Evidence evaluation in fingerprint comparison and automated fingerprint identification systems: modeling between finger variability

In the context of the investigation of the use of automated fingerprint identification systems (AFIS) for the evaluation of fingerprint evidence, the current study presents investigations into the variability of scores from an AFIS system when fingermarks from a known donor are compared to fingerprints that are not from the same source. The ultimate goal is to propose a model, based on likelihood ratios, which allows the evaluation of mark-to-print comparisons. In particular, this model, through its use of AFIS technology, benefits from the possibility of using a large amount of data, as well as from an already built-in proximity measure, the AFIS score. More precisely, the numerator of the LR is obtained from scores issued from comparisons between impressions from the same source and showing the same minutia configuration. The denominator of the LR is obtained by extracting scores from comparisons of the questioned mark with a database of non-matching sources. This paper focuses solely on the assignment of the denominator of the LR. We refer to it by the generic term of between-finger variability. The issues addressed in this paper in relation to between-finger variability are the required sample size, the influence of the finger number and general pattern, as well as that of the number of minutiae included and their configuration on a given finger. Results show that reliable estimation of between-finger variability is feasible with 10,000 scores. These scores should come from the appropriate finger number/general pattern combination as defined by the mark. Furthermore, strategies of obtaining between-finger variability when these elements cannot be conclusively seen on the mark (and its position with respect to other marks for finger number) have been presented. These results immediately allow case-by-case estimation of the between-finger variability in an operational setting.

Biological exposure indicators: quantification of biological variability using toxicokinetic modeling

Compartmental and physiologically based toxicokinetic modeling coupled with Monte Carlo simulation were used to quantify the impact of biological variability (physiological, biochemical, and anatomic parameters) on the values of a series of bio-indicators of metal and organic industrial chemical exposures. A variability extent index and the main parameters affecting biological indicators were identified. Results show a large diversity in interindividual variability for the different categories of biological indicators examined. Measurement of the unchanged substance in blood, alveolar air, or urine is much less variable than the measurement of metabolites, both in blood and urine. In most cases, the alveolar flow and cardiac output were identified as the prime parameters determining biological variability, thus suggesting the importance of workload intensity on absorbed dose for inhaled chemicals.

Preliminary regional rockfall hazard mapping using LIDAR-based slope frequency distribution and CONEFALL modeling

A factor limiting preliminary rockfall hazard mapping at regional scale is often the lack of knowledge of potential source areas. Nowadays, high resolution topographic data (LiDAR) can account for realistic landscape details even at large scale. With such fine-scale morphological variability, quantitative geomorphometric analyses become a relevant approach for delineating potential rockfall instabilities. Using digital elevation model (DEM)-based ?slope families? concept over areas of similar lithology and cliffs and screes zones available from the 1:25,000 topographic map, a susceptibility rockfall hazard map was drawn up in the canton of Vaud, Switzerland, in order to provide a relevant hazard overview. Slope surfaces over morphometrically-defined thresholds angles were considered as rockfall source zones. 3D modelling (CONEFALL) was then applied on each of the estimated source zones in order to assess the maximum runout length. Comparison with known events and other rockfall hazard assessments are in good agreement, showing that it is possible to assess rockfall activities over large areas from DEM-based parameters and topographical elements.

