PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Digital network”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 559 records · Page 31Linked to original sources
PubMed2006

Morphological development of Aspergillus niger in submerged citric acid fermentation as a function of the spore inoculum level. Application of neural network and cluster analysis for characterization of mycelial morphology.

BACKGROUND: Although the citric acid fermentation by Aspergillus niger is one of the most important industrial microbial processes and various aspects of the fermentation appear in a very large number of publications since the 1950s, the effect of the spore inoculum level on fungal morphology is a rather neglected area. The aim of the presented investigations was to quantify the effects of changing spore inoculum level on the resulting mycelial morphology and to investigate the physiology that underlines the phenomena. Batch fermentations were carried out in a stirred tank bioreactor, which were inoculated directly with spores in concentrations ranging from 10(4) to 10(9) spores per ml. Morphological features, evaluated by digital image analysis, were classified using an artificial neural network (ANN), which considered four main object types: globular and elongated pellets, clumps and free mycelial trees. The significance of the particular morphological features and their combination was determined by cluster analysis. RESULTS: Cell volume fraction analysis for the various inoculum levels tested revealed that by rising the spore inoculum level from 10(4) to 10(9) spores per ml, a clear transition from pelleted to dispersed forms occurs. Glucosamine formation and release by the mycelium appears to be related to spore inoculum level. Maximum concentrations detected in fermentations inoculated with 10(4) and 10(5) spores/ml, where pellets predominated. At much higher inoculum levels (10(8), 10(9) spores/ml), lower dissolved oxygen levels during the early fermentation phase were associated with slower ammonium ions uptakes and significantly lower glucosamine concentrations while the mycelium developed in dispersed morphologies. A big increase in the main and total hyphal lengths and branching frequency was observed in mycelial trees as inoculum levels rise from 10(4) to 10(9) spores/ml, while in aggregated forms particle sizes and their compactness decreased. CONCLUSION: The methods used in this study, allowed for the detailed quantification of the transition between the two extreme morphological forms. The impact of spore inoculum level on the detailed characteristics of the particular morphological forms produced was high. Control of mycelial morphology is often regarded as a prerequisite to ensure increased productivities in industrial applications. The research described here demonstrates that adjusting the spore inoculum level controls effectively mycelial morphology.

Journal Article↗
PubMed2000

Boosting neural networks.

Boosting is a general method for improving the performance of learning algorithms. A recently proposed boosting algorithm, AdaBoost, has been applied with great success to several benchmark machine learning problems using mainly decision trees as base classifiers. In this article we investigate whether AdaBoost also works as well with neural networks, and we discuss the advantages and drawbacks of different versions of the AdaBoost algorithm. In particular, we compare training methods based on sampling the training set and weighting the cost function. The results suggest that random resampling of the training data is not the main explanation of the success of the improvements brought by AdaBoost. This is in contrast to bagging, which directly aims at reducing variance and for which random resampling is essential to obtain the reduction in generalization error. Our system achieves about 1.4% error on a data set of on-line handwritten digits from more than 200 writers. A boosted multilayer network achieved 1.5% error on the UCI letters and 8.1% error on the UCI satellite data set, which is significantly better than boosted decision trees.

Algorithms↗
PubMed1997

Resolution requirements for digital images in dermatology.

BACKGROUND: The digital image has become an important tool in dermatology because of the rapid development of computer hardware, networks, and the World Wide Web. OBJECTIVE: Our purpose was to examine the resolution requirements for digital images. METHODS: Slides of eight selected images showing characteristic lesions were produced in five different resolutions each, ranging from 192 x 128 x 24 (192 pixels x, 128 pixels y, and 24-bit color depth) up to 3072 x 2048 x 24. They were compared side by side by a group of six experienced dermatologists using a standardized questionnaire. RESULTS: Images at the resolution of 768 x 512 x 24 were perceived as equivalent to higher resolutions, whereas a clear difference was visible between 768 x 512 x 24 and 384 x 256 x 24. The slide projector had a significant influence on the rating of the images. CONCLUSION: For digital images in dermatology a resolution of 768 x 512 x 24 is suitable to recognize the relevant details of the source image.

