CAD e mamografia: por que usar esta ferramenta?

A mamografia representa o melhor método de detecção precoce do câncer de mama, porém cerca de 10% a 30% das lesões mamárias são perdidas no rastreamento, devido a limitações próprias dos observadores humanos. A detecção auxiliada por computador (computer-aided detection - CAD) é uma tecnologia relativamente nova que tem sido implementada em alguns serviços de mamografia, com o intuito de prover uma dupla leitura. Estudos clínicos têm demonstrado que o CAD aumenta a sensibilidade de detecção do câncer da mama, por radiologistas, em até 21%. Um sistema CAD é útil em situações em que exista alta variabilidade interobservador, falta de observadores treinados, ou na impossibilidade de se realizar a dupla leitura com dois ou mais radiologistas. O objetivo desta revisão está baseado na necessidade de atualizar a comunidade médica acerca desta ferramenta, como um método auxiliar, quantitativo, não operador-dependente, e que visa a melhorar a qualidade do diagnóstico do câncer de mama.

O6-Methylguanine-DNA methyltransferase protein expression by immunohistochemistry in brain and non-brain systemic tumours: systematic review and meta-analysis of correlation with methylation-specific polymerase chain reaction

Background: The DNA repair protein O6-Methylguanine-DNA methyltransferase (MGMT) confers resistance to alkylating agents. Several methods have been applied to its analysis, with methylation-specific polymerase chain reaction (MSP) the most commonly used for promoter methylation study, while immunohistochemistry (IHC) has become the most frequently used for the detection of MGMT protein expression. Agreement on the best and most reliable technique for evaluating MGMT status remains unsettled. The aim of this study was to perform a systematic review and meta-analysis of the correlation between IHC and MSP. Methods A computer-aided search of MEDLINE (1950-October 2009), EBSCO (1966-October 2009) and EMBASE (1974-October 2009) was performed for relevant publications. Studies meeting inclusion criteria were those comparing MGMT protein expression by IHC with MGMT promoter methylation by MSP in the same cohort of patients. Methodological quality was assessed by using the QUADAS and STARD instruments. Previously published guidelines were followed for meta-analysis performance. Results Of 254 studies identified as eligible for full-text review, 52 (20.5%) met the inclusion criteria. The review showed that results of MGMT protein expression by IHC are not in close agreement with those obtained with MSP. Moreover, type of tumour (primary brain tumour vs others) was an independent covariate of accuracy estimates in the meta-regression analysis beyond the cut-off value. Conclusions Protein expression assessed by IHC alone fails to reflect the promoter methylation status of MGMT. Thus, in attempts at clinical diagnosis the two methods seem to select different groups of patients and should not be used interchangeably.

Real-Time Simulation of Mobile and Industrial Machines Using the Multibody Simulation Approach

This thesis introduces a real-time simulation environment based on the multibody simulation approach. The environment consists of components that are used in conventional product development, including computer aided drawing, visualization, dynamic simulation and finite element software architecture, data transfer and haptics. These components are combined to perform as a coupled system on one platform. The environment is used to simulate mobile and industrial machines at different stages of a product life time. Consequently, the demands of the simulated scenarios vary. In this thesis, a real-time simulation environment based on the multibody approach is used to study a reel mechanism of a paper machine and a gantry crane. These case systems are used to demonstrate the usability of the real-time simulation environment for fault detection purposes and in the context of a training simulator. In order to describe the dynamical performance of a mobile or industrial machine, the nonlinear equations of motion must be defined. In this thesis, the dynamical behaviour of machines is modelled using the multibody simulation approach. A multibody system may consist of rigid and flexible bodies which are joined using kinematic joint constraints while force components are used to describe the actuators. The strength of multibody dynamics relies upon its ability to describe nonlinearities arising from wearing of the components, friction, large rotations or contact forces in a systematic manner. For this reason, the interfaces between subsystems such as mechanics, hydraulics and control systems of the mechatronic machine can be defined and analyzed in a straightforward manner.

