Voice recognition has become a more focused and researched field in the last century,and new techniques to identify speech has been introduced. A part of voice recognition isspeaker verification which is divided into Front-end and Back-end. The first componentis the front-end or feature extraction where techniques such as Mel-Frequency CepstrumCoefficients (MFCC) is used to extract the speaker specific features of a speech signal,MFCC is mostly used because it is based on the known variations of the humans ear’scritical frequency bandwidth. The second component is the back-end and handles thespeaker modeling. The back-end is based on the Gaussian Mixture Model (GMM) andGaussian Mixture Model-Universal Background Model (GMM-UBM) methods forenrollment and verification of the specific speaker. In addition, normalization techniquessuch as Cepstral Means Subtraction (CMS) and feature warping is also used forrobustness against noise and distortion. In this paper, we are going to build a speakerverification system and experiment with a variance in the amount of training data for thetrue speaker model, and to evaluate the system performance. And further investigate thearea of security in a speaker verification system then two methods are compared (GMMand GMM-UBM) to experiment on which is more secure depending on the amount oftraining data available.This research will therefore give a contribution to how much data is really necessary fora secure system where the False Positive is as close to zero as possible, how will theamount of training data affect the False Negative (FN), and how does this differ betweenGMM and GMM-UBM.The result shows that an increase in speaker specific training data will increase theperformance of the system. However, too much training data has been proven to beunnecessary because the performance of the system will eventually reach its highest point and in this case it was around 48 min of data, and the results also show that the GMMUBM model containing 48- to 60 minutes outperformed the GMM models.

Malaria is a devastating global health problem that affected 219 million people and caused 660,000 deaths in 2010. Inaccurate estimation of the level of infection may have adverse clinical and therapeutic implications for patients, and for epidemiological endpoint measurements. The level of infection, expressed as the parasite density (PD), is classically defined as the number of asexual parasites relative to a microliter of blood. Microscopy of Giemsa-stained thick blood smears (TBSs) is the gold standard for parasite enumeration. Parasites are counted in a predetermined number of high-power fields (HPFs) or against a fixed number of leukocytes. PD estimation methods usually involve threshold values; either the number of leukocytes counted or the number of HPFs read. Most of these methods assume that (1) the distribution of the thickness of the TBS, and hence the distribution of parasites and leukocytes within the TBS, is homogeneous; and that (2) parasites and leukocytes are evenly distributed in TBSs, and thus can be modeled through a Poisson-distribution. The violation of these assumptions commonly results in overdispersion. Firstly, we studied the statistical properties (mean error, coefficient of variation, false negative rates) of PD estimators of commonly used threshold-based counting techniques and assessed the influence of the thresholds on the cost-effectiveness of these methods. Secondly, we constituted and published the first dataset on parasite and leukocyte counts per HPF. Two sources of overdispersion in data were investigated: latent heterogeneity and spatial dependence. We accounted for unobserved heterogeneity in data by considering more flexible models that allow for overdispersion. Of particular interest were the negative binomial model (NB) and mixture models. The dependent structure in data was modeled with hidden Markov models (HMMs). We found evidence that assumptions (1) and (2) are inconsistent with parasite and leukocyte distributions. The NB-HMM is the closest model to the unknown distribution that generates the data. Finally, we devised a reduced reading procedure of the PD that aims to a better operational optimization and a practical assessing of the heterogeneity in the distribution of parasites and leukocytes in TBSs. A patent application process has been launched and a prototype development of the counter is in process.

