A 3-D CAD tool for CT colonography.

Nowadays, biomedical images of the human body are an invaluable source of information that clinicians can use in order to make a correct and final diagnosis. Their use is widespread in almost every branch of medicine. In some cases, computer-aided diagnosis (CAD) tools have received approval for being used without any human intervention and validation, like, for example, in breast cancer detection relying on X-rays mammograms. In this paper we present a framework for building CAD tools based on the well-known software design patterns and an implementation of the first modules of a CAD tool for computed tomography (CT) colonography.

A decision support system for the acquisition and elaboration of EEG signals: the AmI-GRID environment.

This paper describes a data-driven Decision Support System for Electroencephalography (EEG) signals acquisition, and parallel elaboration based on the integration of an Ambient Intelligent (AmI) [1] platform and a GRID enabled Infrastructure. The paper explores the analysis and design of the environment, the real-time data acquisition, the integration of the acquired data in dedicated EHR, and the EEG processing through parallel analysis algorithm available on the GRID infrastructure. After an overview of background concepts, the paper presents a brief description of the environment architecture, and a detailed analysis of the EEG algorithm. The challenge of the work presented is to effectively show how medical data can be shared and processed by exploiting the resources and capabilities of both the AmI platform and the GRID infrastructure. This particular Decision Support System, shows how it is possible to improve patient safety, quality of care, and efficiency in healthcare delivery.

Bioprofiling over grid for eHealthcare.

A trend in modern medicine is towards individualization of healthcare and, potentially, grid computing can play an important role in this by allowing sharing of resources and expertise to improve the quality of care. In this paper, we present a new test bed, the BIOPATTERN Grid, which aims to fulfil this role in the long term. The main objectives in this paper are 1) to report the development of the BIOPATTERN Grid, for biopattern analysis and bioprofiling in support of individualization of healthcare. The BIOPATTERN Grid is designed to facilitate secure and seamless sharing of geographically distributed bioprofile databases and to support the analysis of bioprofiles to combat major diseases such as brain diseases and cancer within a major EU project, BIOPATTERN (www.biopattern.org); 2) to illustrate how the BIOPATTERN Grid could be used for biopattern analysis and bioprofiling for early detection of dementia and for brain injury assessment on an individual basis. We highlight important issues that would arise from the mobility of citizens in the EU, such as those associated with access to medical data, ethical and security; and 3) to describe two grid services which aim to integrate BIOPATTERN Grid with existing grid projects on crawling service and remote data acquisition which is necessary to underpin the use of the test bed for biopattern analysis and bioprofiling.

Cervical ectopic pregnancy: clinical review.

Cervical pregnancy is a rare kind of intrauterine ectopic pregnancy. Diagnosis and treatment of cervical pregnancy have enormously changed in the last 15 years. Before 1980, diagnosis was made when dilation and curettage for presumed incomplete abortion resulted in sudden and uncontrollable hemorrhage. Hysterectomy was practiced in order to save the patient's life. Today, cervical pregnancy is diagnosed by ultrasound (US) during the Ist trimester of pregnancy, so that the patient's fertility can be preserved. Therefore any physician should consider the possibility of a cervical pregnancy in a woman with abdominal pain and vaginal bleeding during the first trimester of pregnancy. In this study we reviewed the literature on the epidemiology, etiology, diagnosis and treatments of cervical pregnancy.

Cervical pregnancy: a case report.

The case of a patient with cervical pregnancy diagnosed by ultrasound (US) at nine weeks of gestation is described. US showed a cavity of 4.5 x 3.5 cm in diameter located on the right lateral wall of the cervix containing trophoblasts. In order to preserve the patient's fertility conservative treatment was administered: methotrexate (MTX), ligature of the uterine arteries, and hysterosuction. Due to intense bleeding uncontrolled by the use of a Foley's catheter total hysterectomy with conservation of the adnexae was performed.

Analysis of the internal representations developed by neural networks for structures applied to quantitative structure--activity relationship studies of benzodiazepines.

An application of recursive cascade correlation (CC) neural networks to quantitative structure-activity relationship (QSAR) studies is presented, with emphasis on the study of the internal representations developed by the neural networks. Recursive CC is a neural network model recently proposed for the processing of structured data. It allows the direct handling of chemical compounds as labeled ordered directed graphs, and constitutes a novel approach to QSAR. The adopted representation of molecular structure captures, in a quite general and flexible way, significant topological aspects and chemical functionalities for each specific class of molecules showing a particular chemical reactivity or biological activity. A class of 1,4-benzodiazepin-2-ones is analyzed by the proposed approach. It compares favorably versus the traditional QSAR treatment based on equations. To show the ability of the model in capturing most of the structural features that account for the biological activity, the internal representations developed by the networks are analyzed by principal component analysis. This analysis shows that the networks are able to discover relevant structural features just on the basis of the association between the molecular morphology and the target property (affinity).