Genetic diversity of begomovirus infecting tomato and associated weeds in Southeastern Brazil

The genetic diversity of begomovirus isolates from tomato (Lycopersicon esculentum) fields in the Southeastern region of Brazil was analyzed by direct sequencing of PCR fragments amplified by using universal oligonucleotides for the begomovirus DNA-A, and subsequent computer-aided phylogenetic analysis. Samples of tomato plants and associated weeds showing typical symptoms of virus infection were collected at seven locations in the states of Minas Gerais, Espírito Santo and Rio de Janeiro. A total of 137 out of 369 samples were infected with a begomovirus based on PCR analysis. Phylogenetic analysis indicated a high degree of genetic diversity among begomoviruses infecting tomatoes in the sampled area. One species (Tomato chlorotic mottle virus, TCMV) occurs predominantly in Minas Gerais, whereas in Rio de Janeiro and Espírito Santo a distinct species, not yet fully characterized, predominates. Phylogenetic analysis further indicates the presence of an additional four possible new species. This high degree of genetic diversity suggests a recent transfer of indigenous begomovirus from wild hosts into tomatoes. The close phylogenetic relationship verified between begomovirus infecting tomato and associated weeds favors this hypothesis.

Comparison Structure Analysis

This study presents an automatic, computer-aided analytical method called Comparison Structure Analysis (CSA), which can be applied to different dimensions of music. The aim of CSA is first and foremost practical: to produce dynamic and understandable representations of musical properties by evaluating the prevalence of a chosen musical data structure through a musical piece. Such a comparison structure may refer to a mathematical vector, a set, a matrix or another type of data structure and even a combination of data structures. CSA depends on an abstract systematic segmentation that allows for a statistical or mathematical survey of the data. To choose a comparison structure is to tune the apparatus to be sensitive to an exclusive set of musical properties. CSA settles somewhere between traditional music analysis and computer aided music information retrieval (MIR). Theoretically defined musical entities, such as pitch-class sets, set-classes and particular rhythm patterns are detected in compositions using pattern extraction and pattern comparison algorithms that are typical within the field of MIR. In principle, the idea of comparison structure analysis can be applied to any time-series type data and, in the music analytical context, to polyphonic as well as homophonic music. Tonal trends, set-class similarities, invertible counterpoints, voice-leading similarities, short-term modulations, rhythmic similarities and multiparametric changes in musical texture were studied. Since CSA allows for a highly accurate classification of compositions, its methods may be applicable to symbolic music information retrieval as well. The strength of CSA relies especially on the possibility to make comparisons between the observations concerning different musical parameters and to combine it with statistical and perhaps other music analytical methods. The results of CSA are dependent on the competence of the similarity measure. New similarity measures for tonal stability, rhythmic and set-class similarity measurements were proposed. The most advanced results were attained by employing the automated function generation – comparable with the so-called genetic programming – to search for an optimal model for set-class similarity measurements. However, the results of CSA seem to agree strongly, independent of the type of similarity function employed in the analysis.

A reprodutibilidade da medida do volume endometrial por meio do VOCAL: a importância do passo de rotação

OBJETIVOS: avaliar a reprodutibilidade intra-observador e interobservador da medida do volume endometrial utilizando ultra-sonografia tridimensional (3D) e o programa VOCAL® (Virtual Organ Computer-aided AnaLysis). MÉTODOS: um bloco ultra-sonográfico 3D do endométrio foi obtido de cinco pacientes inférteis voluntárias que apresentavam diferentes volumes endometriais. Para cada bloco 3D, o volume endometrial foi calculado utilizando o modo manual em quatro diferentes passos de rotação (30º, 15º, 9º e 6º) por dois diferentes observadores. Dez medidas foram obtidas para cada rotação e por cada observador. Testamos a diferença entre as médias com o one-way ANOVA e o pós-teste de Tukey e a reprodutibilidade com os coeficientes de correlação intraclasse. RESULTADOS: as medidas realizadas com passo de rotação de 30º foram associadas a médias significativamente menores em 3 das 5 pacientes. Não houve diferença entre as médias obtidas pelos passos de rotação de 15º, 9º ou 6º. Em nenhuma das avaliações foi notada diferença entre as médias obtidas pelos dois observadores. Os coeficientes de correlação intraclasse foram significativamente menores com o passo de rotação de 30º (todos abaixo de 0,984) que com os outros passos de rotação (todos acima de 0,996). CONCLUSÕES: o uso de passos de rotação de 15º ou menos apresentou medidas reprodutíveis do volume endometrial: não houve diferença significativa entre as médias obtidas pelos dois observadores, associado ao alto coeficiente de correlação intraclasse (>0,996). É recomendável o uso do passo de rotação de 15º, pois demora menos para ser obtido em comparação com 6º e 9º.

