Jauhatuspiirin ilmakierron optimointi

Luokittuminen erilaisine mekanismeineen aiheuttaa yleisesti ongelmia, kun on kysymyksessä kiintoaineen väliaikainenkin varastointi siilossa. Sitä voidaan vähentää kiintoaineiden, prosessin ja laitesuunnittelun muutoksilla. Tässä työssä tutkittiin mahdollisuuksia vähentää ilmeniitin luokittumista sen jauhatuspiirin ilmakiertoa optimoimalla. Suljetun kuivajauhatuspiirin keskeisimmäksi laitteeksi voitaisiin ajatella siinä oleva luokitin, joka voi olla esim. sykloni. Tässä piirissä tapahtuva kiintoaineen liikkuminen voidaan saada aikaiseksi esim. pneumaattisella kuljetuksella. Ilmeniitin jauhatus tapahtuu suljetussa kuivajauhatuspiirissä, jonka ajavana voimana on siinä oleva ilmakierto. Piirin oleellisia laitteita ovat kuulamylly, luokitin, erotussykloni ja pölykaappi sekä kiertoilma- ja poistoilmapuhaltimet. Ilmakierron optimointia varten suoritettiin kahden vastaavan jauhatuspiirin ainetasemääritykset. Lisäksi määritettiin yhden isomman piirin perustila. Jauhatuspiirien ainetasemäärityksissä määritettiin niiden massa- ja ilmavirrat sekä kiertokuorma ja luokittimen erotusterävyys, kuten myös ilmeniitin hiukkaskokojakaumat. Perustilamittauksissa määritettiin ainoastaan piirin ilmavirrat ja ilmeniitin hiukkaskokojakaumat. Optimointimittauksissa pienennettiin pikkumyllypiirin ilmamäärät vastaamaan kutakuinkin vastaavan toisen piirin määriä. Tällä yritettiin selvittää näiden toisiaan vastaavien piirien ilmamäärien ja varsinkin kiertokuormien eroavuutta. Tämä ilmamäärien pienentäminen ei tuottanut mainittavampaa muutosta piirin ainetaseisiin, joten voitaneen todeta, että piirin ilmamääriä pienentämällä saadaan aikaiseksi säästöjä, lähinnä kiertoilmapuhaltimen tehon alennuksen kautta.

Postnatal ontogenesis of the tibia. Implications for age and sex estimation

The growth of five variables of the tibia (diaphyseal length, diaphyseal length plus distal epiphysis, condylo-malleolar length, sagittal diameter of the proximal epiphysis, maximum breadth of the distal epiphysis) were analysed using polynomial regression in order to evaluate their significance and capacity for age and sex determination during and after growth. Data were collected from 181 (90♂ and 91♀) individuals ranging from birth to 25 years of age and belonging to three documented collections from Western Europe. Results indicate that all five variables exhibit linear behaviour during growth, which can be expressed by a first-degree polynomial function. Sexual significant differences were observed from age 15 onward in the two epiphysis measurements and condylo-malleolar length, suggesting that these three variables could be useful for sex determination in individuals older than 15 years. Strong correlation coefficients were identified between the five tibial variables and age. These results indicate that any of the studied tibial measurements is likely to serve as a useful source for estimating sub-adult age in both archaeological and forensic samples.

Reducción del vector de características de reconocimiento facial

In this present work, we are proposing a characteristics reduction system for a facial biometric identification system, using transformed domains such as discrete cosine transformed (DCT) and discrete wavelets transformed (DWT) as parameterization; and Support Vector Machines (SVM) and Neural Network (NN) as classifiers. The size reduction has been done with Principal Component Analysis (PCA) and with Independent Component Analysis (ICA). This system presents a similar success results for both DWT-SVM system and DWT-PCA-SVM system, about 98%. The computational load is improved on training mode due to the decreasing of input’s size and less complexity of the classifier.

Feature Selection, Ranking of Each Feature and Classification for the Diagnosis of Community Acquired Legionella Pneumonia

Diagnosis of community acquired legionella pneumonia (CALP) is currently performed by means of laboratory techniques which may delay diagnosis several hours. To determine whether ANN can categorize CALP and non-legionella community-acquired pneumonia (NLCAP) and be standard for use by clinicians, we prospectively studied 203 patients with community-acquired pneumonia (CAP) diagnosed by laboratory tests. Twenty one clinical and analytical variables were recorded to train a neural net with two classes (LCAP or NLCAP class). In this paper we deal with the problem of diagnosis, feature selection, and ranking of the features as a function of their classification importance, and the design of a classifier the criteria of maximizing the ROC (Receiving operating characteristics) area, which gives a good trade-off between true positives and false negatives. In order to guarantee the validity of the statistics; the train-validation-test databases were rotated by the jackknife technique, and a multistarting procedure was done in order to make the system insensitive to local maxima.