Dermatology↗
PubMed2005

Friends in the 'hood: Should peer-based health promotion programs target nonschool friendship networks?

PURPOSE: To examine the characteristics of inner city African-American adolescents nonschool-based and school-based friendship networks and to explore the influence of these networks on health risk behavior. METHODS: We assessed close friendships networks in a probability sample of inner city African-American youth living in a single neighborhood and describe the networks and health risk behavior of network members. The initial probability sample was obtained via telephone (Random Digit Dialing [RDD] sampling) and followed up with in-person interviews with telephone respondents (seeds). Subsequently, seeds' friends were recruited and completed an in-person interview. RESULTS: A majority of friendship networks included some nonschool friends (57%) and 24% of networks were composed exclusively of nonschool friends. As expected, youth were more likely to spend time with school-based friends on weekdays. On weekends, youth were equally likely to spend time with both school and nonschool networks. Youth in the same friendship group tended to engage in similar behaviors. Health risk behaviors were high regardless of whether networks were nonschool based, mixed, or school based. CONCLUSIONS: The high proportion of nonschool friendships suggests that out-of-school networks may be an important influence in this population. Youth spend time with their friends, regardless of network type, on weekends, and weekends are a high-risk period for health-damaging behaviors. Levels of experience with health risk behaviors suggest that both school and nonschool environments require intervention. Future social influence prevention efforts that are broad-based are likely to have maximal impact.

Adolescent↗
PubMed2003

Picture archiving and fundus imaging in a glaucoma clinic.

Ophthalmological image archiving and distribution can be automated using a picture archiving and communication system (PACS). A fundus PACS has been in clinical use since February 2000 at the ophthalmology clinic of Tampere University Hospital. It consists of a digital fundus camera, an imaging workstation, from which new patients can be added to the archive, 10 viewing stations and an image archive server. In glaucoma imaging, the fundus images taken from a patient are transferred from the imaging workstation to the image archive server and are then immediately available from the physician's viewing workstation; the transfer time of an average image, of 350 kbit, is 0.0035 s, even though the archive is located 5 km away. After 18 months of operation there were over 16,000 images archived; these took 5.3 GByte of a total storage capacity of 41.9 GByte. The network and archive server achieved 99% reliability in use. Digital imaging makes it possible to shift ophthalmology clinics towards more patient-oriented treatment procedures.

Glaucoma↗
PubMed2000

[AM signals' digital demodulation and its application in physiological signals' remote transmission].

Taking advantage of the characteristics of digital systems, a new digital method is given in this article to overcome the difficulties in demodulating AM signals using digital methods. Physiological signals can't be transmitted through telephone networks because they are full of ultra-low frequency components. However, if they are modulated (such as AM), they can be transmitted to the receiving end through telephone networks. Then phaselocked loops can be employed to trace the frequency of the signals' carriers so that we can demodulate and transmit them in the end.

Algorithms↗
PubMed1997

An open distributed medical image and signal data server network with World Wide Web front-end.

In modern patient care and medical research diagnostic imaging from multiple modalities plays an essential role. In general these digital image data are acquired at different locations. An efficient access to the data via network is often not provided due to technical, organisational and political reasons. This article presents an approach to solve the problem by means of an open distributed data server network with Web-based front-ends and describes the prototype DACHS that has been developed at the University Hospital of the Technical University of Munich.

Computer Communication Networks↗
PubMed2005

A novel microstructural approach in tendon viscoelastic modelling at the fibrillar level.

Novel applications in rehabilitation, surgery and tissue engineering require the knowledge of the mechanical behaviour of the tissues at microstructural level. The aim of this work is to investigate the viscoelastic properties of the tendon from the interaction of its biological constituents in the fibrillar network. Traction, relaxation and creep in-vitro tests have been performed on porcine flexor digital tendons. A viscoelastic constitutive equation at finite deformation is presented. The fibrillar deformation modes are described through a network of adaptive links between collagen type I and decorin. The theoretical predictions fit accurately the experimental data. The results of the model demonstrate the mechanical importance of glycosaminoglycan chains of decorin for the differential recruitment and the activation of fibrillar collagen.

