Search PubMed⌕ Search

Biomedical subjects

Roland Linder

Publications and source records attributed to Roland Linder.

11 recordsLinked to original sources
PubMed2006

Artificial neural network-based classification to screen for dysphonia using psychoacoustic scaling of acoustic voice features.

SUMMARY: For diagnosis and classification of dysphonia, voice specialists can choose from an array of diagnostic tools like perceptual tests or acoustic voice analysis. These methods have in common that they require a high level of specialized training and experience, and therefore are mostly reserved to specialized centers. We aimed at developing an acoustic voice analysis system that could be used as a screening device to monitor, document, and diagnose voice problems that are also encountered by non-voice specialists, such as anesthesiologists, head and neck surgeons, and general surgeons before surgery of the thyroid gland and the upper thoracic aperture. An acoustical feature extraction paradigm that focused on jitter, shimmer, standard deviation of fundamental frequency, and the glottal-to-noise excitation ratio was used to reanalyse 120 voice samples previously analyzed by Schönweiler et al (A Novel Approach to Acoustical Voice Analysis Using Artificial Neural Networks. JARO. 2000:1;270-282). An improved artificial neural network (ANN) was used for classification. Building on this preliminary work, we modified the mathematical algorithm to further improve classification accuracy. Eighty percent of all voice samples could be classified correctly as either healthy or hoarse (sensitivity: 63.0%; specificity: 93.9%; area under the curve: 0.854). The adaptation of the ANN-voice analysis system for mobile use may facilitate its use and acceptance by non-voice specialists for the discovery and documentation of preexisting voice disorders, and may thereby lead to a timely initiation of further diagnosis and therapy by voice specialists.

Diagnosis, Differential↗
PubMed2006

Low prevalence of transfusion transmitted virus (TTV)-like DNA sequences in cystadenolymphoma and pleomorphic adenoma of the salivary glands.

Titers of transfusion transmitted virus (TTV)-like DNA in saliva samples have been reported 100-1,000 times higher than those of the corresponding sera, suggesting viral transmission by saliva droplets. The present study was conducted to determine whether TTV-like DNA sequence elements play a role in the pathogenesis of cystadenolymphoma or pleomorphic adenoma and if the parotid or the submandibular gland is a major source of TTV persistence. Sixty-two archival salivary gland samples (16 cystadenolymphomas, 13 pleomorphic adenomas, and 33 controls) and 23 corresponding saliva samples were examined using a polymerase chain reaction (PCR) for TTV DNA. All PCR products that displayed DNA bands were sequenced. Leder's stain and immunohistochemistry (anti-CD8, anti-CD20, anti-CD45R0, anti-CD68, and anti-Ki67/MiB1) were applied to detect possible changes associated with findings of TTV-like DNA sequences. Tissue displayed TTV-like DNA sequences in 8.1% (5/62; saliva: 47.8%, 11/23). Tissue that contained TTV-like DNA sequences was histologically indistinguishable from samples lacking such DNA. TTV appears to be only a bystander in cystadenolymphoma, pleomorphic adenoma, and other salivary gland affections. Neither of the glands seems to be a major source of TTV persistence.

Adenolymphoma↗
PubMed2006

Classification of patients with pain based on neuropathic pain symptoms: comparison of an artificial neural network against an established scoring system.

Wider use of pain assessment tools that are specifically designed for certain types of pain--such as neuropathic pain--contribute an increasing amount of information which in turn offers the opportunity to employ advanced methods of data analysis. In this manuscript, we present the results of a study where we employed artificial neural networks (ANNs) in an analysis of pain descriptors with the goal of determining how an approach that uses a specific symptoms-based tool would perform with data from the real world of clinical practice. We also used traditional statistics approaches in the form of established scoring systems as well as logistic regression analysis for the purpose of comparison. Our results confirm the clinical experience that groups of pain descriptors rather than single items differentiate between patients with neuropathic and non-neuropathic pain. The accuracy obtained by ANN analysis was only slightly higher than that of the traditional approaches, indicating the absence of nonlinear relationships in this dataset. Data analysis with ANNs provides a framework that extends what current approaches offer, especially for dynamic data, such as the rating of pain descriptors over time.