The Bonenblust-Hille inequality for homogeneous polynomials is hypercontractive

The Bohnenblust-Hille inequality says that the $\ell^{\frac{2m}{m+1}}$ -norm of the coefficients of an $m$-homogeneous polynomial $P$ on $\Bbb{C}^n$ is bounded by $\| P \|_\infty$ times a constant independent of $n$, where $\|\cdot \|_\infty$ denotes the supremum norm on the polydisc $\mathbb{D}^n$. The main result of this paper is that this inequality is hypercontractive, i.e., the constant can be taken to be $C^m$ for some $C>1$. Combining this improved version of the Bohnenblust-Hille inequality with other results, we obtain the following: The Bohr radius for the polydisc $\mathbb{D}^n$ behaves asymptotically as $\sqrt{(\log n)/n}$ modulo a factor bounded away from 0 and infinity, and the Sidon constant for the set of frequencies $\bigl\{ \log n: n \text{a positive integer} \le N\bigr\}$ is $\sqrt{N}\exp\{(-1/\sqrt{2}+o(1))\sqrt{\log N\log\log N}\}$.

Publication trends of shared decision making in 15 high impact medical journals: a full-text review with bibliometric analysis.

BACKGROUND: Shared Decision Making (SDM) is increasingly advocated as a model for medical decision making. However, there is still low use of SDM in clinical practice. High impact factor journals might represent an efficient way for its dissemination. We aimed to identify and characterize publication trends of SDM in 15 high impact medical journals. METHODS: We selected the 15 general and internal medicine journals with the highest impact factor publishing original articles, letters and editorials. We retrieved publications from 1996 to 2011 through the full-text search function on each journal website and abstracted bibliometric data. We included publications of any type containing the phrase "shared decision making" or five other variants in their abstract or full text. These were referred to as SDM publications. A polynomial Poisson regression model with logarithmic link function was used to assess the evolution across the period of the number of SDM publications according to publication characteristics. RESULTS: We identified 1285 SDM publications out of 229,179 publications in 15 journals from 1996 to 2011. The absolute number of SDM publications by journal ranged from 2 to 273 over 16 years. SDM publications increased both in absolute and relative numbers per year, from 46 (0.32% relative to all publications from the 15 journals) in 1996 to 165 (1.17%) in 2011. This growth was exponential (P < 0.01). We found fewer research publications (465, 36.2% of all SDM publications) than non-research publications, which included non-systematic reviews, letters, and editorials. The increase of research publications across time was linear. Full-text search retrieved ten times more SDM publications than a similar PubMed search (1285 vs. 119 respectively). CONCLUSION: This review in full-text showed that SDM publications increased exponentially in major medical journals from 1996 to 2011. This growth might reflect an increased dissemination of the SDM concept to the medical community.

