Účel studie: Hlavním cílem studie bylo odhadnout dosažení patologické kompletní odpovědi nemocných trpících lokálně pokročilým adenokarcinomem rekta po předoperační (neoadjuvantní) chemoradioterapii (CHRT) s využitím individuálních farmakokinetických parametrů 5-fluorouracilu (5-FU). Sekundárním cílem bylo hodnocení bezpečnosti a snášenlivosti léčby. Tato otevřená prospektivní studie podporovaná grantem IGA NS 9693–04/2008 zahrnula 34 dospělých s lokálně pokročilým karcinomem konečníku ozařovaných do celkové dávky 50,4 Gy ve 28 frakcích po 1,8 Gy 10–15 MV paprsky v režimu 5 + 2 (5 dní radioterapie, 2 dny volno) s kontinuálně podávaným 5-FU v i. v. infuzi 200–1000 mg/m2 po dobu 4–5 týdnů kontinuálně 7denní infuzí. Chirurgická resekce následovala 4–6 týdnů po ukončení CHRT a po klinickém restagingu. Hodnocena byla klinická a patologická odpověď na chemoradioterapii pomocí zobrazovacího (MR) a histopatologického vyšetření, vyjádřená v % jako reziduální choroba. Výsledky a závěr: Výsledek dokládá vztah mezi velikostí kumulativní dávky 5-FU a kumulativní AUC 5-FU (r = 0,61, p < 0,001). Podobný vztah byl prokázán mezi kumulativní AUC 5-FU a metabolickým poměrem (poměr dosažených plazmatických koncentrací neaktivního metabolitu dihydrofluorouracilu (5-FUH2) ku koncentracím 5-FU, r = -0,80, p < 0,0001). Kumulativní AUC korelovala se stupněm odpovědi na léčbu (r = – 0,53, p < 0,005) a určovala také stupeň toxicity léčby. Pokud bychom chtěli dosáhnout kompletní patologickou odpověď, pak by denní dávka 5-FU měla být u středně rychlého metabolizéra >350 mg/m2 a kumulativní AUC1–39 dní > 50 mg/L*h. U žádného z nemocných nebyla nalezena mutace v genu pro dihydropyrimidindehydrogenázu (DPD) a multidrug resistance-1 protein (MDR-1), přesto byla nalezena vysoká interindividuální variabilita v dosažených plazmatických koncentracích 5-FU, a to i s ohledem na cirkadiální rytmus kinetiky léčiva., Background and Purpose: The main goal of the present study was to estimate the early patients´response following neoadjuvant chemoradiotherapy (CHRT) based on 5-fluorouracil (5-FU) with curative aim in relation to plasma concentrations and pharmacokinetic parameters of 5-FU. Secondary objectives included evaluation of the safety and tolerability of the regimen. Patients and Methods: This open prospective study enrolles 34 adult patients with locally advanced rectal cancer, who received 5-FU 200 -1000 mg/m2 administered as a continuous i. v. infusion over 4–5 week and radiotherapy delivered with 10–15 MV photon beams at 1.8 Gy/fraction up o 50.4 Gy in 28 daily fractions for 5 days a week. Surgical resection with curative aim followed 4–6 weeks after the completion of CHRT and clinical restaging. Pathologic response evaluation and the rate of tumor regression was evaluated using tumor downstaging by MR, histopathological staging, and expressed as residual disease (%). Results and Conclusion: The outcome evidenced the correlation between the cumulative 5-FU dose and cumulative AUC of 5-FU (r = 0.61; p < 0.001). The similar relationship was demonstrated between the cumulative AUC and metabolic ratio (the plasma concentration od inactive metabolite dihydrofluotouracile 5-FUH2 to 5-FU; r = -0.80; p < 0.0001). The cumulative AUC was correlated with tumor regression rate (r = -0.53; p < 0.005) and determined toxicity grade. To reach pCR, the daily dose of 5-FU in patient with average metabolic ratio of 5-FUH2/5-FU should be >350 mg/m2 and the cumulative AUC1–39days > 50 mg/L*h. No mutation of gene for enzyme dihydropyrimidindehydrogenase (DPD) and mutidrug resistance-1 protein (MDR-1) were identified, although the interindividual variability of 5-FU plasma concentration was high, also with regard to circadial 5-FU pharmacokinetics variability., Jiří Grim, Miloš Hroch, Jaroslav Chládek, Ondřej Slanař, Jiří Petera, Jiřina Martínková, and Literatura 27
The purpose of feature selection in machine learning is at least two-fold - saving measurement acquisition costs and reducing the negative effects of the curse of dimensionality with the aim to improve the accuracy of the models and the classification rate of classifiers with respect to previously unknown data. Yet it has been shown recently that the process of feature selection itself can be negatively affected by the very same curse of dimensionality - feature selection methods may easily over-fit or perform unstably. Such an outcome is unlikely to generalize well and the resulting recognition system may fail to deliver the expectable performance. In many tasks, it is therefore crucial to employ additional mechanisms of making the feature selection process more stable and resistant the curse of dimensionality effects. In this paper we discuss three different approaches to reducing this problem. We present an algorithmic extension applicable to various feature selection methods, capable of reducing excessive feature subset dependency not only on specific training data, but also on specific criterion function properties. Further, we discuss the concept of criteria ensembles, where various criteria vote about feature inclusion/removal and go on to provide a general definition of feature selection hybridization aimed at combining the advantages of dependent and independent criteria. The presented ideas are illustrated through examples and summarizing recommendations are given.
