in different cities. The interactions among these selected systems are complex and unpredictable and present the opportunity for a new methodology of scientific investigation: the connectionistic approach. The dual aim of this study is to: investigate the underlying relationships among the three subsystems with a set of social, economic and physical attributes of the chief towns of a Province in Italy and ; verify if this underlying structure could reproduce the heterogeneity of urban realities, allowing one to distinguish groups of cities with different assets or drawbacks in their sustainability. The Data Base (DB), composed of 43 attributes for 95 cities, was processed by Self-Reflexive Neural Networks (SRNN) (Buscema, 1995). These Networks are a useful instrument of investigation and analogic questioning of the Data Base. Once the SRNN has learned the structure of the weights from the DB, by querying the network with the maximization or minimization of specific groups of attributes, it is possible to read the related properties and to rank the cities' urban profile.

Cities↗

PubMed1998

[Iodine deficiency and iodine prophylaxis experience in Sicily].

Twenty years ago we described an area of goiter endemia in North-Eastern Sicily. In this area endemic goiter was associated to a variable degree of iodine deficiency and, in some places, also to an increased thiocyanate urine excretion. Our studies have demonstrated a strict relationship between iodine deficiency and congenital hypothyroidism (both permanent and transient), an increased prevalence of autonomously functioning thyroid nodules and an increased prevalence of thyroid cancer (follicular and anaplastic histotypes). A number of cases of endemic cretinism have also been described. An active iodine prophylaxis program has been carried in the town of Troina in the years 1979-87, by iodinating the municipal water supply. This intervention caused the disappearance of goiter in schoolchildren in only five years. In the last 20 years the prevalence of goiter has decreased in all endemic areas probably because of the "silent prophylaxis", due to improved socio-economic conditions and industrial food consumption. Today the persistence of endemic goiter confirms the inadequacy of the silent iodine prophylaxis and the need to immediately introduce an active iodine prophylaxis in Sicily.

Biomarkers↗

PubMed1998

Artificial neural networks and complex social systems. I. Theory.

Explore the source record for details and available documents.

Algorithms↗

PubMed1997

A general presentation of artificial neural networks. I.

This article, the first in a series, presents the use of a new processing technology: Artificial Neural Networks (ANN) Modeling. We introduce the reader, in an easy way, to types of problems for which researchers can use this technology. ANN represents a powerful way for understanding, predicting, and simulating complex and chaotic dynamics. We believe that social fields, human psychology, and behaviors, which are nonlinear in their essence, need an adequate methodology such as ANN in order to best investigate and comprehend them.

Computer Simulation↗

PubMed1997

The Sonda Project: prevention, prediction, and psychological disorder.

The Sonda Project began in Italy in 1985 and was created by Semeion Research Center of Sciences of Communication. It has been operating in approximately 30 Italian cities, ranging from large to small, since 1986. The aim of the Sonda Project is to undertake primary prevention measures at the level of personality organization disorder. Cognitive and communication disorders are, in the context of the Sonda Project, the common ground for many expressions of cultural disorders labeled drug addiction, alcoholism, violence, etc. From the point of view of the Sonda Project, measures aimed at combating drug-addiction and alcoholism are no more than superficial strategies incapable of dealing with the causes of self and hetero-destruction (of others). Such abuses are the most evident political symptoms. The methodological aim of the Sonda Project, therefore, consists in understanding, predicting, and acting upon the incubation disorder of human beings; a disorder which subsequently transforms itself into more or less recognizable symptoms. In order to attain its aim, the Sonda Project has made use of two instruments: 1) a series of Intelligent Computerized Observatories, able to predict, in each community, the typology of people who are more likely to exhibit forms of destructive suffering; 2) a local team able to use such advanced technology and able to utilize territorial activation techniques. These are designed to reinsert into the arena of the social game those people who are more "at risk" in terms of self and heterodestruction, and who are most exposed to psychological and sociological levels of relational "fatigue."

Health Promotion↗

PubMed1995

Copper addition prevents the inhibitory effects of interleukin 1-beta on rat pancreatic islets.