Evaluation of Teacher Education Programs by Students and Graduates

The purpose of this research was to do a repeated cross-sectional research on class teachers who study in the 4th year and also graduated at the Faculty of Education, University of Turku between the years of 2000 through 2004. Specifically, seven research questions were addressed to target the main purpose of the study: How do class teacher education masters’ degree senior students and graduates rate “importance; effectiveness; and quality” of training they have received at the Faculty of Education? Are there significant differences between overall ratings of importance; effectiveness and quality of training by year of graduation, sex, and age (for graduates) and sex and age (for senior students)? Is there significant relationship between respondents’ overall ratings of importance; effectiveness and their overall ratings of the quality of training and preparation they have received? Are there significant differences between graduates and senior students about importance, effectiveness, and quality of teacher education programs? And what do teachers’ [Graduates] believe about how increasing work experience has changed their opinions of their preservice training? Moreover the following concepts related to the instructional activities were studied: critical thinking skills, communication skills, attention to ethics, curriculum and instruction (planning), role of teacher and teaching knowledge, assessment skills, attention to continuous professional development, subject matters knowledge, knowledge of learning environment, and using educational technology. Researcher also tried to find influence of some moderator variables e.g. year of graduation, sex, and age on the dependent and independent variables. This study consisted of two questionnaires (a structured likert-scale and an open ended questionnaire). The population in study 1 was all senior students and 2000-2004 class teacher education masters’ degree from the departments of Teacher Education Faculty of Education at University of Turku. Of the 1020 students and graduates the researcher was able to find current addresses of 675 of the subjects and of the 675 graduates contacted, 439 or 66.2 percent responded to the survey. The population in study 2 was all class teachers who graduated from Turku University and now work in the few basic schools (59 Schools) in South- West Finland. 257 teachers answered to the open ended web-based questions. SPSS was used to produce standard deviations; Analysis of Variance; Pearson Product Moment Correlation (r); T-test; ANOVA, Bonferroni post-hoc test; and Polynomial Contrast tests meant to analyze linear trend. An alpha level of .05 was used to determine statistical significance. The results of the study showed that: A majority of the respondents (graduates and students) rated the overall importance, effectiveness and quality of the teacher education programs as important, effective and good. Generally speaking there were only a few significant differences between the cohorts and groups related to the background variables (gender, age). The different cohorts were rating the quality of the programs very similarly but some differences between the cohorts were found in the importance and effectiveness ratings. Graduates of 2001 and 2002 rated the importance of the program significantly higher than 2000 graduates. The effectiveness of the programs was rated significantly higher by 2001 and 2003 graduates than other groups. In spite of these individual differences between cohorts there were no linear trends among the year cohorts in any measure. In respondents’ ratings of the effectiveness of teacher education programs there was significant difference between males and females; females rated it higher than males. There were no significant differences between males’ and females’ ratings of the importance and quality of programs. In the ratings there was only one difference between age groups. Older graduates (35 years or older) rated the importance of the teacher training significantly higher that 25-35 years old graduates. In graduates’ ratings there were positive but relatively low correlations between all variables related to importance, effectiveness and quality of Teacher Education Programs. Generally speaking students’ ratings about importance, effectiveness and quality of teacher education program were very positive. There was only one significant difference related to the background variables. Females rated higher the effectiveness of the program. The comparison of students’ and graduates’ perception about importance, effectiveness, and quality of teacher education programs showed that there were no significant differences between graduates and students in the overall ratings. However there were differences in some individual variables. Students rated higher in importance of “Continuous Professional Development”, effectiveness of “Critical Thinking Skills” and “Using Educational Technology” and quality of “Advice received from the advisor”. Graduates rated higher in importance of “Knowledge of Learning Environment” and effectiveness of “Continuous Professional Development”. According to the qualitative data of study 2 some graduates expressed that their perceptions have not changed about the importance, effectiveness, and quality of training that they received during their study time. They pointed out that teacher education programs have provided them the basic theoretical/formal knowledge and some training of practical routines. However, a majority of the teachers seems to have somewhat critical opinions about the teacher education. These teachers were not satisfied with teacher education programs because they argued that the programs failed to meet their practical demands in different everyday situations of the classroom e.g. in coping with students’ learning difficulties, multiprofessional communication with parents and other professional groups (psychologists and social workers), and classroom management problems. Participants also emphasized more practice oriented knowledge of subject matter, evaluation methods and teachers’ rights and responsibilities. Therefore, they (54.1% of participants) suggested that teacher education departments should provide more practice-based courses and programs as well as closer collaboration between regular schools and teacher education departments in order to fill gap between theory and practice.

MYC-IG rearrangements are negative predictors of survival in DLBCL patients treated with immunochemotherapy: a GELA/LYSA study.

Diffuse large B-cell lymphoma (DLBCL) with MYC rearrangement (MYC-R) carries an unfavorable outcome. We explored the prognostic value of the MYC translocation partner gene in a series of MYC-R de novo DLBCL patients enrolled in first-line prospective clinical trials (Groupe d'Etudes des Lymphomes de l'Adulte/Lymphoma Study Association) and treated with rituximab-anthracycline-based chemotherapy. A total of 774 DLBCL cases characterized for cell of origin by the Hans classifier were analyzed using fluorescence in situ hybridization with BCL2, BCL6, MYC, immunoglobulin (IG)K, and IGL break-apart and IGH/MYC, IGK/MYC, and IGL/MYC fusion probes. MYC-R was observed in 51/574 (8.9%) evaluable DLBCL cases. MYC-R cases were predominantly of the germinal center B-cell-like subtype 37/51 (74%) with no distinctive morphologic and phenotypic features. Nineteen cases were MYC single-hit and 32 cases were MYC double-hit (MYC plus BCL2 and/or BCL6) DLBCL. MYC translocation partner was an IG gene in 24 cases (MYC-IG) and a non-IG gene (MYC-non-IG) in 26 of 50 evaluable cases. Noteworthy, MYC-IG patients had shorter overall survival (OS) (P = .0002) compared with MYC-negative patients, whereas no survival difference was observed between MYC-non-IG and MYC-negative patients. In multivariate analyses, MYC-IG predicted poor progression-free survival (P = .0051) and OS (P = .0006) independently from the International Prognostic Index and the Hans classifier. In conclusion, we show in this prospective randomized trial that the adverse prognostic impact of MYC-R is correlated to the MYC-IG translocation partner gene in DLBCL patients treated with immunochemotherapy. These results may have an important impact on the clinical management of DLBCL patients with MYC-R who should be routinely characterized according to MYC partner gene. These trials are individually registered at www.clinicaltrials.gov as #NCT00144807, #NCT01087424, #NCT00169143, #NCT00144755, #NCT00140660, #NCT00140595, and #NCT00135499.