Imatinib plasma concentrations variability in oncologic patients

Imatinib (Glivec®) has transformed the treatment and short-term prognosis of chronic myeloid leukemia (CML) and gastrointestinal stromal tumor (GIST). However, the treatment must be taken indefinitely, it is not devoid of inconvenience and toxicity. Moreover, resistance or escape from disease control occurs in a significant number of patients. Imatinib is a substrate of the cytochromes P450 CYP3A4/5 and of the multidrug transporter P-glycoprotein (product of the MDR1 gene). Considering the large inter-individual differences in the expression and function of those systems, the disposition and clinical activity of imatinib can be expected to vary widely among patients, calling for dosage individualization. The aim of this exploratory study was to determine the average pharmacokinetic parameters characterizing the disposition of imatinib in the target population, to assess their inter-individual variability, and to identify influential factors affecting them. A total of 321 plasma concentrations, taken at various sampling times after the latest dose, were measured in 59 patients receiving Glivec at diverse regimens, using a validated HPLC-UV method developed for this study. The results were analyzed by non-linear mixed effect modeling (NONMEM). A one-compartment model with first-order absorption appeared appropriate to describe the data, with an average apparent clearance of 12.4 l/h, a distribution volume of 268 l and an absorption constant of 0.47 h-1. The clearance was affected by body weight, age and sex. No influences of interacting drugs were found. DNA samples were used for pharmacogenetic explorations. At present, only the MDR1 polymorphism has been assessed and seems to affect the pharmacokinetic parameters of imatinib. Large inter-individual variability remained unexplained by the demographic covariates considered, both on clearance (40 %) and distribution volume (71 %). Together with intra-patient variability (34 %), this translates into an 8-fold width of the 90 %-prediction interval of plasma concentrations expected under a fixed dosing regimen. This is a strong argument to further investigate the possible usefulness of a therapeutic drug monitoring program for imatinib. It may help to individualize the dosing regimen before overt disease progression or observation of treatment toxicity, thus improving both the long-term therapeutic effectiveness and tolerability of this drug.

The variability of peridotite composition across a mantle shear zone (Lanzo massif, Italy): interplay of melt focusing and deformation

In this paper we present new data on the spatial variability of peridotite composition across a kilometer-scale mantle shear zone within the Lanzo massif (Western Alps, Italy). The shear zone separates the central from the northern part of the massif. Plagioclase peridotite shows gradually increasing deformation towards the shear zone, from porphyroclastic to mylonitic textures in the central body, while the northern body is composed of porphyroclastic rocks. The peridotite displays a large range of compositions, from fertile peridotite to refractory harzburgite and dunite. Deformed peridotites (proto-mylonite and mylonites) tend to be compositionally more homogeneous and fertile than weakly deformed peridotites. The composition of most plagioclase peridotites show rather high and constant (Ce/Yb) (N) ratios, and Yb (N) that cannot be explained by any simple melting model. Instead, refertilization modeling, consisting of melt increments from spinel peridotite sources, particularly with E-MORB melt, reasonably reproduces the plagioclase peridotite whole rock composition. Combined with constraints from Ce-Nb and Ce-Th systematics, we speculate that peridotites such as those from Lanzo record pervasive refertilization processes in the thermal boundary layer. In this scenario, mantle shear zones might act as important areas of melt focusing in the upper mantle that separates the thermal boundary layer from the conductively cooled mantle.

Carbon and oxygen isotope study of hydrothermal carbonates in the zinc-lead deposits of the San Vicente district, central Peru: A quantitative modeling on mixing processes and CO2 degassing

Carbon and oxygen isotope studies of the host and gangue carbonates of Mississippi Valley-type zinc-lead deposits in the San Vicente District hosted in the Upper Triassic to Lower Jurassic dolostones of the Pucara basin (central Peru) were used to constrain models of the ore formation. A mixing model between an incoming hot saline slightly acidic radiogenic (Pb, Sr) fluid and the native formation water explains the overall isotopic variation (delta(13)C = - 11.5 to + 2.5 parts per thousand relative to PDB and delta(18)O = + 18.0 to + 24.3 parts per thousand relative to SMOW) of the carbonate generations. The dolomites formed during the main ore stage show a narrower range (delta(13)C = - 0.1 to + 1.7 parts per thousand and delta(18)O = + 18.7 to + 23.4 parts per thousand) which is explained by exchange between the mineralizing fluids and the host carbonates combined with changes in temperature and pressure. This model of fluid-rock interaction explains the pervasive alteration of the host dolomite I and precipitation of sphalerite I. The open-space filling hydrothermal white sparry dolomite and the coexisting sphalerite II formed by prolonged fluid-host dolomite interaction and limited CO2 degassing. Late void-filling dolomite III (or calcite) and the associated sphalerite III formed as the consequence of CO2 degassing and concomitant pH increase of a slightly acidic ore fluid. Widespread brecciation is associated to CO2 outgassing. Consequently, pressure variability plays a major role in the ore precipitation during the late hydrothermal events in San Vicente. The presence of native sulfur associated with extremely carbon-light calcites replacing evaporitic sulfates (e.g., delta(13)C = - 11.5 parts per thousand), altered native organic matter and heavier hydrothermal bitumen (from - 27.0 to - 23.0 parts per thousand delta(13)C) points to thermochemical reduction of sulfate and/or thiosulfate. The delta(13)C- and delta(18)O-values of the altered host dolostone and hydrothermal carbonates, and the carbon isotope composition of the associated organic matter show a strong regional homogeneity. These results coupled with the strong mineralogical and petrographic similarities of the different MVT occurrences perhaps reflects the fact that the mineralizing processes were similar in the whole San Vicente belt, suggesting the existence of a common regional mineralizing hydrothermal system with interconnected plumbing.

