Probabilistic Neural Network Analysis Research Articles

Predicting the gas resource potential in reservoir C-sand interval of Lower Goru Formation, Middle Indus Basin, Pakistan

Abstract The integrated study of seismic attributes and inversion analysis can provide a better understanding for predicting the hydrocarbon-bearing zones even in extreme heterogeneous reservoirs. This study aims to delineate and characterize the gas saturated zone within the reservoir (Cretaceous C-sand) interval of Sawan gas field, Middle Indus Basin, Pakistan. The hydrocarbon bearing zone is well identified through the seismic attribute analysis along a sand channel. The sparse-spike inversion analysis has efficiently captured the variations in reservoir parameter (P-impedance) for gas prospect. Inversion results indicated that the relatively lower P-impedance values are encountered along the predicted sand channel. To further characterize the reservoir, geostatistical techniques comprising multiattribute regression and probabilistic neural network (PNN) analysis are applied to predict the effective porosity of reservoir. Comparatively, the PNN analysis predicted the targeted property more efficiently and applied its estimations on entire seismic volume. Furthermore, the geostatistical estimations of PNN analysis significantly predicted the gas-bearing zones and confirmed the sand channel as a major contributor of gas accumulation in the area. These estimates are in appropriate agreement with each other, and the workflow adopted here can be applied to various South Asian regions and in other parts of the world for improved characterization of gas reservoirs.

Anti-Cancer Effect of Melatonin via Downregulation of Delta-like Ligand 4 in Estrogen-Responsive Breast Cancer Cells.

The Notch signaling pathway has a key role in angiogenesis and Delta - Like Ligand 4 (DLL4) is one of the main ligands of Notch involved in cell proliferation in sprouting vessels. In this study, we aimed to evaluate the expression of DLL4 in primary breast tumors and to examine the effect of melatonin on DLL4 expression in vitro. Eighty-five breast tumor and paired adjacent non-tumor tissue samples were collected. Apoptosis assay was performed on breast cancer cells to evaluate melatonin effects. Western blot and quantitative RT-PCR were used to measure DLL4 expression. Then, we investigated the effect of melatonin on the expression of DLL4 in four breast cancer cell lines at RNA and protein levels. We also performed a probabilistic neural network analysis to study genes closely associated with DLL4 expression. Our results showed a significantly higher expression of DLL4 in tumor tissues compared to non-tumor tissues (P = 0.027). Melatonin treatment substantially attenuated DLL4 expression in BT474 and MCF-7 cells, but not in SK-BR-3 and MDA-MB-231 cells. Also, melatonin induced apoptosis in all four cell lines. Network analysis revealed a set of 15 genes that had close association and interaction with DLL4. DLL4 was overexpressed in breast cancer tissues as compared to the non-tumor tissues. It can be concluded that melatonin treatment attenuated DLL4 expression only in estrogen- responsive breast cancer cells and is able to induce apoptosis in breast cancer cells.

Use of maximum likelihood sparse spike inversion and probabilistic neural network for reservoir characterization: a study from F-3 block, the Netherlands

Maximum likelihood sparse spike inversion (MLSSI) method is commonly used in the seismic industry to estimate petrophysical parameters in inter-well region. In present study, maximum likelihood sparse spike inversion technique is applied to the processed 3D post-stack seismic data from the F-3 block, the Netherlands, for estimation of acoustic impedance in the region between the wells. The analysis shows that the impedance varies from 2500 to 6200 m/s/*g/cc in the region which is relatively low and indicates the presence of loose formation in the area. The correlation between synthetic seismic trace and original seismic trace is found to be 0.93 and the synthetic relative error as 0.369, which indicate good performance of the algorithm. The analysis also shows low-impedance anomaly in between 600 and 700 ms time interval which may be due to the presence of sand formation. Thereafter, the probabilistic neural network analysis is performed to predict porosity along with multi-attribute transform analysis to estimate P-wave velocity and porosity in inter-well region. These parameters strengthen the seismic data interpretation which is very crucial step of any exploration and production project. The method is first applied to the composite traces near to well locations, and results are compared with well log data. After getting reasonable results, the whole seismic section is inverted for the P-wave velocity and porosity volume. The analysis shows anomaly in between 600 and 700 ms time interval which corroborates well with the low-impedance zone which may correspond to the reservoir. This is preliminarily interpretation; however to confirm a reservoir, there is need for more petrophysical parameters to be studied.