Volume do embrião estimado pela ultra-sonografia tridimensional entre a sétima e a décima semana de gestação

OBJETIVO: avaliar a evolução do volume do embrião (VE) entre a sétima e a décima semana de gestação por meio da ultra-sonografia tridimensional. MÉTODOS: realizou-se um estudo de corte transversal com 63 gestantes normais entre a sétima e a décima semana. Os exames ultra-sonográficos foram realizados por meio de um transdutor endocavitário volumétrico. Para o cálculo do VE, utilizou-se o método VOCAL (Virtual Organ Computer-aided Analysis) com ângulo de rotação de 12º, com delimitação de 15 planos seqüenciais. Para o VE foram calculadas médias, medianas, desvios padrão e valores máximo e mínimo em todas as idades gestacionais. Para se avaliar a correlação entre o VE e o comprimento cabeça-nádega (CCN) foi criado gráfico de dispersão, sendo o ajuste realizado pelo coeficiente de determinação (R²). Para se determinarem intervalos de referência do VE em função do CCN, utilizou-se a seguinte fórmula: percentil =VE+K versus dp, com K=1,96. RESULTADOS: o CCN variou de 9,0 a 39,7 mm, com média de 23,9 mm (±7,9 mm), enquanto o VE variou de 0,1 a 7,6 cm³, com média de 2,7 cm³ (±3,2 cm³). O VE foi altamente correlacionado com o CCN, sendo que o melhor ajuste foi obtido com regressão quadrática (VE=0,165 - 0,055 x CCN + 0,005 x CCN²; R²=0,853). O VE médio variou de 0,1 (-0,3 a 0,5 cm³) a 6,7 cm³ (3,8 a 9,7 cm³) no intervalo de 9 a 40 mm do CCN. Neste intervalo o VE aumentou 67 vezes, enquanto o CCN aumentou apenas 4,4 vezes. CONCLUSÕES: o VE é um parâmetro mais sensível que o CCN para avaliar o crescimento embrionário entre a sétima e a décima semana de gestação.

Business intelligence viewpoint on managerial decision-making : Ill-structured problems and decision support systems

Vaikka liiketoimintatiedon hallintaa sekä johdon päätöksentekoa on tutkittu laajasti, näiden kahden käsitteen yhteisvaikutuksesta on olemassa hyvin rajallinen määrä tutkimustietoa. Tulevaisuudessa aiheen tärkeys korostuu, sillä olemassa olevan datan määrä kasvaa jatkuvasti. Yritykset tarvitsevat jatkossa yhä enemmän kyvykkyyksiä sekä resursseja, jotta sekä strukturoitua että strukturoimatonta tietoa voidaan hyödyntää lähteestä riippumatta. Nykyiset Business Intelligence -ratkaisut mahdollistavat tehokkaan liiketoimintatiedon hallinnan osana johdon päätöksentekoa. Aiemman kirjallisuuden pohjalta, tutkimuksen empiirinen osuus tunnistaa liiketoimintatiedon hyödyntämiseen liittyviä tekijöitä, jotka joko tukevat tai rajoittavat johdon päätöksentekoprosessia. Tutkimuksen teoreettinen osuus johdattaa lukijan tutkimusaiheeseen kirjallisuuskatsauksen avulla. Keskeisimmät tutkimukseen liittyvät käsitteet, kuten Business Intelligence ja johdon päätöksenteko, esitetään relevantin kirjallisuuden avulla – tämän lisäksi myös dataan liittyvät käsitteet analysoidaan tarkasti. Tutkimuksen empiirinen osuus rakentuu tutkimusteorian pohjalta. Tutkimuksen empiirisessä osuudessa paneudutaan tutkimusteemoihin käytännön esimerkein: kolmen tapaustutkimuksen avulla tutkitaan sekä kuvataan toisistaan irrallisia tapauksia. Jokainen tapaus kuvataan sekä analysoidaan teoriaan perustuvien väitteiden avulla – nämä väitteet ovat perusedellytyksiä menestyksekkäälle liiketoimintatiedon hyödyntämiseen perustuvalle päätöksenteolle. Tapaustutkimusten avulla alkuperäistä tutkimusongelmaa voidaan analysoida tarkasti huomioiden jo olemassa oleva tutkimustieto. Analyysin tulosten avulla myös yksittäisiä rajoitteita sekä mahdollistavia tekijöitä voidaan analysoida. Tulokset osoittavat, että rajoitteilla on vahvasti negatiivinen vaikutus päätöksentekoprosessin onnistumiseen. Toisaalta yritysjohto on tietoinen liiketoimintatiedon hallintaan liittyvistä positiivisista seurauksista, vaikka kaikkia mahdollisuuksia ei olisikaan hyödynnetty. Tutkimuksen merkittävin tulos esittelee viitekehyksen, jonka puitteissa johdon päätöksentekoprosesseja voidaan arvioida sekä analysoida. Despite the fact that the literature on Business Intelligence and managerial decision-making is extensive, relatively little effort has been made to research the relationship between them. This particular field of study has become important since the amount of data in the world is growing every second. Companies require capabilities and resources in order to utilize structured data and unstructured data from internal and external data sources. However, the present Business Intelligence technologies enable managers to utilize data effectively in decision-making. Based on the prior literature, the empirical part of the thesis identifies the enablers and constraints in computer-aided managerial decision-making process. In this thesis, the theoretical part provides a preliminary understanding about the research area through a literature review. The key concepts such as Business Intelligence and managerial decision-making are explored by reviewing the relevant literature. Additionally, different data sources as well as data forms are analyzed in further detail. All key concepts are taken into account when the empirical part is carried out. The empirical part obtains an understanding of the real world situation when it comes to the themes that were covered in the theoretical part. Three selected case companies are analyzed through those statements, which are considered as critical prerequisites for successful computer-aided managerial decision-making. The case study analysis, which is a part of the empirical part, enables the researcher to examine the relationship between Business Intelligence and managerial decision-making. Based on the findings of the case study analysis, the researcher identifies the enablers and constraints through the case study interviews. The findings indicate that the constraints have a highly negative influence on the decision-making process. In addition, the managers are aware of the positive implications that Business Intelligence has for decision-making, but all possibilities are not yet utilized. As a main result of this study, a data-driven framework for managerial decision-making is introduced. This framework can be used when the managerial decision-making processes are evaluated and analyzed.