Biomechanical Phenomena↗
PubMed1993

Social support and cancer screening among older black Americans.

BACKGROUND: Age-adjusted cancer mortality is 27% higher for Black Americans than for the general U.S. population, which may result from inappropriate use of cancer detection tests. Social support has been shown to affect adjustment to breast cancer and survival, but it has not been studied as a predictor of use of preventive health care services in the older population. Our hypothesis is that larger social networks are associated with greater utilization of cancer-screening tests in the older population. PURPOSE: The objective of this study was to examine the relationship between social support and use of cancer-screening tests among older Black Americans. METHODS: Data for this study were obtained from a 1986 baseline survey evaluation of a community intervention program to increase cancer awareness and a 1991 end-point survey of use of cancer detection tests. Our study sample consisted of 617 Black Americans aged 55 years or older who lived in San Francisco (Calif.), the control community, and in Oakland (Calif.), the target community for intervention. The survey included measures of 1) social network characteristics, as determined by a modified version of Berkman and Syme's Social Network Index; 2) demographic characteristics; and 3) use of six cancer-screening tests--mammography, occult blood stool examination, cervical smear, clinical breast examination, digital rectal examination, and sigmoidoscopy. RESULTS: Multiple logistic regression analysis of the Social Network Index results indicated statistically significant positive associations of social support with the use of mammography and occult blood stool examination but not with the other cancer-screening tests. There were statistically significant associations between having HMO (Health Maintenance Organization) insurance and increased use of mammography and occult blood stool examination, compared with having Medi-Cal or other insurance. The interval between the surveys had a statistically significant positive association with use of mammography. These significant associations were not explained by differences in the other variables, which included health status, age, gender, education, type of health insurance, interval between the surveys, and a regular source of care. CONCLUSIONS: Social support seems to be associated with increased use of mammography and occult blood stool examinations among older Black Americans. IMPLICATIONS: Interventions designed to increase utilization of social networks may be an effective way to increase use of cancer screening, which may ultimately lead to reduced mortality from cancer.

Black or African American↗
PubMed2000

FPNA: interaction between FPGA and neural computation.

Neural networks are usually considered as naturally parallel computing models. But the number of operators and the complex connection graph of standard neural models can not be directly handled by digital hardware devices. More particularly, several works show that programmable digital hardware is a real opportunity for flexible hardware implementations of neural networks. And yet many area and topology problems arise when standard neural models are implemented onto programmable circuits such as FPGAs, so that the fast FPGA technology improvements can not be fully exploited. Therefore neural network hardware implementations need to reconcile simple hardware topologies with complex neural architectures. The theoretical and practical framework developed, allows this combination thanks to some principles of configurable hardware that are applied to neural computation: Field Programmable Neural Arrays (FPNA) lead to powerful neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme. This paper shows how FPGAs have led to the definition of the FPNA computation paradigm. Then it shows how FPNAs contribute to current and future FPGA-based neural implementations by solving the general problems that are raised by the implementation of complex neural networks onto FPGAs.

Computers↗
PubMed2005

Significance analysis of qualitative mammographic features, using linear classifiers, neural networks and support vector machines.