Qualitative and quantitative comparison of geostatistical techniques of porosity prediction from the seismic and logging data: a case study from the Blackfoot Field, Alberta, Canada

In present study, three recently developed geostatistical methods, single attribute analysis, multi-attribute analysis and probabilistic neural network algorithm have been used to predict porosity in inter well region for Blackfoot field, Alberta, Canada, an offshore oil field. These techniques make use of seismic attributes, generated by model based inversion and colored inversion techniques. The principle objective of the study is to find the suitable combination of seismic inversion and geostatistical techniques to predict porosity and identification of prospective zones in 3D seismic volume. The porosity estimated from these geostatistical approaches is corroborated with the well log porosity. The results suggest that all the three implemented geostatistical methods are efficient and reliable to predict the porosity but the multi-attribute and probabilistic neural network analysis provide more accurate and high resolution porosity sections. A low impedance (6000–8000 m/s g/cc) and high porosity (> 15%) zone is interpreted from inverted impedance and porosity sections respectively between 1060 and 1075 ms time interval and is characterized as reservoir. The qualitative and quantitative results demonstrate that of all the employed geostatistical methods, the probabilistic neural network along with model based inversion is the most efficient method for predicting porosity in inter well region.

A model-based approach for integration analysis of well log and seismic data for reservoir characterization

Accurate reservoir model requires complete information of subsurface properties, specifically porosity and permeability. Reservoir heterogeneity is the fundamental challenge for geoscientists to predict these properties which may affect the reservoir performance and their well productivity. Porosity is one of the key parameters for accumulation of hydrocarbon but its prediction is difficult due to significant variation over a reservoir volume. A spatial distribution of porosity can be investigated by integrating the 3-D seismic and well log attributes which may help in determining such reservoir variation. In addition, nonlinear multivariable regression techniques such as multivariable transform, Genetic algorithm, and Probabilistic Neural Network analysis have also been implemented to achieve high correlation coefficients between well log properties and seismic data. Results from nonlinear regression have better correlations than linear regression. In this study, a 3-D low frequency model (LFM) is proposed which can be estimated by kriging interpolation of resultant impedance values from well log data. Furthermore, “seismic inversion” is adopted for extracting correlated attributes to merge with the LFM so as to better construct a pseudo log volume. A polynomial neural network (PNN*1) is utilized to convert resultant acoustic impedance values into a distinct reservoir property such as porosity. PNN* is trained, tested and validated by using gammaray and resultant acoustic impedance values as input and effective porosity values as a target. The trained PNN* is then applied over the whole reservoir volume to generate a pseudo log volume. In the proposed low frequency model, an attempt has been made to achieve high correlation between the predicted and measured porosity logs. It will improve the reservoir characterization and lead to better estimation of hydrocarbon reserves. This low frequency model achieves better correlation between the predicted and true porosity log even with a minimum number of measured well logs.

Effects of Curative Colorectal Cancer Surgery on Exhaled Volatile Organic Compounds and Potential Implications in Clinical Follow-up.

The aim of this study was to determine whether the volatile organic compounds (VOCs) pattern in colorectal cancer (CRC) patients is modified by curative surgery for a potential application in the oncologic follow-up. CRC has been proved to induce metabolic derangements detectable by high through-output techniques in exhaled breath showing a specific pattern of VOCs. Forty-eight CRC patients and 55 healthy controls (HC) entered the study. Thirty-two patients (M/F: 1.4; mean age 63 years) attended the oncologic follow-up (mean 24 months) and were found disease-free. Breath samples were collected under similar environmental conditions into a Tedlar bags and processed offline by thermal-desorption gas chromatography-mass spectrometry (TD-GC-MS). VOCs were selected by U test to build a Probabilistic Neural Network (PNN) model to set-up a training phase, which was cross-validated using the leave-one out method. A total of 11 VOCs were finally selected for their excellent discriminant performance in identifying disease-free patients in follow-up from CRC patients before surgery, (sensitivity 100%, specificity 97.92%, accuracy 98.75%, and AUC: 1). The same VOCs pattern discriminated follow-up patients from HC, with a sensitivity of 100%, specificity of 90.91%, accuracy of 94.25%, and AUC 0.959. Exhaled VOCs pattern from CRC patients is modified by cancer removal confirming the tight relationship between tumor metabolism and exhaled VOCs. PNN analysis provides a high discriminatory tool to identify patients disease-free after curative surgery suggesting potential implications in CRC screening and secondary prevention.

The Operational Monitor of Air-Conditioners by Using Probabilistic Neural Network

This paper uses the probabilistic neural network (PNN) to monitor the operation statues for the compressor of air-conditioners. The field data including the high/low pressures and the high/low temperatures of refrigerants are measured in a practical system. PNN analyses the refrigerants’ pressures/temperatures of air-conditioners to monitor the operation conditions of compressor. PNN method is suitable for application in a dynamic environment by using new data-set and new hidden without doing any computed iteration. Computer simulations were conducted with refrigerants’ records, test results showed the effectiveness of the proposed system.