Stability of nanoparticle agglomerates under mechanical stress and its effects on their release into the air

L'exposition professionnelle aux nanomatériaux manufacturés dans l'air présente des risques potentiels pour la santé des travailleurs dans les secteurs de la nanotechnologie. Il est important de comprendre les scénarios de libération des aérosols de nanoparticules dans les processus et les activités associées à l'exposition humaine. Les mécanismes de libération, y compris les taux de libération et les propriétés physico-chimiques des nanoparticules, déterminent leurs comportements de transport ainsi que les effets biologiques néfastes. La distribution de taille des particules d'aérosols est l'un des paramètres les plus importants dans ces processus. La stabilité mécanique d'agglomérats de nanoparticules affecte leurs distributions de tailles. Les potentiels de désagglomération de ces agglomérats déterminent les possibilités de leur déformation sous énergies externes. Cela rend les changements possibles dans leur distribution de taille et de la concentration en nombre qui vont finalement modifier leurs risques d'exposition. Les conditions environnementales, telles que l'humidité relative, peuvent influencer les processus de désagglomération par l'adhérence de condensation capillaire de l'humidité. L'objectif général de cette thèse était d'évaluer les scénarios de libération des nanomatériaux manufacturés des processus et activités sur le lieu de travail. Les sous-objectifs étaient les suivants: 1. Etudier les potentiels de désagglomération des nanoparticules dans des conditions environnementales variées. 2. Etudier la libération des nano-objets à partir de nanocomposites polymères; 3. Evaluer la libération de nanoparticules sur le lieu de travail dans des situations concrètes. Nous avons comparé différents systèmes de laboratoire qui présentaient différents niveau d'énergie dans l'aérosolisation des poudres. Des nanopoudres de TiO2 avec des hydrophilicités de surface distinctes ont été testées. Un spectromètre à mobilité électrique (SMPS), un spectromètre à mobilité aérodynamique (APS) et un spectromètre optique (OPC) ont été utilisés pour mesurer la concentration de particules et la distribution de taille des particules. La microscopie électronique à transmission (TEM) a été utilisée pour l'analyse morphologique d'échantillons de particules dans l'air. Les propriétés des aérosols (distribution de taille et concentration en nombre) étaient différentes suivant la méthode employée. Les vitesses des flux d'air d'aérosolisation ont été utilisées pour estimer le niveau d'énergie dans ces systèmes, et il a été montré que les tailles modales des particules étaient inversement proportionnelles à la vitesse appliquée. En général, les particules hydrophiles ont des diamètres plus grands et des nombres inférieurs à ceux des particules hydrophobes. Toutefois, cela dépend aussi des méthodes utilisées. La vitesse de l'air peut donc être un paramètre efficace pour le classement de l'énergie des procédés pour des systèmes d'aérosolisation similaires. Nous avons développé un système laboratoire pour tester les potentiels de désagglomération des nanoparticules dans l'air en utilisant des orifices critiques et un humidificateur. Sa performance a été comparée à un système similaire dans un institut partenaire. Une variété de nanopoudres différentes a été testée. Le niveau d'énergie appliquée et l'humidité ont été modifiés. Le SMPS et l'OPC ont été utilisés pour mesurer la concentration de particules et la distribution de la taille. Un TEM a été utilisé pour l'analyse morphologique d'échantillons de particules dans l'air. Le diamètre moyen des particules a diminué et la concentration en nombre s'est accrue lorsque des énergies externes ont été appliquées. Le nombre de particules inférieures à 100 nm a été augmenté, et celui au-dessus de 350 nm réduits. Les conditions humides ont faits exactement le contraire, en particulier pour les petites particules. En outre, ils ont réduits les effets de la différence de pression due à l'orifice. Les résultats suggèrent que la désagglomération d'agglomérats de nanoparticules dans l'air est possible dans la gamme d'énergie appliquée. Cependant, l'atmosphère humide peut favoriser leur agglomération et améliorer leurs stabilités en réduisant la libération de nanoparticules dans l'environnement. Nous proposons d'utiliser notre système pour le test de routine des potentiels de désagglomération des nanomatériaux manufacturés et de les classer. Un tel classement faciliterait la priorisation de l'exposition et du risque encouru en fonction du niveau d'ENM. Un système de perçage automatique et un système de sciage manuel ont été développés pour étudier la libération de nanoparticules à partir de différents types de nanocomposites. La vitesse de perçage et taille de la mèche ont été modifiées dans les expériences. La distribution de taille des particules et leur concentration en nombre ont été mesurées par un SMPS et un miniature diffusion size classifier (DISCmini). Les distributions de nanoparticules dans les composites et les particules libérées ont été analysés par un TEM et un microscope électronique à balayage (SEM). Les tests de perçage ont libérés un plus grand nombre de particules que le sciage. Des vitesses de perçage plus rapide et les mèches plus grandes ont augmentés la génération de particules. Les charges de nanoparticules manufacturées dans les composites ne modifient pas leurs comportements de libération dans les expériences de perçage. Toutefois, le sciage différencie les niveaux de libération entre les composites et les échantillons blancs. De plus, les vapeurs de polymères ont été générées par la chaleur de sciage. La plupart des particules libérées sont des polymères contenant des nanoparticules ou sur leurs surface. Les résultats ont souligné l'importance du type de processus et paramètres pour déterminer la libération de nanoparticules de composites. Les émissions secondaires telles que les fumées polymères appellent à la nécessité d'évaluations de l'exposition et de risque pour de tels scénarios. Une revue systématique de la littérature sur le sujet de libérations de nanoparticules dans l'air dans les secteurs industriels et laboratoires de recherche a été effectuée. Des stratégies de recherche des informations pertinentes et de stockage ont été développées. Les mécanismes de libération, tels que la taille de particules d'aérosol et de leur concentration en nombre, ont été comparés pour différentes activités. La disponibilité de l'information contextuelle qui est pertinente pour l'estimation de l'exposition humaine a été évaluée. Il a été constaté que les données relatives à l'exposition ne sont pas toujours disponibles dans la littérature actuelle. Les propriétés des aérosols libérés semblent dépendre de la nature des activités. Des procédés à haute énergie ont tendance à générer des plus hauts niveaux de concentrations de particules dans les gammes de plus petite taille. Les résultats peuvent être utiles pour déterminer la priorité des procédés industriels pour l'évaluation les risques associés dans une approche à plusieurs niveaux. Pour l'évaluation de l'exposition, la disponibilité de l'information peut être améliorée par le développement d'une meilleure méthode de communication des données.