Discriminant pattern recognition using transformation-invariant neurons.

To overcome the problem of invariant pattern recognition, Simard, LeCun, and Denker (1993) proposed a successful nearest-neighbor approach based on tangent distance, attaining state-of-the-art accuracy. Since this approach needs great computational and memory effort, Hastie, Simard, and Säckinger (1995) proposed an algorithm (HSS) based on singular value decomposition (SVD), for the generation of nondiscriminant tangent models. In this article we propose a different approach, based on a gradient-descent constructive algorithm, called TD-Neuron, that develops discriminant models. We present as well comparative results of our constructive algorithm versus HSS and learning vector quantization (LVQ) algorithms. Specifically, we tested the HSS algorithm using both the original version based on the two-sided tangent distance and a new version based on the one-sided tangent distance. Empirical results over the NIST-3 database show that the TD-Neuron is superior to both SVD- and LVQ-based algorithms, since it reaches a better trade-off between error and rejection.

Supervised neural networks for the classification of structures.

Standard neural networks and statistical methods are usually believed to be inadequate when dealing with complex structures because of their feature-based approach. In fact, feature-based approaches usually fail to give satisfactory solutions because of the sensitivity of the approach to the a priori selection of the features, and the incapacity to represent any specific information on the relationships among the components of the structures. However, we show that neural networks can, in fact, represent and classify structured patterns. The key idea underpinning our approach is the use of the so called "generalized recursive neuron", which is essentially a generalization to structures of a recurrent neuron. By using generalized recursive neurons, all the supervised networks developed for the classification of sequences, such as backpropagation through time networks, real-time recurrent networks, simple recurrent networks, recurrent cascade correlation networks, and neural trees can, on the whole, be generalized to structures. The results obtained by some of the above networks (with generalized recursive neurons) on the classification of logic terms are presented.

NEUREX: a tutorial expert system for the diagnosis of neurogenic diseases of the lower limbs.

Specialist tutors have to transfer two types of knowledge to doctors who are specialising in a particular clinic: public declarative knowledge, including facts, notions, principles in that particular field; and their own private procedural knowledge acquired in years of direct experience. Embodying this knowledge into an expert system means that this information can be shared more rapidly, and tutoring is easier. This paper presents a tutorial expert system for neurological clinics which can emulate the diagnostic process of an expert neurologist for neurogenic diseases of the lower limbs, assist users in planning the optimal sequence of NG and EMG tests, interpret the results of these tests, and help users to achieve the most suitable diagnosis.

Sex differences in surface EMG interference pattern power spectrum.

Sex differences in the spectral parameters of the surface electromyogram (EMG) power spectrum were studied during voluntary muscle contractions of different strength with rest in between. The influence of two different types of leads (unipolar and bipolar) on the values of the spectral parameters was also investigated under the same experimental conditions. The subjects were 15 healthy female and 15 healthy male volunteers. The relationship between the amplitude (root mean square) of the EMG and the force developed was not linear. The mean values of the median power frequency were lower in women than in men. With both types of lead, the increase in force was accompanied by a progressive increase in median power frequency in male and female subjects. The significant differences in spectral parameters observed in the two sexes are probably correlated with anatomic differences.

Antigliadin antibody, D-xylose, and cellobiose/mannitol permeability tests as indicators of mucosal damage in children with coeliac disease.

A dual sugar (cellobiose/mannitol) permeability test using an iso-osmolar solution was performed, to compare its ability to predict small-bowel mucosal damage in children affected by coeliac disease with the determination of serum levels of D-xylose and antigliadin antibody. Eighty-three children (67 on gluten-containing diet and 16 on gluten-free diet) were investigated. The D-xylose and the serum antigliadin antibody test predicted accurately 70% and 78% of the small-bowel biopsy results, respectively, whereas the cellobiose-mannitol permeability test predicted 93%. These data confirm the superiority of the permeability test over the D-xylose test, although the former cannot be advocated as a substitute for jejunal biopsy. Our results suggest a complementary use of the permeability test and the antigliadin antibody measurement as screening tests for coeliac disease before applying more invasive procedures.

Differences between surface EMG in male and female subjects evidenced by automatic analysis.