Since copper [Cu(II)] is a necessary cofactor for both intra-mitochondrial enzymes involved in energy production and hydroxyl scavenger enzymes, two hypothesised mechanisms for action of interleukin-I beta (IL-1 beta), we studied whether Cu(II) addition could prevent the inhibitory effect of IL-1 beta on insulin release and glucose oxidation in rat pancreatic islets. Islets were incubated with or without 50 U/ml IL-1 beta, in the presence or absence of various concentrations of Cu(II)-GHL (Cu(II) complexed with glycyl-L-histidyl-L-lysine, a tripeptide known to enhance copper uptake into cultured cells). CuSO4 (1-1000 ng/ml) was used as a control for Cu(II) effect when present as an inorganic salt. At the end of the incubation period, insulin secretion was evaluated in the presence of either 2.8 mmol/l (basal insulin secretion) or 16.7 mmol/l glucose (glucose-induced release). In control islets basal insulin secretion was 92.0 +/- 11.4 pg.islet-1 h-1 (mean +/- SEM, n = 7) and glucose-induced release was 2824.0 +/- 249.0 pg.islet-1 h-1. In islets pre-exposed to 50 U/ml IL-1 beta, basal insulin release was not significantly affected but glucose-induced insulin release was greatly reduced (841.2 +/- 76.9, n = 7, p < 0.005). In islets incubated with IL-1 beta and Cu-GHL (0.4 mumol/l, maximal effect) basal secretion was 119.0 +/- 13.1 pg.islet-1 h-1 and glucose-induced release was 2797.2 +/- 242.2, (n = 7, p < 0.01 in respect to islets exposed to IL-1 beta alone).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1995

Interleukin-1 beta inhibition of insulin release in rat pancreatic islets: possible involvement of G-proteins in the signal transduction pathway.

In vitro exposure of rat pancreatic beta cells to interleukin-1 beta (IL-1 beta) inhibits glucose-stimulated insulin release (2140 +/- 239 and 323 +/- 80 pg.islet-1.h-1 at glucose levels of 16.7 mmol/l in control and IL-1 beta-exposed islets, respectively, n = 7, p < 0.001). Cholera toxin (2 micrograms/ml) or pertussis toxin (0.5 microgram/ml) potentiated, as expected, glucose-induced insulin release in control islets, but, in addition, when added together with IL-1 beta, were able to prevent the IL-1 beta mediated inhibition of glucose-stimulated insulin secretion (2087 +/- 301 and 1662 +/- 173 pg.islet-1.h-1, respectively, p < 0.05 vs islets exposed to IL-1 beta alone). To investigate the mechanism by which the toxins prevent the IL-1 beta effect, we then measured nitrite levels, glucose oxidation and Ca2+ uptake. Nitrite levels in the culture medium were 4.2 +/- 1.4 and 24.0 +/- 5 pmol.islet-1.24 h-1 in control islets and in IL-1 beta-exposed islets, respectively (n = 6, p = 0.05). In islets exposed to IL-1 beta and cholera or pertussis toxins, nitrite levels were 9.1 +/- 3 and 12.4 +/- 6 pmol.islet-1.24 h-1, respectively (n = 6, NS vs control islets). Glucose oxidation at 16.7 mmol/l glucose was 31.1 +/- 2.9 pmol.islet-1.120 min-1 in control islets and 16.8 +/- 2.7 pmol.islet-1.120 min-1 in IL-1 beta-treated islets (p < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1995

Glucose transport, phosphorylation, and utilization in isolated porcine pancreatic islets.