Background for Designing an Improved Digital Musical Keyboard Instrument

Electrical keyboard instruments and computer-aided music-making generally base on the piano keyboard that was developed for a tuning system no longer used. Alternative keyboard layout offers at least easier playing, faster adopting, new ways to play and better ergonomics. This thesis explores the development of keyboard instruments and tunings, and different keyboard layouts. This work is preliminary research for an electrical keyboard instrument to be implemented later on.

Training graduate students to be teachers

Pedagogic education of graduate students, when and where it exists, is restricted to theoretical courses or to the participation of the students as teachers' assistants. This model is essentially reproductive and offers few opportunities for any significant curriculum innovation. To open an opportunity for novelty we have introduced a new approach in "Biochemistry Teaching", a course included in the Biochemistry Graduate Program of the Biochemistry Department (Universidade Estadual de Campinas and Universidade de São Paulo). The content of the course consists of a) choosing the theme, b) selecting and organizing the topics, c) preparing written material, d) establishing the methodological strategies, e) planning the evaluation tools and, finally, f) as teachers, conducting the course as an optional summer course for undergraduate students. During the first semester the graduate students establish general and specific educational objectives, select and organize contents, decide on the instructional strategies and plan evaluation tools. The contents are explored using a wide range of strategies, which include computer-aided instruction, laboratory classes, small group teaching, a few lectures and round table discussions. The graduate students also organize printed class notes to be used by the undergraduate students. Finally, as a group, they teach the summer course. In the three versions already developed, the themes chosen were Biochemistry of Exercise (UNICAMP), Biochemistry of Nutrition (UNICAMP) and Molecular Biology of Plants (USP). In all cases the number of registrations greatly exceeded the number of places and a selection had to be made. The evaluation of the experience by both graduate and undergraduate students was very positive. Graduate students considered this experience to be unique and recommended it to their schoolmates; the undergraduate students benefited from a more flexible curriculum (more options) and gave very high scores to both the courses and the teachers.

