Improving generalization capabilities of dynamic neural networks

This work addresses the problem of improving the generalization capabilities of continuous recurrent neural networks. The learning task is transformed into an optimal control framework in which the weights and the initial network state are treated as unknown controls. A new learning algorithm based on a variational formulation of Pontrayagin's maximum principle is proposed. Under reasonable assumptions, its convergence is discussed. Numerical examples are given that demonstrate an essential improvement of generalization capabilities after the learning process of a dynamic network.     

New indicator for optimal preprocessing and wavelength selection of near-infrared spectra

Preprocessing of near-infrared spectra to remove unwanted, i.e., non-related spectral variation and selection of informative wavelengths is considered to be a crucial step prior to the construction of a quantitative calibration model. The standard methodology when comparing various preprocessing techniques and selecting different wavelengths is to compare prediction statistics computed with an independent set of data not used to make the actual calibration model. When the errors of reference value are large, no such values are available at all, or only a limited number of samples are available, other methods exist to evaluate the preprocessing method and wavelength selection. In this work we present a new indicator (SE) that only requires blank sample spectra, i.e., spectra of samples that are mixtures of the interfering constituents (everything except the analyte), a pure analyte spectrum, or alternatively, a sample spectrum where the analyte is present. The indicator is based on computing the net analyte signal of the analyte and the total error, i.e., instrumental noise and bias. By comparing the indicator values when different preprocessing techniques and wavelength selections are applied to the spectra, the optimal preprocessing technique and the optimal wavelength selection can be determined without knowledge of reference values, i.e., it minimizes the non-related spectral variation. The SE indicator is compared to two other indicators that also use net analyte signal computations. To demonstrate the feasibility of the SE indicator, two near-infrared spectral data sets from the pharmaceutical industry were used, i.e., diffuse reflectance spectra of powder samples and transmission spectra of tablets. Especially in pharmaceutical spectroscopic applications, it is expected beforehand that the non-related spectral variation is rather large and it is important to remove it. The indicator gave excellent results with respect to wavelength selection and optimal preprocessing. The SE indicator performs better than the two other indicators, and it is also applicable to other situations where the Beer-Lambert law is valid.     

A Simple Method to Extract Essential Oils from Tissue Samples by Using Microwave Radiation

A microwave protocol to extract lipophilic substances from tissue was modified to extract essential oils (EOs) from plant tissue and insect feculae. The material, in a solvent transparent to microwave radiation, is exposed for a short time to steam in a microwave oven. EO extracts are analyzed directly by GC or GC-MS when plant material is fresh and terpenes contained in glandular structures on leaf surfaces are readily released into the solvent. For dried material or insect feculae, mechanical means are utilized first to break up tissue; however, the complete procedure is carried out inside the same vial to reduce losses. Statistical analysis shows that the reproducibility of the modified method is high. Several samples can be run within an hour with this method.     

Contributions to Oil-Spill Detection and Analysis with Radar and Microwave Radiometry: Results of the Archimedes II Campaign

During the Archimedes II campaign in November 1985 (conducted and sponsored by JRC ISPRA) different DFVLR-instruments were flown, an X-band SLAR, an elementary L-band SAR, and microwaveradiometers at 32 and 90 GHz. The objective of these measurements was to evaluate the possibility of detection, localization, and qualification of oil pollution with microwave sensors. Examples of measurement results obtained are presented. As a main result, it can be stated that all of these microwave instruments are valuable and necessary tools for oil pollution detection andoil collection systems.     

Traumatic epidural hematoma of the cervical spine: magnetic resonance imaging diagnosis and spontaneous resolution: case report

OBJECTIVE AND IMPORTANCE: Demonstration of interest of medical treatment of cervical epidural hematomas. CLINICAL PRESENTATION: A young patient developed cervical pain after experiencing cervical trauma. Computed tomography and magnetic resonance imaging demonstrated an epidural cervical hematoma. A spontaneous resolution of the clinical symptoms and the radiological abnormalities was observed. CONCLUSION: Although surgical decompression is generally regarded as mandatory in selected patients with incomplete and nonprogressing deficits, conservative management may be possible.