Made available in DSpace on 2014-07-29T15:29:20Z (GMT). No. of bitstreams: 1 Dissertacao Edna Joana Claudio Manrique.pdf: 2113525 bytes, checksum: 8bf51386d65f55712ef0067bf65249de (MD5) Previous issue date: 2008-03-16 / This study compared the effectiveness of quick review 100%, 10% random review and revision of smears selected based on clinical criteria as a method of internal quality control of cervical smears negative on routine screening and found that the false-negative results identified by review rapid vary with the quality of the sample and the woman's age. Was based on the female population user of the Unified Health System in Goiânia-GO, which was subjected to cytological examination performed at the Laboratory of Clinical Analysis of the Romulus Rock School of Pharmacy, Federal University of Goiás The 5,530 smears classified as negative scrutiny underwent routine method for quick review 100%, then the smears were selected based on clinical criteria and 10% of negative smears and submitted to the respective revisães. Four cytologists alternated during revisions. After the revisions concordant results were considered final diagnosis, whereas the divergent were analyzed by a third cytologist, which consensus meeting defined the final diagnosis. All steps were blindly and the results classified according to the Bethesda System. For statistical analysis, the variables were analyzed descriptively and applied the chi-square test. A Quick Review of 100% smears identified 141 suspects, 84 of these (59.6%) tested positive for the final diagnosis, of which 36 (25.5%) were classified as atypical squamous cells of undetermined significance (ASC-US) five (3.5%) as atypical squamous cells can not exclude high-grade lesion (ASC-H), 34 (24.1%) as squamous intraepithelial lesions of low-grade (LSIL), six (4.3 %) as squamous intraepithelial lesion, high-grade (HSIL) and three (2.1%) atypical glandular cells (AGC). Smears of 84 suspected and confirmed the final diagnosis, 62 (73.8%) were classified as satisfactory and 22 (26.2%) satisfactory, but with some limitation, but no significant difference was with the woman's age. The method of rapid review identified 19 as unsatisfactory smears, 16 of these (84.2%) were confirmed in the final diagnosis. Of the 1,279 smears reviewed based on clinical criteria, 24 were considered abnormal. Of these 19 (79.2%) were confirmed as positive by the final diagnosis, nine (47.3%) classified as ASC-US, one (5.3%) ASC-H, eight (42.1%) and LSIL One (5.3%) HSIL. Of the 560 smears examined by reviewing 10%, 13 were considered abnormal, these six (46.2%) were confirmed in the final diagnosis, four (30.8%) classified as ASC-US, one (7.7%) and LSIL (7.7%) HSIL. Therefore, a quick review of 100% is a more efficient alternative for the detection of false-negative results of cervical screening as a method of internal quality control, than the random review of 10% and revision based on clinical criteria. It was observed also that the quick review showed better performance when the sample was classified as adequate for analysis, but does not vary with the age of the woman. / This study compared the effectiveness of quick review 100%, 10% random review and revision of smears selected based on clinical criteria as a method of internal quality control of cervical smears negative on routine screening and found that the false-negative results identified by review rapid vary with the quality of the sample and the woman's age. Was based on the female population user of the Unified Health System in Goiânia-GO, which was subjected to cytological examination performed at the Laboratory of Clinical Analysis of the Romulus Rock School of Pharmacy, Federal University of Goiás The 5,530 smears classified as negative scrutiny underwent routine method for quick review 100%, then the smears were selected based on clinical criteria and 10% of negative smears and submitted to the respective revisães. Four cytologists alternated during revisions. After the revisions concordant results were considered final diagnosis, whereas the divergent were analyzed by a third cytologist, which consensus meeting defined