Meesauunin kannatuksien suunnittelutyökalun kehitys

Meesauunia käytetään sellun tuotantoprosessissa kemiallisessa reaktiossa, jossa kalsiumkarbonaatti muutetaan kalsiumoksidiksi. Reaktio on osa sellutehtaan kemikaalikiertoa, missä sellun valmistamisessa käytettävät kemikaalit kierrätetään uusiokäyttöön. Tässä diplomityössä on kehitetty meesauunin kannatusten automatisoitu suunnittelutyökalu vanhan AutoLispillä kehitetyn ohjelman perusteella. Kehitetyn suunnittelutyökalun tärkeimmät osat ovat Excelissä tehdyt kuormituslaskenta ja mitoituslaskenta. Näillä ohjataan Autodeskin Inventoriin kehitettyä cad-mallia ja valmistuspiirustusta. Laskentaohjelmat ja cad-malli on kehitetty mahdollistamaan uusien uunikokojen helpon lisäämisen ohjelmaan. Diplomityössä on perehdytty kuormituslaskentaan standardien mukaisesti. Käytettyjä standardeja ovat esimerkiksi Eurokoodi ja Uniform Building Code. Epätavallisten tilanteiden kuormituslaskennassa käytettiin apuna FE-analyysilla kehitettyjä laskentamalleja. Näitä käytettiin uunin taipumisesta aiheutuvien voimien laskennassa ja lovetun kannatinpyörän aiheuttamien impulssivoimien suuruuden määrittämisessä. Lisäksi diplomityössä perehdyttiin suunnitteluautomaatin kehittämiseen Exceliä apuna käyttäen. Suunnittelutyökalun toimivuutta on verifioitu suunnittelemalla kehitetyllä suunnittelutyökalulla standardiuuneja ja verrattu tuloksia ja valmistuspiirustuksia vanhalla ohjelmalla tulostettuihin tuloksiin. Tulosten perusteella uutta suunnittelutyökalua voidaan käyttää meesauunin kannatuksien suunnitteluun.

Legitimacy Building in Public Disclosures on Strategic Energy Investments

Renewable energy investments play a key role in energy transition. While studies have suggested that social acceptance may form a barrier for renewable energy investments, the ways in which companies perceive and attempt to gain the acceptance have received little attention. This study aims to fill the gap by exploring how large electric utilities justify their strategic investments in their press releases and how do the justifications differ between renewable and non-renewable energy investments. The study bases on legitimacy theory and aims at contributing to the research on legitimation in institutional change. As its research method, the study employs an inductive mixed method content analysis. The study has two parts: a qualitative content analysis that explores and identifies the themes and legitimation strategies of the press releases and a quantitative computer-aided analysis that compares renewable and non-renewable energy investments. The sample of the study consists of 396 press releases representing the strategic energy investments of 34 electric utilities from the list of the world’s 250 largest and financially most successful energy companies. The data is collected from the period of 2010–2014. The study reveals that most important justifications for strategic energy investments are fit with the strategy and environmental and social benefits. Justifications address especially the expectations of market. Investments into non-renewable energy are justified more and they use more arguments addressing the proprieties and performance of power plants whereas renewable energy investments are legitimized by references to past actions and commonly accepted morals and norms. The findings support the notion that validity-addressing and propriety-addressing legitimation strategies are used differently in stable and unstable institutional settings.

Les végétaux comme éléments du paysage sonore : relations qualitatives entre conditions environnementales, morphologie et organisation dans la production d’ambiances sonores

Le paysage sonore est une forme de perception de notre environnement qui nous permet d’identifier les composantes sonores de notre quotidien. Ce projet de recherche porte sur une thématique particulière, les sons produits par les végétaux et leurs rôles dans les ambiances sonores paysagères. C’est la perspective que nous avons explorée in situ, en comparant les différentes espèces végétales; cette collection d’informations nous permet de proposer une typologie d’ambiances sonores des végétaux. Dans la première partie, des notions rattachées au « monde sonore » telles que l’objet sonore, le paysage sonore et les effets sonores justifient d’établir, dans la méthodologie, une grille d’analyse comportant différentes échelles d’écoute. Une lecture multidisciplinaire propose, d’une part, de réunir de l’information sur le son et les végétaux, la morphologie de ces derniers, l’aménagement au site, les conditions climatiques et, d’autre part, de retrouver ce qui a trait au son dans l’histoire des jardins, dont les jardins sensoriels, thérapeutiques, technologiques, et des sentiers d’interprétation sonore, sous l’angle du son comme projet. De plus, une liste de végétaux recevant les chants et cris de la faune vient introduire la notion de biodiversité sonore. Une enquête sociale de terrain, par la méthode des parcours commentés, et une enquête « experte » ont été réalisées au Jardin botanique de Montréal. Ces deux enquêtes nous ont permis de constituer une grille d’analyse croisée comprenant plusieurs échelles d’écoute : textures, actions sonores, effets sonores... De là, des générateurs d’ambiance (morphologie, organisation, climat) ont été relevés pour déterminer les aspects de récurrences et de différenciations d’un type d’ambiance sonore à l’autre. Des associations se sont formées en fonction de onze types de végétaux, chacun comprenant plusieurs sous-catégories. Celles-ci proposent des ambiances sonores spécifiques, des échelles d’écoute à considérer pour chaque type d’ambiance et l’énumération d’espèces à utiliser. Cette recherche ouvre la voie à un autre type de lecture sonore, par thématique d’ambiance (les sons du végétal dans notre cas), afin d’offrir de nouveaux outils de conception pour les professionnels, en profonde relation avec les perceptions sonores d’usagers sur le terrain et l’agencement spécifique d’un site.