Evaluation of occupational exposure: comparison of biological and environmental variabilities using physiologically based toxicokinetic modeling

PURPOSE: Few studies compare the variabilities that characterize environmental (EM) and biological monitoring (BM) data. Indeed, comparing their respective variabilities can help to identify the best strategy for evaluating occupational exposure. The objective of this study is to quantify the biological variability associated with 18 bio-indicators currently used in work environments. METHOD: Intra-individual (BV(intra)), inter-individual (BV(inter)), and total biological variability (BV(total)) were quantified using validated physiologically based toxicokinetic (PBTK) models coupled with Monte Carlo simulations. Two environmental exposure profiles with different levels of variability were considered (GSD of 1.5 and 2.0). RESULTS: PBTK models coupled with Monte Carlo simulations were successfully used to predict the biological variability of biological exposure indicators. The predicted values follow a lognormal distribution, characterized by GSD ranging from 1.1 to 2.3. Our results show that there is a link between biological variability and the half-life of bio-indicators, since BV(intra) and BV(total) both decrease as the biological indicator half-lives increase. BV(intra) is always lower than the variability in the air concentrations. On an individual basis, this means that the variability associated with the measurement of biological indicators is always lower than the variability characterizing airborne levels of contaminants. For a group of workers, BM is less variable than EM for bio-indicators with half-lives longer than 10-15 h. CONCLUSION: The variability data obtained in the present study can be useful in the development of BM strategies for exposure assessment and can be used to calculate the number of samples required for guiding industrial hygienists or medical doctors in decision-making.

Correction for heart rate variability during 3D whole heart MR coronary angiography.

PURPOSE: To evaluate the effect of a real-time adaptive trigger delay on image quality to correct for heart rate variability in 3D whole-heart coronary MR angiography (MRA). MATERIALS AND METHODS: Twelve healthy adults underwent 3D whole-heart coronary MRA with and without the use of an adaptive trigger delay. The moment of minimal coronary artery motion was visually determined on a high temporal resolution MRI. Throughout the scan performed without adaptive trigger delay, trigger delay was kept constant, whereas during the scan performed with adaptive trigger delay, trigger delay was continuously updated after each RR-interval using physiological modeling. Signal-to-noise, contrast-to-noise, vessel length, vessel sharpness, and subjective image quality were compared in a blinded manner. RESULTS: Vessel sharpness improved significantly for the middle segment of the right coronary artery (RCA) with the use of the adaptive trigger delay (52.3 +/- 7.1% versus 48.9 +/- 7.9%, P = 0.026). Subjective image quality was significantly better in the middle segments of the RCA and left anterior descending artery (LAD) when the scan was performed with adaptive trigger delay compared to constant trigger delay. CONCLUSION: Our results demonstrate that the use of an adaptive trigger delay to correct for heart rate variability improves image quality mainly in the middle segments of the RCA and LAD.

Statistical Modeling of the Eye for Multimodal Treatment Planning for External Beam Radiation Therapy of Intraocular Tumors.