40 healthy volunteers (20 males and 20 females) have been studied by an automatic analysis of their surface EMG. The power density spectrum (PDS) of the electromyographic signal, derived from the tibialis anterior muscle, was used to evaluate the RMS values of the EMG developed during maximal voluntary (Vc) and evoked (Vm) contractions. The ratios between Vc, calculated over each of 5 frequency bands (5-15, 20-40, 45-70, 75-110, 115-160 Hz), and the total Vc have also been calculated. No significant differences emerge in the Vm values for males and females, whereas the Vc values for female subjects are found to be significantly reduced (P less than 0.001) with respect to the corresponding values for males. Significant differences have also been found concerning the percentage distribution of power in the above mentioned frequency bands for men and women (P less than 0.001). It can thus, be hypothesized that there are two different modalities of motor unit recruitment and that different sociological and cultural traditions may be more important in producing these differences than sexually determined physiological differences.

Automatic analysis of surface EMG (preliminary findings in healthy subjects and in patients with neurogenic motor diseases).

Supramaximal electrical stimulation of a motor nerve produces a full contraction of a muscle and the corresponding compound action potential can be recorded. Recent studies appear to support the view that all the motor units are activated during voluntary maximal contraction, at least in the tibialis anterior muscle. The compound action potential and the EMG interference pattern in the tibialis anterior are regarded as two different manifestations of the activation of all the motor units. A method has been developed which compares these EMG activities, by automatic analysis, in order to obtain useful parameters for clinical applications.

EMG responses of the vertical eye muscles to dynamic and static natural vestibular stimulation about different axes in alert rabbits.

The EMG responses of the vertical eye muscles have been recorded in alert intact rabbits submitted to both dynamic and static natural vestibular stimulations about different axes. Following dynamic vestibular stimulation, the phase lead of the EMG response with respect to head position as well as its sensitivity increased with stimulus frequency in all the stimulation modalities. This indicates a progressive recruitment by the stimulus of the second-order vestibular neurons related to semicircular canals. The sensitivity of the response was consistently higher in the effective intermediate roll-pitch modality for all the 4 muscles. Following static stimulation, the EMG response showed an asymmetric modulation. The excitatory response was always higher than the inhibitory one and linearly related with the stimulus. Also for the static stimulation the amplitude of the response was significantly higher when the animal was tilted about the effective intermediate roll-pitch axis for all the 4 muscles. The presence of a maximal EMG response in the same modality for both static and dynamic stimulation indicates a similar spatial organization of those subgroups of ampullar and macular receptors projecting to the same eye muscle.

Phase changes induced by ketamine in the vertical vestibulo-ocular reflex in the rabbit.

The effect of ketamine has been tested on the phase of the vertical vestibulo-ocular reflex of rabbits sinusoidally oscillated at various frequencies. A significant phase lag, predominantly affecting the macular component of the reflex, was observed. This action resembles that induced by Nembutal in the same preparation. A specific action of ketamine on synaptic transmission is suggested. Erroneous phase relationship between natural stimuli responses can be obtained in experiments employing ketamine.

The dynamic discharge properties of oculomotor neurons in rabbits with intact and canal-plugged labyrinths during sinusoidal lateral tilt.

The characteristics of the control exerted by macular and ampullar vestibular receptors on oculomotor neurons (OMN) have been investigated by submitting unanesthetized, encéphale isolé rabbits, with intact and canal-plugged labyrinths, to sinusoidal lateral tilts of varying frequencies (0.013-0.2 Hz). For frequencies of tilt above 0.025 Hz, corresponding to peak angular accelerations suprathreshold for canal related vestibular neurons, both phase and sensitivity of the OMN response differ significantly in the two types of preparation. At lower frequencies of tilt the characteristics of the OMN response are, on the contrary, very similar, thus indicating that also in the intact animals the control of the vertical vestibulo-ocular reflex, in this frequency range, is exerted almost exclusively by the otoliths.

Behavior of the oculomotor neurons discharge during sinusoidal lateral tilt in the rabbit.

The characteristics of the control exerted by macular and ampullar vestibular receptors on oculomotor neurons (OMN) have been investigated by submitting unanesthetized, encéphale isolé rabbits to sinusoidal lateral tilts of varying frequencies (0.013-0.2 Hz). The phase of the response exhibited a progressive shift towards head velocity with increasing frequencies of tilt. The sensitivity of the OMN significantly increased at frequencies above 0.025 Hz, corresponding to peak accelerations suprathreshold for canals related vestibular neurons. The convergent action of macular and ampullar vestibular receptors in the control of vertical eye movements is discussed in relation with stimulus frequency.