We summarize the main results on probabilistic neural networks recently published in a series of papers. Considering the framework of statistical pattern recognition we assume approximation of class-conditional distributions by finite mixtures of product components. The probabilistic neurons correspond to mixture components and can be interpreted in neurophysiological terms. In this way we can find possible theoretical background of the functional properties of neurons. For example, the general formula for synaptical weights provides a statistical justification of the well known Hebbian principle of learning. Similarly, the mean effect of lateral inhibition can be expressed by means of a formula proposed by Perez as a measure of dependence tightness of involved variables.
The colorectal cancer ranks high among the malignant tumours in incidence and mortality and irinotecan is standardly used in palliative treatment of metastatic disease in every therapeutic line. Unfortunately, the treatment with irinotecan is often associated with severe toxicities, especially neutropenia and diarrhea. The majority of the toxic manifestation is caused by the insufficient deactivation (glucuronidation) of irinotecan active metabolite SN-38 by UGT1A enzyme. The elevated SN-38 plasma concentration is responsible for the hematological and gastrointestinal toxicity that can become life-threatening. The patients carrying the mutation of the gene encoding UGT1A enzyme lack the ability of bilirubin glucuronidation, and suffer from the inherited un-conjugated hyperbilirubinemia (Gilbert syndrome, Crigler-Najjar type 1 and 2 syndrome). The mutations in other enzyme systems also play role in the etiopathogenesis of the irinotecan toxicity: CYP3A (cytochrome P-450), ABC family of transmembrane transporters (adenosine-triphosphate binding cassette). The goal of the contemporary research is to determine the predictive factors that will enable the individual adjustment of the individual drug dosage while minimising the adverse effects and maintaining the treatment benefit. and A. Paulík, J. Grim, S. Filip
During the last decade we have introduced probabilistic mixture models into image modelling area, which present highly atypical and extremely demanding applications for these models. This difficulty arises from the necessity to model tens thousands correlated data simultaneously and to reliably learn such unusually complex mixture models. Presented paper surveys these novel generative colour image models based on multivariate discrete, Gaussian or Bernoulli mixtures, respectively and demonstrates their major advantages and drawbacks on texture modelling applications. Our mixture models are restricted to represent two-dimensional visual information. Thus a measured 3D multi-spectral texture is spectrally factorized and corresponding multivariate mixture models are further learned from single orthogonal mono-spectral components and used to synthesise and enlarge these mono-spectral factor components. Texture synthesis is based on easy computation of arbitrary conditional distributions from the model. Finally single synthesised mono-spectral texture planes are transformed into the required synthetic multi-spectral texture. Such models can easily serve not only for texture enlargement but also for segmentation, restoration, and retrieval or to model single factors in unusually complex seven dimensional Bidirectional Texture Function (BTF) space models. The strengths and weaknesses of the presented discrete, Gaussian or Bernoulli mixture based approaches are demonstrated on several colour texture examples.
Considering the statistical recognition of multidimensional binary observations we approximato the unknown class-conditioiial probability distributions by multivariate Bernoulli mixtures. We show that both the parameter optimization and the resulting Bayesian decision-making can be realized by a probabilistic neural network having strictly modular properties. In particular, the process of learning based on the EM algorithm can be perfomied by means of a sequential autonomous adaptation of neurons involving only the infomiation from the input synapses and the interior of neurons. In this sense the probabilistic neural network can be designed automatically. The properties of the sequential strictly modular learning procedure are illustrated by mumerical exainples.