Adult↗
PubMed2005

A computer-based approach to assess the perception of composite odour intensity: a step towards automated olfactometry calibration.

The 2004 Nobel Prize in Physiology or Medicine laureates, Richard Axel and Linda Buck, have made smell a less enigmatic sense to study. In clinical routine, olfactory function is assessed using defined concentrations of a single defined substance, a setting which is uncommon in daily life. The present study was therefore conducted to evaluate the applicability of composite odours. Air was contaminated with different quantities of cyclohexanol, cyclohexanone and cyclohexane to generate 73 gas mixtures (one component: n = 21, two components: n = 40, three components: n = 12). The intensity of perception was estimated for each mixture by an average of 60.3 healthy individuals (4,403 assessments). An artificial neural network (ANN) was trained and validated using the contaminants' concentrations with the corresponding estimated intensities. The inter-rater variability was low, as 75.7% of the assessments did not exceed a difference beyond 0.5 from the corresponding median (considered correct predictions). The ANN correctly estimated 78.1% of the gas mixtures, and in terms of the regression task the ANN demonstrated a sufficient prediction performance (Pearson's correlation coefficient r = 0.883; R(2) = 0.757) and outperformed linear regression (r = 0.770; R(2) = 0.667). Evaluating extra ANNs for gas mixtures comprising one, two or three components, the predictive power did not decrease when complexity increased. The aforementioned results reflect nonlinearity in human perception. ANN technology helps simulate human perception of composite odour intensity which may be applicable to olfactometry calibration and systems biological mathematical modelling. The use of composite odours may represent real-life problems more adequately than single substances.

Adult↗
PubMed2005

Breastfeeding duration is determined by only a few factors.

BACKGROUND: According to a representative German study prepared between 1996 and 1998 (SuSe-Study) 90% of the mothers were willing to breastfeed, but only 33% (10%) of the babies were exclusively breastfed up to 4 (6) months of age. Bearing in mind that the period currently recommended for breastfeeding is 6 months, this discrepancy highlights the need for action to identify the causes. The present study investigates the behavioural pattern of mothers 4-5 months after delivery. METHODS: The investigation was carried out as a cross-sectional study of mothers who intended to breastfeed their babies (n = 52). All mothers delivered in a hospital, with maximum medical care, and were interviewed later by telephone using a semi-standardised questionnaire. The differences between actual breastfeeding mothers (BF, n = 30) and non-breastfeeding mothers (not BF, n = 22) were identified. The factors influencing decision-making were determined. RESULTS: Significant differences between the two groups tested (BF and not BF) were identified as the perception of the simplicity of breastfeeding, planned pregnancy, marital status, as well as participation at birth preparation classes. If only factors known prior to birth are applied, the decision to breastfeed can be correctly forecast as being 81%. CONCLUSION: In order to allocate consulting resources more effectively, appropriate concepts need to be developed and promoted. If the present results could be verified by a study with a larger sample, the practical use for resource optimisation in breastfeeding consultations would be very beneficial.

Adolescent↗
PubMed2004

The 'subsequent artificial neural network' (SANN) approach might bring more classificatory power to ANN-based DNA microarray analyses.

MOTIVATION: Human decisions often proceed in two steps. Initially those most preferred are chosen followed by a subsequent choice of these preferences. Applying one artificial neural network (ANN), a classification is limited to the preselection process. The final categorization is only possible by a subsequent ANN that distinguishes the pre-chosen classes. Existing strategies using coupled ANNs are discussed and a new approach particularly suited for multiclass classification problems is introduced ('Subsequent ANN', SANN). RESULTS: Evaluating a simulated data base comprising 3 classes, classification results of SANN were obviously superior to those achieved by ANN. To evaluate a real-world data base the microarray benchmark GCM (14 classes) was chosen. The ANN results reached 72%, comparable to previous results. Using SANN, up to 81% of the tumors were correctly classified. AVAILABILITY: Programs used in this work and numerical results are available upon request.