PURPOSE: Ocular anatomy and radiation-associated toxicities provide unique challenges for external beam radiation therapy. For treatment planning, precise modeling of organs at risk and tumor volume are crucial. Development of a precise eye model and automatic adaptation of this model to patients' anatomy remain problematic because of organ shape variability. This work introduces the application of a 3-dimensional (3D) statistical shape model as a novel method for precise eye modeling for external beam radiation therapy of intraocular tumors. METHODS AND MATERIALS: Manual and automatic segmentations were compared for 17 patients, based on head computed tomography (CT) volume scans. A 3D statistical shape model of the cornea, lens, and sclera as well as of the optic disc position was developed. Furthermore, an active shape model was built to enable automatic fitting of the eye model to CT slice stacks. Cross-validation was performed based on leave-one-out tests for all training shapes by measuring dice coefficients and mean segmentation errors between automatic segmentation and manual segmentation by an expert. RESULTS: Cross-validation revealed a dice similarity of 95% ± 2% for the sclera and cornea and 91% ± 2% for the lens. Overall, mean segmentation error was found to be 0.3 ± 0.1 mm. Average segmentation time was 14 ± 2 s on a standard personal computer. CONCLUSIONS: Our results show that the solution presented outperforms state-of-the-art methods in terms of accuracy, reliability, and robustness. Moreover, the eye model shape as well as its variability is learned from a training set rather than by making shape assumptions (eg, as with the spherical or elliptical model). Therefore, the model appears to be capable of modeling nonspherically and nonelliptically shaped eyes.

Dosage optimization of treatments using population pharmacokinetic modeling and simulation.

Pharmacokinetic variability in drug levels represent for some drugs a major determinant of treatment success, since sub-therapeutic concentrations might lead to toxic reactions, treatment discontinuation or inefficacy. This is true for most antiretroviral drugs, which exhibit high inter-patient variability in their pharmacokinetics that has been partially explained by some genetic and non-genetic factors. The population pharmacokinetic approach represents a very useful tool for the description of the dose-concentration relationship, the quantification of variability in the target population of patients and the identification of influencing factors. It can thus be used to make predictions and dosage adjustment optimization based on Bayesian therapeutic drug monitoring (TDM). This approach has been used to characterize the pharmacokinetics of nevirapine (NVP) in 137 HIV-positive patients followed within the frame of a TDM program. Among tested covariates, body weight, co-administration of a cytochrome (CYP) 3A4 inducer or boosted atazanavir as well as elevated aspartate transaminases showed an effect on NVP elimination. In addition, genetic polymorphism in the CYP2B6 was associated with reduced NVP clearance. Altogether, these factors could explain 26% in NVP variability. Model-based simulations were used to compare the adequacy of different dosage regimens in relation to the therapeutic target associated with treatment efficacy. In conclusion, the population approach is very useful to characterize the pharmacokinetic profile of drugs in a population of interest. The quantification and the identification of the sources of variability is a rational approach to making optimal dosage decision for certain drugs administered chronically.

Distribution of metals exposure and associations with cardiometabolic risk factors in the "Modeling the Epidemiologic Transition Study".

BACKGROUND: Metals are known endocrine disruptors and have been linked to cardiometabolic diseases via multiple potential mechanisms, yet few human studies have both the exposure variability and biologically-relevant phenotype data available. We sought to examine the distribution of metals exposure and potential associations with cardiometabolic risk factors in the "Modeling the Epidemiologic Transition Study" (METS), a prospective cohort study designed to assess energy balance and change in body weight, diabetes and cardiovascular disease risk in five countries at different stages of social and economic development. METHODS: Young adults (25-45 years) of African descent were enrolled (N = 500 from each site) in: Ghana, South Africa, Seychelles, Jamaica and the U.S.A. We randomly selected 150 blood samples (N = 30 from each site) to determine concentrations of selected metals (arsenic, cadmium, lead, mercury) in a subset of participants at baseline and to examine associations with cardiometabolic risk factors. RESULTS: Median (interquartile range) metal concentrations (μg/L) were: arsenic 8.5 (7.7); cadmium 0.01 (0.8); lead 16.6 (16.1); and mercury 1.5 (5.0). There were significant differences in metals concentrations by: site location, paid employment status, education, marital status, smoking, alcohol use, and fish intake. After adjusting for these covariates plus age and sex, arsenic (OR 4.1, 95% C.I. 1.2, 14.6) and lead (OR 4.0, 95% C.I. 1.6, 9.6) above the median values were significantly associated with elevated fasting glucose. These associations increased when models were further adjusted for percent body fat: arsenic (OR 5.6, 95% C.I. 1.5, 21.2) and lead (OR 5.0, 95% C.I. 2.0, 12.7). Cadmium and mercury were also related with increased odds of elevated fasting glucose, but the associations were not statistically significant. Arsenic was significantly associated with increased odds of low HDL cholesterol both with (OR 8.0, 95% C.I. 1.8, 35.0) and without (OR 5.9, 95% C.I. 1.5, 23.1) adjustment for percent body fat. CONCLUSIONS: While not consistent for all cardiometabolic disease markers, these results are suggestive of potentially important associations between metals exposure and cardiometabolic risk. Future studies will examine these associations in the larger cohort over time.

