Adam Jozwik

Phase distinction in semi-insulating polycrystalline silicon by pattern recognition of X-ray photoelectron spectroscopy/X-ray-induced Auger electron spectroscopy data

Abstract X-ray photoelectron spectroscopy (XPS) and X-ray-induced Auger electron spectroscopy (XAES) supported with the lineshape analysis by the pattern recognition (PR) method and the fuzzy k-nearest neighbor rule (kNN FR) were applied to study semiinsulating polycrystalline silicon layers (SIPOS). The aim of the present work was to obtain the qualitative and quantitative information about the surface region of as-received SIPOS layers. For the purpose of qualitative analysis the binding energies (BE), binding energy shifts (ΔBE), the half widths (FWHM) and the lineshapes of the Si 2p, O 1s and O KLL lines were analysed. The quantitative analysis was performed on the basis of the XPS using sensitivity factor method, multiline (ML) approach and the kNN FR. The performance of the kNN rule is possible after selecting the proper set of reference standard materials to which the rule refers during identification of an ambiguous chemical state. By selecting the reference samples supplying the information about the chemical state of Si, SiOx and SiO2, the kNN rule allowed to distinguish qualitatively these three different phases in SIPOS samples. The particular application of the kNN FR makes possible the quantitative analysis by referring the fuzzy probability of classification for the given chemical state to the concentration of particular constituents in the investigated SIPOS. All the methods applied are consistent in revealing the quantitative results and show that SIPOS is a deeply non-homogeneous material, consisting of two phases: silicon and silicon oxide.

Extraction of biomedical traits for patients with Amyotrophic Lateral Sclerosis using parallel and hierarchical classifiers

This paper concerns the analysis and interpretation of results obtained from biomedical experiments. The biomedical approach allows differentiating patients by testing their EPO concentration in serum and cerebrospinal fluids. The diagnostic tests can be applied in patient monitoring, or future therapeutic research. The paper presents the possibility of recognition of the clinical status of patients with Amyotrophic Lateral Sclerosis (ALS) by the use of parallel and hierarchical classifiers. The proposed method can be used in biometric identification of ALS patients. The results demonstrate that the pattern recognition methods may be helpful in evaluation of the clinical progress of ALS.

Recognition of the Ventilatory Response to the Intermittent Chemical Stimuli in Awake Animals

In this study we examined whether it could be possible to recognize a type of chemical stimuli, given in intermittent cycles, on the basis of observed changes in the breathing pattern in an animal model. Ventilatory responses to three chemical stimulus - normoxic cycles (3-min administration of stimulus/8-min normoxic recovery) in awake rats were investigated. Two types of chemical stimuli were given: (a) gas mixtures of 14% or 9% O2 in N2 (i.e. hypoxia), and (b) 5% or 10% CO2 in O2 (i.e. hypercapnia), each one in a separate run of the intermittent stimulus. Ventilatory features: respiratory frequency, tidal volume, minute ventilation, inspired and expired times, were used for recognition of ventilatory responses to exposures of the intermittent stimuli. The quality of recognition was evaluated by a probability of misclassification that was estimated experimentally. As a classifier we used the k nearest neighbor (k-NN) rule that is one of most powerful method offered by the pattern recognition theory. Satisfactory recognition was obtained for recovery periods and stronger stimuli. The best recognition was observed for the intermittent hypercapnia. In conclusion, the approach based on k-NN rule has appeared to be useful tool for recognition changes of ventilatory responses to exposures of the intermittent chemical stimuli.

Recognition of Cycles of Repeated Hypoxia on the Basis of Time Periods in Biological Model

Recognition possibility of expositions (cycles) on the basis the respiratory response to repetitive hypoxia in experimental model (named as the intermittent hypoxia model) was the aim of the study. Variables of a breathing pattern: respiratory frequency, tidal volume, duration of inspiration and expiration (F, V T, TI, TE, respectively), and phrenic nerve activity (A phr) were measured in one-minute periods in biological experiments. The data set was analyzed using k-nearest neighbor decision rule (k-NN). It was found that the variables, measured after sequential hypoxic exposures for the recovery phases, allow recognizing cycles with 6.5% of misclassification rate. However, the recognition based on two selected variables F, V T offered significantly lower error rate of 2%. The obtained results have shown the usefulness of pattern recognition approach to diagnosis of the changes that reflect the respiratory plasticity developing on short-term exposures of the intermittent hypoxia, such as that simulated in the biological model.