Pubic growth study: applicability in sexual and age diagnostic.

The growth of five variables of the ischiopubic area was analyzed from bone material from birth to old age. The main purpose was to evaluate its significance and capacity for age and sex determination during and after growth. The material used consisted of 327 specimens from four documented Western European collections. Growth curves were calculated by polynomial regression for two classical variables of the ischiopubic area (pubis length and ischiopubic index) and three new variables of the pubic acetabular area (horizontal and vertical diameter of the pubic acetabular area and the pubic acetabular index). None of the curves showed lineal growth, with the exception of the ischiopubic index and the masculine vertical diameter of the pubis acetabular area. Pubis length has the most complicated growth, expressed by a five-degree polynomial. All the variables are useful for adult sex determination, except the pubic acetabular index. The ischopubic index, vertical diameter of the pubic acetabular area and the pubic acetabular index seem to be good variables for sub-adult sex determination. For age estimation the best variables, in both archaeological and forensic remains, are the absolute measurements (pubic length, vertical and horizontal diameter of the pubis). However, pubis length is the best variable for age estimation because it can be applied until 25 years of age.

DNA Topoisomerase I Gene Copy Number and mRNA Expression Assessed as Predictive Biomarkers for Adjuvant Irinotecan in Stage II/III Colon Cancer.