Algorithms↗
PubMed2004

Factors influencing infant feeding method in an urban community.

The benefits of breastfeeding are well established. However, despite this fact, rates of breastfeeding continue to be low, falling far below the goals of Healthy People 2010. Rates are even lower among ethnic minority and low-income women. In this study, we attempt to identify the factors that most influence a mother's choice of infant feeding method in an urban predominately African-American population. Phone interviews of 70 women who delivered full-term infants at an urban tertiary care hospital were conducted in order to explore knowledge, attitudes, and beliefs about breastfeeding of the mothers and that of members of their social support network. Ten mothers (14%) exclusively breastfed. Older, caucasian, and married women were more likely to breastfeed. Breastfeeding mothers reported more partner support as well as more family knowledge about breastfeeding and had more positive attitudes about breastfeeding. Healthcare providers were not directly influential in mother's feeding choice. From this study, we conclude that in this population, the mother's partner and family are most influential in the choice of infant feeding method and, thus, should be included in breastfeeding promotion programs.

Adolescent↗
PubMed2004

Prevalence of the spondyloarthritides in patients with uveitis.

OBJECTIVE: To determine the prevalence of the spondyloarthritides in patients with uveitis and to establish criteria for certain subtypes of this group of diseases. METHODS: In a university-based prospective study, we examined 433 patients with different types of uveitis for signs of spondyloarthritides as defined by the classification criteria of the European Spondylarthropathy Study Group (ESSG). RESULTS: Forty-four of the 433 uveitis patients (36 with acute anterior uveitis and 8 with other types of uveitis) had spondyloarthritis according to the ESSG classification criteria. Nineteen of these 44 patients had ankylosing spondylitis (AS), 11 had undifferentiated spondyloarthritis, 9 reactive arthritis, and 5 psoriatic arthritis. CONCLUSION: Uveitis is associated not only with AS and reactive arthritis, but also with undifferentiated spondyloarthritis and psoriatic arthritis. The ESSG criteria are helpful in identifying these spondyloarthritides in uveitis patients.

Adolescent↗
PubMed2003

A novel method for diagnosing chronic rhinosinusitis based on an electronic nose.

The nasal out-breath of persons with chronic nasal and/or paranasal infections may have characteristic strange odors, which in our experience are in most cases related to bacterial and/or fungal infections of the sinuses. The objective of the present study was to examine nasal out-breath samples from patients with chronic rhinosinusitis (CRS) (with or without polyposis) and healthy control volunteers using the electronic-nose (EN) technology. We developed a simple technique for collecting samples of nasal out-breath in disposable sterile plastic sacks with a tight closing seal. The principal component analysis correctly classified all individual EN patterns for CRS patients and misclassified 2 samples from the healthy controls (80.0% successful classification rate). The artificial neural network analysis correctly classified 60.0% of the patterns of both groups. We believe that the use of methodologies based on EN technology, combined with conventional clinical examinations, may improve the diagnosis of chronic rhinosinusitis.

Adult↗
PubMedDate not supplied

Simulating soft data to make soft data applicable to simulation.

BACKGROUND: Biomedical processes are often influenced by measures considered "non-crisp", "soft" or "subjective". Despite the growing awareness of the importance of such measures, they are rarely considered in biomedical simulation. This study introduces an input generator for soft data (input generator SD) that makes soft data applicable to simulation. MATERIALS AND METHODS: Machine learning approaches and standard regression techniques were applied to simulate odour intensity ratings. RESULTS: The performance of all the applied methods was satisfactory and the results can be used to modify systems biological mathematical models. CONCLUSION: Soft data should no longer be discounted in systems biological simulations. Exemplarily, it can be demonstrated that the input generator SD produces results that are similar to those that the simulated system can generate. Machine learning and/or appropriate conventional mathematical approaches may be applied to simulate noncrisp processes that can be used to modify mathematical models of any granularity.

Adult↗