the final diagnosis. All steps were blindly and the results classified according to the Bethesda System. For statistical analysis, the variables were analyzed descriptively and applied the chi-square test. A Quick Review of 100% smears identified 141 suspects, 84 of these (59.6%) tested positive for the final diagnosis, of which 36 (25.5%) were classified as atypical squamous cells of undetermined significance (ASC-US) five (3.5%) as atypical squamous cells can not exclude high-grade lesion (ASC-H), 34 (24.1%) as squamous intraepithelial lesions of low-grade (LSIL), six (4.3 %) as squamous intraepithelial lesion, high-grade (HSIL) and three (2.1%) atypical glandular cells (AGC). Smears of 84 suspected and confirmed the final diagnosis, 62 (73.8%) were classified as satisfactory and 22 (26.2%) satisfactory, but with some limitation, but no significant difference was with the woman's age. The method of rapid review identified 19 as unsatisfactory smears, 16 of these (84.2%) were confirmed in the final diagnosis. Of the 1,279 smears reviewed based on clinical criteria, 24 were considered abnormal. Of these 19 (79.2%) were confirmed as positive by the final diagnosis, nine (47.3%) classified as ASC-US, one (5.3%) ASC-H, eight (42.1%) and LSIL One (5.3%) HSIL. Of the 560 smears examined by reviewing 10%, 13 were considered abnormal, these six (46.2%) were confirmed in the final diagnosis, four (30.8%) classified as ASC-US, one (7.7%) and LSIL (7.7%) HSIL. Therefore, a quick review of 100% is a more efficient alternative for the detection of false-negative results of cervical screening as a method of internal quality control, than the random review of 10% and revision based on clinical criteria. It was observed also that the quick review showed better performance when the sample was classified as adequate for analysis, but does not vary with the age of the woman. / Este estudo comparou a eficiência da revisão rápida de 100%, revisão aleatória de 10% e revisão de esfregaços selecionados com base em critérios clínicos como método de controle interno da qualidade dos esfregaços cervicais negativos no escrutínio de rotina e verificou se os resultados falso-negativos identificados pela revisão rápida variam com a adequabilidade da amostra e com a idade da mulher. Teve como base à população feminina usuária do Sistema Único de Saúde de Goiânia-GO, que se submeteu ao exame citopatológico realizado no Laboratório de Análises Clínicas Rômulo Rocha da Faculdade de Farmácia da Universidade Federal de Goiás. Os 5.530 esfregaços classificados como negativos no escrutínio de rotina foram submetidos ao método de revisão rápida de 100%, em seguida, foram selecionados os esfregaços com base em critérios clínicos e 10% do total de esfregaços negativos e submetidos às respectivas revisães. Quatro citologistas se alternaram durante as revisões. Após as revisões os resultados concordantes foram considerados como diagnóstico final, enquanto que, os divergentes foram analisados por um terceiro citologista, que em reunião de consenso definiu o diagnóstico final. Todas as etapas foram às cegas e os resultados classificados de acordo com o Sistema de Bethesda. Para análise estatística as variáveis foram estudadas de maneira descritiva e aplicado o Teste do Qui-quadrado. A revisão rápida de 100% identificou 141 esfregaços suspeitos, desses 84 (59,6%) foram considerados positivos pelo diagnóstico final, dos quais 36 (25,5%) foram classificados como células escamosas atípicas de significado indeterminado (ASC-US), cinco (3,5%) como células escamosas atípicas, não podendo excluir lesão de alto grau (ASC-H), 34 (24,1%) como lesão intra-epitelial escamosa de baixo grau (LSIL), seis (4,3%) como lesão intra-epitelial de alto grau (HSIL) e três (2,1%) células glandulares atípicas (AGC). Dos 84 esfregaços suspeitos e confirmados pelo diagnóstico final, 62 (73,8%) foram classificados como satisfatório e 22 (26,2%) satisfatórios, porém com alguma limitação, mas não observou diferença significativa com a idade da mulher. O método de revisão rápida identificou 19 esfregaços como insatisfatórios, desses 16 (84,2%) foram confirmados pelo