Assessment of multivariate classification analysis methods on 3‐dimensional anatomical distribution of functionally identified cat pre‐Bötzinger complex microinjection sites (712.15)

Previous work from our laboratory has demonstrated that multiple patterns of phrenic motor output can be elicited by chemical stimulation of anatomically distinct sub‐regions of the in vivo adult cat pre‐Bötzinger complex. To gain further insight into the anatomical distribution of the microinjection sites yielding these different discharge patterns, we implemented two multivariate classification analysis methods: Fisher Linear Discriminant (FLD) and probabilistic neural network (PNN). For FLD analysis, the multivariate data were transformed into univariate observations by computing the inverse of a pooled variance‐covariance matrix, which was then translated back to the histological coordinates to predict physical sub‐region boundaries. For PNN analysis, a 3‐layered radial basis neural network was created with (1) the input vector for each discharge pattern yielding an output value according to how close the input vector was to each pattern weight vector, (2) the contributions for each class of inputs was summed to produce a net output vector of probabilities; and (3) a transfer function was applied to the outputs to identify the maximum probabilities and produce a classification decision. While each analysis method has its advantages (and disadvantages), both were able to predicte all of the observed phrenic discharge patterns based solely on the 3‐D coordinates of the microinjection sites. We suggest, however, that FLD analysis may be better suited for these data.Grant Funding Source: Supported by HL63175

Exhaled volatile organic compounds identify patients with colorectal cancer

An effective screening tool for colorectal cancer is still lacking. Analysis of the volatile organic compounds (VOCs) linked to cancer is a new frontier in cancer screening, as tumour growth involves several metabolic changes leading to the production of specific compounds that can be detected in exhaled breath. This study investigated whether patients with colorectal cancer have a specific VOC pattern compared with the healthy population. Exhaled breath was collected in an inert bag (Tedlar(®) ) from patients with colorectal cancer and healthy controls (negative at colonoscopy), and processed offline by thermal-desorber gas chromatography-mass spectrometry to evaluate the VOC profile. During the trial phase VOCs of interest were identified and selected, and VOC patterns able to discriminate patients from controls were set up; in the validation phase their discriminant performance was tested on blinded samples. A probabilistic neural network (PNN) validated by the leave-one-out method was used to identify the pattern of VOCs that better discriminated between the two groups. Some 37 patients and 41 controls were included in the trial phase. Application of a PNN to a pattern of 15 compounds showed a discriminant performance with a sensitivity of 86 per cent, a specificity of 83 per cent and an accuracy of 85 per cent (area under the receiver operating characteristic (ROC) curve 0·852). The accuracy of PNN analysis was confirmed in the validation phase on a further 25 subjects; the model correctly assigned 19 patients, giving an overall accuracy of 76 per cent. The pattern of VOCs in patients with colorectal cancer was different from that in healthy controls. The PNN in this study was able to discriminate patients with colorectal cancer with an accuracy of over 75 per cent. Breath VOC analysis appears to have potential clinical application in colorectal cancer screening, although further studies are required to confirm its reliability in heterogeneous clinical settings.

Classification of Geographical Origin by PNN Analysis of Fatty Acid Data and Level of Contaminants in Oils From Peruvian Anchovy

AbstractThe aim of this study was to examine Peruvian anchovy oil fatty acid (FA) compositions, and to test the possibility of using the FA data to classify the oils according to geographical origin along the Peruvian coast. The levels of contaminants in a representative set of samples were determined to examine the general levels and investigate if such measurements could aid in future discrimination between oils. The FA results showed that the two known stocks of Peruvian anchovy displayed different levels of docosahexaenoic acid (DHA, 22:6n‐3) (southern stock; 14.4 ± 0.8% versus central‐northern stock; 9.9 ± 1.2%). However, principal component analysis (PCA) of the FA data indicated clusters according to three regions; North, Center and South. Using a data set of 57 anchovy samples and 21 FA as input, a probabilistic neural network (PNN) was constructed. For the validation data sets, “North” oils was predicted accurately 100% of the time, “Center” oils 100% and “South” oils 83% of the time. The levels of contaminants in the oils determined were low in all but one sample.