Environmental data mining and modeling based on machine learning algorithms and geostatistics

The paper presents some contemporary approaches to spatial environmental data analysis. The main topics are concentrated on the decision-oriented problems of environmental spatial data mining and modeling: valorization and representativity of data with the help of exploratory data analysis, spatial predictions, probabilistic and risk mapping, development and application of conditional stochastic simulation models. The innovative part of the paper presents integrated/hybrid model-machine learning (ML) residuals sequential simulations-MLRSS. The models are based on multilayer perceptron and support vector regression ML algorithms used for modeling long-range spatial trends and sequential simulations of the residuals. NIL algorithms deliver non-linear solution for the spatial non-stationary problems, which are difficult for geostatistical approach. Geostatistical tools (variography) are used to characterize performance of ML algorithms, by analyzing quality and quantity of the spatially structured information extracted from data with ML algorithms. Sequential simulations provide efficient assessment of uncertainty and spatial variability. Case study from the Chernobyl fallouts illustrates the performance of the proposed model. It is shown that probability mapping, provided by the combination of ML data driven and geostatistical model based approaches, can be efficiently used in decision-making process. (C) 2003 Elsevier Ltd. All rights reserved.

Imatinib plasma concentrations variability in hemato-oncologic patients

Abstract Imatinib (Glivec~ has transformed the treatment and prognosis of chronic myeloid leukaemia (CML) and of gastrointestinal stromal tumor (GIST). However, the treatment must be taken indefinitely and is not devoid of inconvenience and toxicity. Moreover, resistance or escape from disease control occurs. Considering the large interindividual differences in the function of the enzymatic and transport systems involved in imatinib disposition, exposure to this drug can be expected to vary widely among patients. Among those known systems is a cytochrome P450 (CYI'3A4) that metabolizes imatinib, the multidrug transporter P-glycoprotein (P-gp; product of the MDR1 gene) that expels imatinib out of cells, and al-acid glycoprotein (AGP), a circulating protein binding imatinib in the plasma. The aim of this observational study was to explore the influence of these covariates on imatinib pharmacokinetics (PK), to assess the interindividual variability of the PK parameters of the drug, and to evaluate whether imatinib use would benefit from a therapeutic drug monitoring (TDM) program. A total of 321 plasma concentrations were measured in 59 patients receiving imatinib, using a validated chromatographic method developed for this study (HPLC-LTV). The results were analyzed by non-linear mixed effect modeling (NONMEM). A one-compartment pharmacokinetic model with first-order absorption appropriately described the data, and a large interindividual variability was observed. The MDK> polymorphism 3435C>T and the CYP3A4 activity appeared to modulate the disposition of imatinib, albeit not significantly. A hyperbolic relationship between plasma AGP levels and oral clearance, as well as volume of distribution, was observed. A mechanistic approach was built up, postulating that only the unbound imatinib concentration was able to undergo first-order elimination. This approach allowed determining an average free clearance (CL,~ of 13101/h and a volume of distribution (Vd) of 301 1. By comparison, the total clearance determined was 141/h (i.e. 233 ml/min). Free clearance was affected by body weight and pathology diagnosis. The estimated variability of imatinib disposition (17% for CLu and 66% for Vd) decreased globally about one half with the model incorporating the AGP impact. Moreover, some associations were observed between PK parameters of the free imatinib concentration and its efficacy and toxicity. Finally, the functional influence of P-gp activity has been demonstrated in vitro in cell cultures. These elements are arguments to further investigate