Que pensent les enseignants et les élèves du secondaire des TIC? : une étude des représentations sociales au Niger

Dans les pays africains en général et au Niger en particulier, l’ordinateur commence à trouver sa place au sein de l’école. Ce processus d’intégration pédagogique des technologies de l’information et de la communication est fortement dépendant des paramètres technologiques, pédagogiques et humains. Réussir ce processus, c’est connaitre les jugements et les attitudes des acteurs de cette école vis-à-vis de ces technologies. À partir des représentations sociales, nous pouvons connaître la manière de penser, de s'approprier, d'interpréter la réalité quotidienne d’un groupe d’individus vis-à-vis d’un objet, en l’occurrence ici, les TIC. Le premier objectif de notre recherche est de connaitre les représentations sociales que les enseignants du secondaire ont de l’ordinateur. Pour cela, nous avons mené une investigation auprès de vingt enseignants. À l’issue des entrevues, nous avons identifié des représentations sociales sur les attitudes des enseignants à l’égard de l’ordinateur, sur les risques de l’ordinateur à l’école et enfin sur les avantages de l’ordinateur à l’école. L’ensemble de ces représentations sociales met en évidence une attitude positive des enseignants vis-à-vis de l’ordinateur à l’école même si pour certains enseignants, l’ordinateur est un outil qui favorise la démotivation des élèves et qui est vu comme un concurrent potentiel. Notre second objectif est de connaître les représentations sociales des TIC chez des élèves. Cinquante élèves provenant de deux écoles ont constitué la population de notre étude. Après avoir soumis les entretiens recueillis à une analyse de contenu, nous sommes parvenu à des résultats qui mettent en évidence des représentations sociales sur les connaissances des TIC et sur son utilisation dans le cadre de l’apprentissage. Notre recherche devait aussi déterminer si les représentations sociales étaient fonction de la formation aux TIC. Les résultats du test d’hypothèse du Khi-carré montrent que la formation aux TIC n’influence pas l’émergence des représentations sociales. Le troisième objectif de notre recherche porte sur la construction d’une échelle de mesure des représentations sociales des TIC chez des élèves. Après avoir été élaborée l’échelle de mesure a été administrée à 265 élèves de deux lycées. Les résultats de l’analyse factorielle exploratoire montrent que l’échelle a une assez bonne cohérence interne. L’ensemble des résultats garantit la validité et la fidélité de l’échelle. L’ensemble des résultats montre que, bien que son utilisation soit très récente dans les écoles nigériennes, des représentations sociales ont été engendrées à l’endroit des TIC. Les contenus de ces représentations montrent des attitudes favorables aux TIC dans l’apprentissage.

Scoliosis curve type classification from 3D trunk image

Adolescent idiopathic scoliosis (AIS) is a deformity of the spine manifested by asymmetry and deformities of the external surface of the trunk. Classification of scoliosis deformities according to curve type is used to plan management of scoliosis patients. Currently, scoliosis curve type is determined based on X-ray exam. However, cumulative exposure to X-rays radiation significantly increases the risk for certain cancer. In this paper, we propose a robust system that can classify the scoliosis curve type from non invasive acquisition of 3D trunk surface of the patients. The 3D image of the trunk is divided into patches and local geometric descriptors characterizing the surface of the back are computed from each patch and forming the features. We perform the reduction of the dimensionality by using Principal Component Analysis and 53 components were retained. In this work a multi-class classifier is built with Least-squares support vector machine (LS-SVM) which is a kernel classifier. For this study, a new kernel was designed in order to achieve a robust classifier in comparison with polynomial and Gaussian kernel. The proposed system was validated using data of 103 patients with different scoliosis curve types diagnosed and classified by an orthopedic surgeon from the X-ray images. The average rate of successful classification was 93.3% with a better rate of prediction for the major thoracic and lumbar/thoracolumbar types.