An empirical study of machine learning techniques applied to image watermarking

A study of how the machine learning technique, known as gentleboost, could improve different digital watermarking methods such as LSB, DWT, DCT2 and Histogram shifting.

The Early Toarcian anoxia, a synchronous event in the Western Tethys? An approach by quantitative biochronology (Unitary Associations), applied on calcareous nannofossils

During the Early Toarcian, major paleoenvironnemental and paleoceanographical changes occurred, leading to an oceanic anoxic event (OAE) and to a perturbation of the carbon isotope cycle. Although the standard biochronology of the Lower Jurassic is essentially based upon ammonites, in recent years biostratigraphy based on calcareous nannofossils and dinoflagellate cysts is increasingly used to date Jurassic rocks. However, the precise dating and correlation of the Early Toarcian OAE, and of the associated delta C-13 anomaly in different settings of the western Tethys, are still partly problematic, and it is still unclear whether these events are synchronous or not. In order to allow more accurate correlations of the organic rich levels recorded in the Lower Toarcian OAE, this account proposes a new biozonation based on a quantitative biochronology approach, the Unitary Associations (UA), applied to calcareous nannofossils. This study represents the first attempt to apply the UA method to Jurassic nannofossils. The study incorporates eighteen sections distributed across western Tethys and ranging from the Pliensbachian to Aalenian, comprising 1220 samples and 72 calcareous nannofossil taxa. The BioGraph [Savary, J., Guex, J., 1999. Discrete biochronological scales and unitary associations: description of the Biograph Computer program. Memoires de Geologie de Lausanne 34, 282 pp] and UA-Graph (Copyright Hammer O., Guex and Savary, 2002) softwares provide a discrete biochronological framework based upon multi-taxa concurrent range zones in the different sections. The optimized dataset generates nine UAs using the co-occurrences of 56 taxa. These UAs are grouped into six Unitary Association Zones (UA-Z), which constitute a robust biostratigraphic synthesis of all the observed or deduced biostratigraphic relationships between the analysed taxa. The UA zonation proposed here is compared to ``classic'' calcareous nannofossil biozonations, which are commonly used for the southern and the northern sides of Tethys. The biostratigraphic resolution of the UA-Zones varies from one nannofossil subzone or part of it to several subzones, and can be related to the pattern of calcareous nannoplankton originations and extinctions during the studied time interval. The Late Pliensbachian - Early Toarcian interval (corresponding to the UA-Z II) represents a major step in the Jurassic nannoplankton radiation. The recognized UA-Zones are also compared to the carbon isotopic negative excursion and TOC maximum in five sections of central Italy, Germany and England, with the aim of providing a more reliable correlation tool for the Early Toarcian OAE, and of the associated isotopic anomaly, between the southern and northern part of western Tethys. The results of this work show that the TOC maximum and delta C-13 negative excursion correspond to the upper part of the UA-Z II (i.e., UA 3) in the sections analysed. This suggests that the Early Toarcian OAE was a synchronous event within the western Tethys. (c) 2006 Elsevier B.V. All rights reserved.

Personalized modeling for drug concentration prediction using Support Vector Machine

Building a personalized model to describe the drug concentration inside the human body for each patient is highly important to the clinical practice and demanding to the modeling tools. Instead of using traditional explicit methods, in this paper we propose a machine learning approach to describe the relation between the drug concentration and patients' features. Machine learning has been largely applied to analyze data in various domains, but it is still new to personalized medicine, especially dose individualization. We focus mainly on the prediction of the drug concentrations as well as the analysis of different features' influence. Models are built based on Support Vector Machine and the prediction results are compared with the traditional analytical models.

Towards a user-friendly webservice architecture for statistical machine translation in the PANACEA project

This paper presents a webservice architecture for Statistical Machine Translation aimed at non-technical users. A workfloweditor allows a user to combine different webservices using a graphical user interface. In the current state of this project,the webservices have been implemented for a range of sentential and sub-sententialaligners. The advantage of a common interface and a common data format allows the user to build workflows exchanging different aligners.

Interobserver and intraobserver variability of the apparent diffusion coefficient in treated malignant hepatic lesions on a 3.0T machine: measurements in the whole lesion versus in the area with the most restricted diffusion.