diagnóstico final. Dos 1.279 esfregaços revisados com base em critérios clínicos, 24 foram considerados alterados. Desses 19 (79,2%) foram confirmados como positivos pelo diagnóstico final, sendo nove (47,3%) classificados como ASC-US, um (5,3%) ASC-H, oito (42,1%) LSIL e um (5,3%) HSIL. Dos 560 esfregaços analisados pela revisão de 10%, 13 foram considerados alterados, desses seis (46,2%) foram confirmados pelo diagnóstico final, sendo quatro (30,8%) classificados como ASC-US, um (7,7%) LSIL e um (7,7%) HSIL. Portanto, a revisão rápida de 100% é uma alternativa mais eficiente na detecção de resultados falso-negativos dos exames citopatológicos, como método de controle interno da qualidade, do que a revisão aleatória de 10% e revisão com base em critérios clínicos. Observou-se, também que a revisão rápida apresentou melhor desempenho quando a amostra foi classificada como satisfatória para análise, porém não varia com a idade da mulher. / Este estudo comparou a eficiência da revisão rápida de 100%, revisão aleatória de 10% e revisão de esfregaços selecionados com base em critérios clínicos como método de controle interno da qualidade dos esfregaços cervicais negativos no escrutínio de rotina e verificou se os resultados falso-negativos identificados pela revisão rápida variam com a adequabilidade da amostra e com a idade da mulher. Teve como base à população feminina usuária do Sistema Único de Saúde de Goiânia-GO, que se submeteu ao exame citopatológico realizado no Laboratório de Análises Clínicas Rômulo Rocha da Faculdade de Farmácia da Universidade Federal de Goiás. Os 5.530 esfregaços classificados como negativos no escrutínio de rotina foram submetidos ao método de revisão rápida de 100%, em seguida, foram selecionados os esfregaços com base em critérios clínicos e 10% do total de esfregaços negativos e submetidos às respectivas revisães. Quatro citologistas se alternaram durante as revisões. Após as revisões os resultados concordantes foram considerados como diagnóstico final, enquanto que, os divergentes foram analisados por um terceiro citologista, que em reunião de consenso definiu o diagnóstico final. Todas as etapas foram às cegas e os resultados classificados de acordo com o Sistema de Bethesda. Para análise estatística as variáveis foram estudadas de maneira descritiva e aplicado o Teste do Qui-quadrado. A revisão rápida de 100% identificou 141 esfregaços suspeitos, desses 84 (59,6%) foram considerados positivos pelo diagnóstico final, dos quais 36 (25,5%) foram classificados como células escamosas atípicas de significado indeterminado (ASC-US), cinco (3,5%) como células escamosas atípicas, não podendo excluir lesão de alto grau (ASC-H), 34 (24,1%) como lesão intra-epitelial escamosa de baixo grau (LSIL), seis (4,3%) como lesão intra-epitelial de alto grau (HSIL) e três (2,1%) células glandulares atípicas (AGC). Dos 84 esfregaços suspeitos e confirmados pelo diagnóstico final, 62 (73,8%) foram classificados como satisfatório e 22 (26,2%) satisfatórios, porém com alguma limitação, mas não observou diferença significativa com a idade da mulher. O método de revisão rápida identificou 19 esfregaços como insatisfatórios, desses 16 (84,2%) foram confirmados pelo diagnóstico final. Dos 1.279 esfregaços revisados com base em critérios clínicos, 24 foram considerados alterados. Desses 19 (79,2%) foram confirmados como positivos pelo diagnóstico final, sendo nove (47,3%) classificados como ASC-US, um (5,3%) ASC-H, oito (42,1%) LSIL e um (5,3%) HSIL. Dos 560 esfregaços analisados pela revisão de 10%, 13 foram considerados alterados, desses seis (46,2%) foram confirmados pelo diagnóstico final, sendo quatro (30,8%) classificados como ASC-US, um (7,7%) LSIL e um (7,7%) HSIL. Portanto, a revisão rápida de 100% é uma alternativa mais eficiente na detecção de resultados falso-negativos dos exames citopatológicos, como método de controle interno da qualidade, do que a revisão aleatória de 10% e revisão com base em critérios clínicos. Observou-se, também que a revisão rápida apresentou melhor desempenho quando a amostra foi classificada como satisfatória para análise, porém não varia com a idade da mulher.