Modeling the competitive market efficiency of Egyptian companies: A probabilistic neural network analysis

Understanding efficiency levels is crucial for understanding the competitive structure of a market and/or segments of a market. This study uses two artificial neural networks (NN) and a traditional statistical classification method to classify the relative efficiency of top listed Egyptian companies. Accuracy indices derived from the application of a non-parametric data envelopment analysis approach are used to assess the classification accuracy of the models. Results indicate that the NN models are superior to the traditional statistical methods. The study shows that the NN models have a great potential for the classification of companies’ relative efficiency due to their robustness and flexibility of modeling algorithms. The implications of these results for potential efficiency programs are discussed.

Cerebral metabolite differences in adolescents with low birth weight: assessment with in vivo proton MR spectroscopy

Children with very low birth weight (VLBW) have a significantly increased risk of later neurodevelopmental problems, while infants born small for gestational age (SGA) at term are also at some risk of developing neurological impairment. To investigate possible brain metabolite differences in adolescents with VLBW, SGA at term and controls by proton in vivo magnetic resonance spectroscopy (MRS) at 1.5 T. MR spectra were acquired from volumes localized in the left frontal lobe, containing mainly white matter (54 subjects). Peak areas of N-acetyl aspartate (NAA), choline (Cho) and creatine (Cr) were determined, and the peak area ratio of NAA to Cr, total Cho to Cr, or NAA to Cho calculated. Probabilistic neural network (PNN) analysis was performed utilizing the chemical shift region containing resonances from NAA, Cho and Cr as inputs. No significant difference in the peak area ratios could be found using the Kruskal-Wallis test. By application of PNN, a correct classification of 52 of the 54 adolescents with a sensitivity and specificity exceeding 93% for all groups was achieved. Small, yet systematic, differences in brain metabolite distribution among the groups were confirmed by PNN analysis.

Transformer-fault diagnosis by integrating field data and standard codes with training enhancible adaptive probabilistic network

A transformer-fault diagnosis system (TFDS) using a probabilistic neural network (PNN) and IEC/Cigre standard codes is proposed. Many artificial neural networks (ANNs) have been proposed before, however, the slow repeated iterative process and poor adaptation capability for structural data restrains the ANN applications. An effective and flexible PNN could overcome these drawbacks. In this paper, a PNN analyses the transformer's dissolved gas content to identify faults, while using the gas ratios of the oil and cellulosic decomposition to create training examples. Retraining can be done by adding new examples and new hidden nodes for easy adaptation without doing any computed iteration. The commonly used Excel was integrated to provide a convenient man–machine interface. Computer simulations were conducted with diagnostic gas records, to show the effectiveness of the proposed system.

Metabonomic characterization of genetic variations in toxicological and metabolic responses using probabilistic neural networks.

Current emphasis on efficient screening of novel therapeutic agents in toxicological studies has resulted in the evaluation of novel analytical technologies, including genomic (transcriptomic) and proteomic approaches. We have shown that high-resolution 1H NMR spectroscopy of biofluids and tissues coupled with appropriate chemometric analysis can also provide complementary data for use in in vivo toxicological screening of drugs. Metabonomics concerns the quantitative analysis of the dynamic multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modification [Nicholson, J. K., Lindon, J. C., and Holmes, E. (1999) Xenobiotica 11, 1181-1189]. In this study, we have used 1H NMR spectroscopy to characterize the time-related changes in the urinary metabolite profiles of laboratory rats treated with 13 model toxins and drugs which predominantly target liver or kidney. These 1H NMR spectra were data-reduced and subsequently analyzed using a probabilistic neural network (PNN) approach. The methods encompassed a database of 1310 samples, of which 583 comprised a training set for the neural network, with the remaining 727 (independent cases) employed as a test set for validation. Using these techniques, the 13 classes of toxicity, together with the variations associated with strain, were distinguishable to >90%. Analysis of the 1H NMR spectral data by multilayer perceptron networks and principal components analysis gave a similar but less accurate classification than PNN analysis. This study has highlighted the value of probabilistic neural networks in developing accurate NMR-based metabonomic models for the prediction of xenobiotic-induced toxicity in experimental animals and indicates possible future uses in accelerated drug discovery programs. Furthermore, the sensitivity of this tool to strain differences may prove to be useful in investigating the genetic variation of metabolic responses and for assessing the validity of specific animal models.

Application of probabilistic neural network analysis to a disease with complex inheritance: the GAW11 simulated data.

A probabilistic neural network analysis was applied to the simulated GAW11 data. Six replicates drawn at random from one of the simulated populations were used to generate training and test vectors for pairs of siblings. The vectors incorporated two environmental indicators as well as identical-by-descent allele sharing scores from each of 300 genetic markers. The performance of a 'naïve' probabilistic neural network (PNN) was fair. However, by combining a traditional linkage analysis with a PNN which incorporated gene x environment interaction, the performance was considerably enhanced.