the possible usefulness of a TDM program for imatinib. It may help in individualizing the dosing regimen before overt disease progression or development of treatment toxicity, thus improving both the long-term therapeutic effectiveness and tolerability of this drug. Résumé L'imatinib (Glivec ®) a révolutionné le traitement et le pronostic de la leucémie myéloïde chronique (LMC) et des tumeurs stromales d'origine digestive (GIST). Il s'agit toutefois d'un traitement non dénué d'inconvénients et de toxicité, et qui doit être pris indéfiniment. Par ailleurs, une résistance, ou des échappements au traitement, sont également rencontrés. Le devenir de ce médicament dans l'organisme dépend de systèmes enzymatiques et de transport connus pour présenter de grandes différences interindividuelles, et l'on peut s'attendre à ce que l'exposition à ce médicament varie largement d'un patient à l'autre. Parmi ces systèmes, on note un cytochrome P450 (le CYP3A4) métabolisant l'imatinib, la P-glycoprotéine (P-gp ;codée par le gène MDR1), un transporteur d'efflux expulsant le médicament hors des cellules, et l'atglycoprotéine acide (AAG), une protéine circulante sur laquelle se fixe l'imatinib dans le plasma. L'objectif de la présente étude clinique a été de déterminer l'influence de ces covariats sur la pharmacocinétique (PK) de l'imatinib, d'établir la variabilité interindividuelle des paramètres PK du médicament, et d'évaluer dans quelle mesure l'imatinib pouvait bénéficier d'un programme de suivi thérapeutique (TDM). En utilisant une méthode chromatographique développée et validée à cet effet (HPLC-UV), un total de 321 concentrations plasmatiques a été dosé chez 59 patients recevant de l'imatinib. Les résultats ont été analysés par modélisation non linéaire à effets mixtes (NONMEM). Un modèle pharmacocinétique à un compartiment avec absorption de premier ordre a permis de décrire les données, et une grande variabilité interindividuelle a été observée. Le polymorphisme du gène MDK1 3435C>T et l'activité du CYP3A4 ont montré une influence, toutefois non significative, sur le devenir de l'imatinib. Une relation hyperbolique entre les taux plasmatiques d'AAG et la clairance, comme le volume de distribution, a été observée. Une approche mécanistique a donc été élaborée, postulant que seule la concentration libre subissait une élimination du premier ordre. Cette approche a permis de déterminer une clairance libre moyenne (CLlibre) de 13101/h et un volume de distribution (Vd) de 301 l. Par comparaison, la clairance totale était de 141/h (c.à.d. 233 ml/min). La CLlibre est affectée par le poids corporel et le type de pathologie. La variabilité interindividuelle estimée pour le devenir de l'imatinib (17% sur CLlibre et 66% sur Vd) diminuait globalement de moitié avec le modèle incorporant l'impact de l'AAG. De plus, une certaine association entre les paramètres PK de la concentration d'imatinib libre et l'efficacité et la toxicité a été observée. Finalement, l'influence fonctionnelle de l'activité de la P-gp a été démontrée in nitro dans des cultures cellulaires. Ces divers éléments constituent des arguments pour étudier davantage l'utilité potentielle d'un programme de TDM appliqué à l'imatinib. Un tel suivi pourrait aider à l'individualisation des régimes posologiques avant la progression manifeste de la maladie ou l'apparition de toxicité, améliorant tant l'efficacité que la tolérabilité de ce médicament. Résumé large public L'imatinib (un médicament commercialisé sous le nom de Glivec ®) a révolutionné le traitement et le pronostic de deux types de cancers, l'un d'origine sanguine (leucémie) et l'autre d'origine digestive. Il s'agit toutefois d'un traitement non dénué d'inconvénients et de toxicité, et qui doit être pris indéfiniment. De plus, des résistances ou des échappements au traitement sont également rencontrés. Le devenir de ce médicament dans le corps humain (dont l'étude relève de la discipline appelée pharmacocinétique) dépend de systèmes connus pour présenter de grandes différences entre les individus, et l'on peut s'attendre à ce que l'exposition à ce médicament varie largement d'un patient à l'autre. Parmi ces systèmes, l'un est responsable de