PURPOSE: Prospective-retrospective assessment of theTOP1gene copy number andTOP1mRNA expression as predictive biomarkers for adjuvant irinotecan in stage II/III colon cancer. EXPERIMENTAL DESIGN: Formalin-fixed, paraffin-embedded tissue microarrays were obtained from an adjuvant colon cancer trial (PETACC3) where patients were randomized to 5-fluorouracil/folinic acid with or without additional irinotecan.TOP1copy number status was analyzed by fluorescencein situhybridization (FISH) using aTOP1/CEN20 dual-probe combination.TOP1mRNA data were available from previous analyses. RESULTS: TOP1FISH and follow-up data were obtained from 534 patients.TOP1gain was identified in 27% using a single-probe enumeration strategy (≥4TOP1signals per cell) and in 31% when defined by aTOP1/CEN20 ratio ≥ 1.5. The effect of additional irinotecan was not dependent onTOP1FISH status.TOP1mRNA data were available from 580 patients with stage III disease. Benefit of irinotecan was restricted to patients characterized byTOP1mRNA expression ≥ third quartile (RFS: HRadjusted, 0.59;P= 0.09; OS: HRadjusted, 0.44;P= 0.03). The treatment byTOP1mRNA interaction was not statistically significant, but in exploratory multivariable fractional polynomial interaction analysis, increasingTOP1mRNA values appeared to be associated with increasing benefit of irinotecan. CONCLUSIONS: In contrast to theTOP1copy number, a trend was demonstrated for a predictive property ofTOP1mRNA expression. On the basis ofTOP1mRNA, it might be possible to identify a subgroup of patients where an irinotecan doublet is a clinically relevant option in the adjuvant setting of colon cancer.Clin Cancer Res; 22(7); 1621-31. ©2015 AACR.

Shared feature extraction for nearest neighbor face recognition

In this paper, we propose a new supervised linearfeature extraction technique for multiclass classification problemsthat is specially suited to the nearest neighbor classifier (NN).The problem of finding the optimal linear projection matrix isdefined as a classification problem and the Adaboost algorithmis used to compute it in an iterative way. This strategy allowsthe introduction of a multitask learning (MTL) criterion in themethod and results in a solution that makes no assumptions aboutthe data distribution and that is specially appropriated to solvethe small sample size problem. The performance of the methodis illustrated by an application to the face recognition problem.The experiments show that the representation obtained followingthe multitask approach improves the classic feature extractionalgorithms when using the NN classifier, especially when we havea few examples from each class

Classification of textures seen from different distances and under varying illumination direction

Changes in the angle of illumination incident upon a 3D surface texture can significantly alter its appearance, implying variations in the image texture. These texture variations produce displacements of class members in the feature space, increasing the failure rates of texture classifiers. To avoid this problem, a model-based texture recognition system which classifies textures seen from different distances and under different illumination directions is presented in this paper. The system works on the basis of a surface model obtained by means of 4-source colour photometric stereo, used to generate 2D image textures under different illumination directions. The recognition system combines coocurrence matrices for feature extraction with a Nearest Neighbour classifier. Moreover, the recognition allows one to guess the approximate direction of the illumination used to capture the test image

Using appearance and context for outdoor scene object classification

We propose a probabilistic object classifier for outdoor scene analysis as a first step in solving the problem of scene context generation. The method begins with a top-down control, which uses the previously learned models (appearance and absolute location) to obtain an initial pixel-level classification. This information provides us the core of objects, which is used to acquire a more accurate object model. Therefore, their growing by specific active regions allows us to obtain an accurate recognition of known regions. Next, a stage of general segmentation provides the segmentation of unknown regions by a bottom-strategy. Finally, the last stage tries to perform a region fusion of known and unknown segmented objects. The result is both a segmentation of the image and a recognition of each segment as a given object class or as an unknown segmented object. Furthermore, experimental results are shown and evaluated to prove the validity of our proposal

Modeling and Classifying Breast Tissue Density in Mammograms

We present a new approach to model and classify breast parenchymal tissue. Given a mammogram, first, we will discover the distribution of the different tissue densities in an unsupervised manner, and second, we will use this tissue distribution to perform the classification. We achieve this using a classifier based on local descriptors and probabilistic Latent Semantic Analysis (pLSA), a generative model from the statistical text literature. We studied the influence of different descriptors like texture and SIFT features at the classification stage showing that textons outperform SIFT in all cases. Moreover we demonstrate that pLSA automatically extracts meaningful latent aspects generating a compact tissue representation based on their densities, useful for discriminating on mammogram classification. We show the results of tissue classification over the MIAS and DDSM datasets. We compare our method with approaches that classified these same datasets showing a better performance of our proposal