Advances in modern technologies and computers have enabled digital image processing to become a vital tool in conventional clinical practice, including mammography. However, the core problem of the clinical evaluation of mammographic tumors remains a highly demanding cognitive task. In order for these automated diagnostic systems to perform in levels of sensitivity and specificity similar to that of human experts, it is essential that a robust framework on problem-specific design parameters is formulated. This study is focused on identifying a robust set of clinical features that can be used as the base for designing the input of any computer-aided diagnosis system for automatic mammographic tumor evaluation. A thorough list of clinical features was constructed and the diagnostic value of each feature was verified against current clinical practices by an expert physician. These features were directly or indirectly related to the overall morphological properties of the mammographic tumor or the texture of the fine-scale tissue structures as they appear in the digitized image, while others contained external clinical data of outmost importance, like the patient's age. The entire feature set was used as an annotation list for describing the clinical properties of mammographic tumor cases in a quantitative way, such that subsequent objective analyses were possible. For the purposes of this study, a mammographic image database was created, with complete clinical evaluation descriptions and positive histological verification for each case. All tumors contained in the database were characterized according to the identified clinical features' set and the resulting dataset was used as input for discrimination and diagnostic value analysis for each one of these features. Specifically, several standard methodologies of statistical significance analysis were employed to create feature rankings according to their discriminating power. Moreover, three different classification models, namely linear classifiers, neural networks and support vector machines, were employed to investigate the true efficiency of each one of them, as well as the overall complexity of the diagnostic task of mammographic tumor characterization. Both the statistical and the classification results have proven the explicit correlation of all the selected features with the final diagnosis, qualifying them as an adequate input base for any type of similar automated diagnosis system. The underlying complexity of the diagnostic task has justified the high value of sophisticated pattern recognition architectures.

Analysis of Variance↗
PubMed2001

Distance learning in the digital environment.

The expansion of radiology departments and divisions often can not occur in adjacent geographic locations. This leads to a greater separation of staff and residents, as well as workers in similar divisions. This makes traditional teaching difficult in academic institutions. The economic drive forcing many departments to investigate more isolated outpatient imaging centers has further hindered the ability to continue effective academic training at many facilities. The ability to easily share a digital environment across physical distance can greatly enhance the teaching experience, as well as be a valuable tool for consultation and case discussion with referring clinicians. The transition to a filmless environment with picture archiving and communication systems (PACS) can be utilized for distance learning in addition to the clinical arena. It is possible to take advantage of the digital transformation to PACS and case-viewing browser programs to conduct improved interactions with referring clinicians as well as radiologic teaching with relatively minimal hardware and software demands. The integration of web-based teleradiology programs with business networking software can be used for effective distance learning in the digital environment, sufficiently closing the distance on our rapidly expanding departments. This same technology allows for greater interaction with referring clinicians for real-time consultation and enhanced case discussion to entrench a supportive referral base for the radiologic community.

Humans↗
PubMedDate not supplied

Multimedia authenticity protection with ICA watermarking and digital bacteria vaccination.

We propose the application of independent component analysis (ICA), via unsupervised neural networks, to authenticity protection for multimedia products. We give an overview of the current state of multimedia authenticity protection, including the requirements of various multimedia applications, current approaches to the problem, and the robustness of the approaches. For watermark security, a covert independent-component watermarking signal can serve as a 'vaccination' against a dormant digital 'bacteria' protecting the multimedia data. An unauthorized removal of the watermark triggers the bacteria payload, which then degrades the quality of the unauthorized data. We argue that such digital bacteria meet all the established requirements for beneficial virus-like programs, and their payload would merely affect pirated media. We show how these new approaches contribute to a flexible, robust, and secure system for protecting the authenticity of multimedia products.

Computer Security↗
PubMed1997

Predicting joint moments and angles from EMG signals.

Electromyographic (EMG) signals are a direct reflection of muscle activity. Efforts have been made to extract additional information from them regarding the kinematic results of muscle use, with limited success. The purpose of this study was to conduct a wheelchair propulsion experiment using a limited number of subjects to gather information for developing a neural network model to correlate EMG signals to kinematic features of the wrist, elbow, and shoulder joints. The trained model was used to predict joint dynamics from previously unseen EMG data. EMG signals were collected from four muscle groups while five able-bodied subjects propelled a wheelchair. The procedure was videotaped with the kinematics and dynamics of the three joints obtained by digitizing and transforming the images into numerical data. A neural network model, based on the back-propagation algorithm, was built and trained on the collected data. The net's predicted values were close to the observed values, particularly for the joint moments (within 7% of actual values). The results demonstrate the potential of neural networks for predicting the movement patterns of wheelchair users from a small number of subjects.

Adult↗