la dégradation du médicament dans le foie (métabolisme), l'autre de l'expulsion du médicament hors des cellules cibles, alors que le dernier consiste en une protéine (dénommée AAG) qui transporte l'imatinib dans le sang. L'objectif de notre étude a été de déterminer l'influence de ces différents systèmes sur le comportement pharmacocinétique de l'imatinib chez les patients, et d'étudier dans quelle mesure le devenir de ce médicament dans l'organisme variait d'un patient à l'autre. Enfin, cette étude avait pour but d'évaluer à quel point la surveillance des concentrations d'imatinib présentes dans le sang pourrait améliorer le traitement des patients cancéreux. Une telle surveillance permet en fait de connaître l'exposition effective de l'organisme au médicament (concept abrégé par le terme anglais TDM, pour Therapeutic Drag Monitoring. Ce projet de recherche a d'abord nécessité la mise au point d'une méthode d'analyse pour la mesure des quantités (ou concentrations) d'imatinib présentes dans le sang. Cela nous a permis d'effectuer régulièrement des mesures chez 59 patients. Il nous a ainsi été possible de décrire le devenir du médicament dans le corps à l'aide de modèles mathématiques. Nous avons notamment pu déterminer chez ces patients la vitesse à laquelle l'imatinib est éliminé du sang et l'étendue de sa distribution dans l'organisme. Nous avons également observé chez les patients que les concentrations sanguines d'imatinib étaient très variables d'un individu à l'autre pour une même dose de médicament ingérée. Nous avons pu aussi mettre en évidence que les concentrations de la protéine AAG, sur laquelle l'imatinib se lie dans le sang, avait une grande influence sur la vitesse à laquelle le médicament est éliminé de l'organisme. Ensuite, en tenant compte des concentrations sanguines d'imatinib et de cette protéine, nous avons également pu calculer les quantités de médicament non liées à cette protéine (= libres), qui sont seules susceptibles d'avoir une activité anticancéreuse. Enfin, il a été possible d'établir qu'il existait une certaine relation entre ces concentrations, l'effet thérapeutique et la toxicité du traitement. Tous ces éléments constituent des arguments pour approfondir encore l'étude de l'utilité d'un programme de TDM appliqué à l'imatinib. Comme chaque patient est différent, un tel suivi pourrait aider à l'ajustement des doses du médicament avant la progression manifeste de la maladie ou l'apparition de toxicité, améliorant ainsi tant son efficacité que son innocuité.

Functional error modeling for uncertainty quantification in hydrogeology

Approximate models (proxies) can be employed to reduce the computational costs of estimating uncertainty. The price to pay is that the approximations introduced by the proxy model can lead to a biased estimation. To avoid this problem and ensure a reliable uncertainty quantification, we propose to combine functional data analysis and machine learning to build error models that allow us to obtain an accurate prediction of the exact response without solving the exact model for all realizations. We build the relationship between proxy and exact model on a learning set of geostatistical realizations for which both exact and approximate solvers are run. Functional principal components analysis (FPCA) is used to investigate the variability in the two sets of curves and reduce the dimensionality of the problem while maximizing the retained information. Once obtained, the error model can be used to predict the exact response of any realization on the basis of the sole proxy response. This methodology is purpose-oriented as the error model is constructed directly for the quantity of interest, rather than for the state of the system. Also, the dimensionality reduction performed by FPCA allows a diagnostic of the quality of the error model to assess the informativeness of the learning set and the fidelity of the proxy to the exact model. The possibility of obtaining a prediction of the exact response for any newly generated realization suggests that the methodology can be effectively used beyond the context of uncertainty quantification, in particular for Bayesian inference and optimization.