PURPOSE: To assess the inter/intraobserver variability of apparent diffusion coefficient (ADC) measurements in treated hepatic lesions and to compare ADC measurements in the whole lesion and in the area with the most restricted diffusion (MRDA). MATERIALS AND METHODS: Twenty-five patients with treated malignant liver lesions were examined on a 3.0T machine. After agreeing on the best ADC image, two readers independently measured the ADC values in the whole lesion and in the MRDA. These measurements were repeated 1 month later. The Bland-Altman method, Spearman correlation coefficients, and the Wilcoxon signed-rank test were used to evaluate the measurements. RESULTS: Interobserver variability for ADC measurements in the whole lesion and in the MRDA was 0.17 x 10(-3) mm(2)/s [-0.17, +0.17] and 0.43 x 10(-3) mm(2)/s [-0.45, +0.41], respectively. Intraobserver limits of agreement could be as low as [-0.10, +0.12] 10(-3) mm(2)/s and [-0.20, +0.33] 10(-3) mm(2)/s for measurements in the whole lesion and in the MRDA, respectively. CONCLUSION: A limited variability in ADC measurements does exist, and it should be considered when interpreting ADC values of hepatic malignancies. This is especially true for the measurements of the minimal ADC.

A polynomial algorithm for special case of the one-machine scheduling problem with time-lags

The standard one-machine scheduling problem consists in schedulinga set of jobs in one machine which can handle only one job at atime, minimizing the maximum lateness. Each job is available forprocessing at its release date, requires a known processing timeand after finishing the processing, it is delivery after a certaintime. There also can exists precedence constraints between pairsof jobs, requiring that the first jobs must be completed beforethe second job can start. An extension of this problem consistsin assigning a time interval between the processing of the jobsassociated with the precedence constrains, known by finish-starttime-lags. In presence of this constraints, the problem is NP-hardeven if preemption is allowed. In this work, we consider a specialcase of the one-machine preemption scheduling problem with time-lags, where the time-lags have a chain form, and propose apolynomial algorithm to solve it. The algorithm consist in apolynomial number of calls of the preemption version of the LongestTail Heuristic. One of the applicability of the method is to obtainlower bounds for NP-hard one-machine and job-shop schedulingproblems. We present some computational results of thisapplication, followed by some conclusions.

Individual Classification of Mild Cognitive Impairment Subtypes by Support Vector Machine Analysis of White Matter DTI.

BACKGROUND AND PURPOSE: MCI was recently subdivided into sd-aMCI, sd-fMCI, and md-aMCI. The current investigation aimed to discriminate between MCI subtypes by using DTI. MATERIALS AND METHODS: Sixty-six prospective participants were included: 18 with sd-aMCI, 13 with sd-fMCI, and 35 with md-aMCI. Statistics included group comparisons using TBSS and individual classification using SVMs. RESULTS: The group-level analysis revealed a decrease in FA in md-aMCI versus sd-aMCI in an extensive bilateral, right-dominant network, and a more pronounced reduction of FA in md-aMCI compared with sd-fMCI in right inferior fronto-occipital fasciculus and inferior longitudinal fasciculus. The comparison between sd-fMCI and sd-aMCI, as well as the analysis of the other diffusion parameters, yielded no significant group differences. The individual-level SVM analysis provided discrimination between the MCI subtypes with accuracies around 97%. The major limitation is the relatively small number of cases of MCI. CONCLUSIONS: Our data show that, at the group level, the md-aMCI subgroup has the most pronounced damage in white matter integrity. Individually, SVM analysis of white matter FA provided highly accurate classification of MCI subtypes.

Beyond Smith's rule: An optimal dynamic index, rule for single machine stochastic scheduling with convex holding costs

Most research on single machine scheduling has assumedthe linearity of job holding costs, which is arguablynot appropriate in some applications. This motivates ourstudy of a model for scheduling $n$ classes of stochasticjobs on a single machine, with the objective of minimizingthe total expected holding cost (discounted or undiscounted). We allow general holding cost rates that are separable,nondecreasing and convex on the number of jobs in eachclass. We formulate the problem as a linear program overa certain greedoid polytope, and establish that it issolved optimally by a dynamic (priority) index rule,whichextends the classical Smith's rule (1956) for the linearcase. Unlike Smith's indices, defined for each class, ournew indices are defined for each extended class, consistingof a class and a number of jobs in that class, and yieldan optimal dynamic index rule: work at each time on a jobwhose current extended class has larger index. We furthershow that the indices possess a decomposition property,as they are computed separately for each class, andinterpret them in economic terms as marginal expected cost rate reductions per unit of expected processing time.We establish the results by deploying a methodology recentlyintroduced by us [J. Niño-Mora (1999). "Restless bandits,partial conservation laws, and indexability. "Forthcomingin Advances in Applied Probability Vol. 33 No. 1, 2001],based on the satisfaction by performance measures of partialconservation laws (PCL) (which extend the generalizedconservation laws of Bertsimas and Niño-Mora (1996)):PCL provide a polyhedral framework for establishing theoptimality of index policies with special structure inscheduling problems under admissible objectives, which weapply to the model of concern.