Porcine islets have been proposed as a donor source for human transplantation, mainly because of both structural and biological similarities of porcine and human insulin. However, the in vitro function of these islets is poorly characterized. In the present study, we first examined insulin release in response to glucose in static incubation experiments. Increasing glucose concentrations up to 8.3 mmol/L stimulated insulin release; however, this elevation was only twofold, and a paradoxical decline was observed at glucose concentrations higher than 8.3 mmol/L. In cultured porcine islets, a greater insulin secretion may be elicited by agents that increase intracellular cyclic adenosine monophosphate (cAMP) levels. To investigate the possible reasons for the porcine islet low response to glucose in vitro, we then evaluated in parallel experiments glucose transport, phosphorylation, and utilization. Glucose transport studies (using 3-O-methyl glucose uptake at 15 degrees C for 15 seconds) indicated the presence of both a high-affinity (Km, 1.2 +/- 0.6 mmol/L) and a low-affinity (Km, 11.8 +/- 1.9 nmol/L, n = 5) component. Glucose phosphorylation, evaluated by measuring the rate of glucose-6-phosphate formation in a fluorimetric assay, indicated that glucokinase activity had a maximum (Vmax) of 7.97 +/- 0.94 nmol/microgram DNA/h and a Km of 8.3 +/- 0.9 mmol/L (mean +/- SE, n = 8).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1993

Different effects of glucose and glyburide on insulin secretion in rat pancreatic islets pre-exposed to interleukin-1 beta. Possible involvement of K+ and Ca2+ channels.

In vitro islet exposure to interleukin 1 beta inhibits the beta-cell response to glucose. We have studied whether a similar inhibition also occurs in response to the sulphonylurea glyburide. Rat pancreatic islets were cultured for 24 h in the presence or absence of 50 U/ml interleukin 1 beta and then stimulated with either glucose or glyburide for 1 h at 37 degrees C. In control islets basal insulin secretion was 117 +/- 32 pg.islet-1.h-1 (mean +/- SEM, n = 7) and greatly increased in response to 16.7 mmol/l glucose (2140 +/- 293) or 10 mumol/l glyburide (1464 +/- 234). When islets were pre-exposed to interleukin 1 beta, insulin release was significantly reduced in response to glucose (323 +/- 80, p < 0.001) but not in response to glyburide (1316 +/- 185). Since both glucose and glyburide influence beta-cell K+ and Ca2+ efflux, to further investigate this different response in islets exposed to interleukin 1 beta we measured both Rb+ efflux (as index of the ATP-sensitive K+ channel activity) and Ca2+ uptake. In control islets, the increased insulin secretion in response to 16.7 mmol/l glucose or 10 mumol/l glyburide was associated with a reduction of 86Rb efflux (decrement of -50 +/- 1.2% and -49 +/- 2.3%, respectively, mean +/- SEM, n = 5). In contrast, in interleukin 1 beta pre-exposed islets both glucose and glyburide stimulation only slightly modified 86Rb efflux (decrement of -19 +/- 1.9% and -5.3 +/- 3.1%, respectively, n = 5, p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1993

Inhibition of the high-affinity glucose transporter GLUT 1 affects the sensitivity to glucose in a hamster-derived pancreatic beta cell line (HIT).

HIT is a hamster-derived beta-cell line which in contrast to normal beta cells that only express the high Km GLUT-2 glucose transporter, also expresses the low Km glucose transporter GLUT 1. In HIT cells the abnormal glucose transport mechanism is associated with a marked shift to the left of the glucose-induced insulin release dose-response curve. We have used this cell model to investigate whether changes in glucose transport affect the glucose-induced insulin release. HIT cells were first incubated with a concentration of cytochalasin B (0.4 mumol/l) that selectively inhibits the GLUT-1 but not the GLUT-2 transporter. The consequences of blocking glucose phosphorylation and insulin release were studied. Exposure to 0.4 mumol/l cytochalasin B for 1 h caused a selective loss of the low Km transport: the calculated Vmax of GLUT 1 was reduced from 1726 +/- 98 to 184 +/- 14 pmol.mg protein-1 5 min-1 (mean +/- SEM, n = 6, p < 0.005), while no major difference in the high Km (GLUT-2) transport was observed. In cytochalasin B exposed HIT cells the glucose phosphorylating activity (due to hexokinase and glucokinase) was unaffected.(ABSTRACT TRUNCATED AT 250 WORDS)

3-O-Methylglucose↗