ART-EMAP: A Neural Network Architecture for Learning and Prediction by Evidence Accumulation

This paper introduces ART-EMAP, a neural architecture that uses spatial and temporal evidence accumulation to extend the capabilities of fuzzy ARTMAP. ART-EMAP combines supervised and unsupervised learning and a medium-term memory process to accomplish stable pattern category recognition in a noisy input environment. The ART-EMAP system features (i) distributed pattern registration at a view category field; (ii) a decision criterion for mapping between view and object categories which can delay categorization of ambiguous objects and trigger an evidence accumulation process when faced with a low confidence prediction; (iii) a process that accumulates evidence at a medium-term memory (MTM) field; and (iv) an unsupervised learning algorithm to fine-tune performance after a limited initial period of supervised network training. ART-EMAP dynamics are illustrated with a benchmark simulation example. Applications include 3-D object recognition from a series of ambiguous 2-D views.

A Neural Network Architecture for Autonomous Learning, Recognition, and Prediction in a Nonstationary World

In a constantly changing world, humans are adapted to alternate routinely between attending to familiar objects and testing hypotheses about novel ones. We can rapidly learn to recognize and narne novel objects without unselectively disrupting our memories of familiar ones. We can notice fine details that differentiate nearly identical objects and generalize across broad classes of dissimilar objects. This chapter describes a class of self-organizing neural network architectures--called ARTMAP-- that are capable of fast, yet stable, on-line recognition learning, hypothesis testing, and naming in response to an arbitrary stream of input patterns (Carpenter, Grossberg, Markuzon, Reynolds, and Rosen, 1992; Carpenter, Grossberg, and Reynolds, 1991). The intrinsic stability of ARTMAP allows the system to learn incrementally for an unlimited period of time. System stability properties can be traced to the structure of its learned memories, which encode clusters of attended features into its recognition categories, rather than slow averages of category inputs. The level of detail in the learned attentional focus is determined moment-by-moment, depending on predictive success: an error due to over-generalization automatically focuses attention on additional input details enough of which are learned in a new recognition category so that the predictive error will not be repeated. An ARTMAP system creates an evolving map between a variable number of learned categories that compress one feature space (e.g., visual features) to learned categories of another feature space (e.g., auditory features). Input vectors can be either binary or analog. Computational properties of the networks enable them to perform significantly better in benchmark studies than alternative machine learning, genetic algorithm, or neural network models. Some of the critical problems that challenge and constrain any such autonomous learning system will next be illustrated. Design principles that work together to solve these problems are then outlined. These principles are realized in the ARTMAP architecture, which is specified as an algorithm. Finally, ARTMAP dynamics are illustrated by means of a series of benchmark simulations.

Prediction and prevention of venous thrombosis in pregnancy

Venous thromboembolism (VTE) remains the leading cause of maternal mortality. Reports identified further research is required in obese and women post caesarean section (CS). Risk factors for VTE during pregnancy are periodically absent indicating the need for a simple and effective screening tool for pregnancy. Perturbation of the uteroplacental haemostasis has been implicated in placenta mediated pregnancy complications. This thesis had 4 main aims: 1) To investigate anticoagulant effects following a fixed thromboprophylaxis dose in healthy women post elective CS. 2) To evaluate the calibrated automated thrombogram (CAT) assay as a potential predictive tool for thrombosis in pregnancy. 3) To compare the anticoagulant effects of fixed versus weight adjusted thromboprophylaxis dose in morbidly obese pregnant women. 4) To investigate the LMWH effects on human haemostatic gene and antigen expression in placentae and plasma from the uteroplacental , maternal and fetal circulation. Tissue factor pathway inhibitor (TFPI), thrombin antithrombin (TAT), CAT and anti-Xa levels were analysed. Real-time PCR and ELISA were used to quantify mRNA and protein expression of TFPI and TF in placental tissue. In women post CS, anti-Xa levels do not reflect the full anticoagulant effects of LMWH. LMWH thromboprophylaxis in this healthy cohort of patients appears to have a sustained effect in reducing excess thrombin production post elective CS. The results of this study suggest that predicting VTE in pregnant women using CAT assay is not possible at present time. The prothrombotic state in pregnant morbidly obese women was substantially attenuated by weight adjusted but not at fixed LMWH doses. LMWH may be effective in reducing in- vivo thrombin production in the uteroplacental circulation of thrombophilic women. All these results collectively suggest that at appropriate dosage, LMWH is effective in attenuating excess thrombin generation, in low risk pregnant women post caesarean section or moderate to high risk pregnant women who are morbidly obese or tested positive for thrombophilia. The results of the studies provided data to inform evidence-based practice to improve the outcome for pregnant women at risk of thrombosis.

Prediction in dynamic models with time-dependent conditional variances

This paper considers forecasting the conditional mean and variance from a single-equation dynamic model with autocorrelated disturbances following an ARMA process, and innovations with time-dependent conditional heteroskedasticity as represented by a linear GARCH process. Expressions for the minimum MSE predictor and the conditional MSE are presented. We also derive the formula for all the theoretical moments of the prediction error distribution from a general dynamic model with GARCH(1, 1) innovations. These results are then used in the construction of ex ante prediction confidence intervals by means of the Cornish-Fisher asymptotic expansion. An empirical example relating to the uncertainty of the expected depreciation of foreign exchange rates illustrates the usefulness of the results. © 1992.

An integrated approach to the prediction of chemotherapeutic response in patients with breast cancer.

BACKGROUND: A major challenge in oncology is the selection of the most effective chemotherapeutic agents for individual patients, while the administration of ineffective chemotherapy increases mortality and decreases quality of life in cancer patients. This emphasizes the need to evaluate every patient's probability of responding to each chemotherapeutic agent and limiting the agents used to those most likely to be effective. METHODS AND RESULTS: Using gene expression data on the NCI-60 and corresponding drug sensitivity, mRNA and microRNA profiles were developed representing sensitivity to individual chemotherapeutic agents. The mRNA signatures were tested in an independent cohort of 133 breast cancer patients treated with the TFAC (paclitaxel, 5-fluorouracil, adriamycin, and cyclophosphamide) chemotherapy regimen. To further dissect the biology of resistance, we applied signatures of oncogenic pathway activation and performed hierarchical clustering. We then used mRNA signatures of chemotherapy sensitivity to identify alternative therapeutics for patients resistant to TFAC. Profiles from mRNA and microRNA expression data represent distinct biologic mechanisms of resistance to common cytotoxic agents. The individual mRNA signatures were validated in an independent dataset of breast tumors (P = 0.002, NPV = 82%). When the accuracy of the signatures was analyzed based on molecular variables, the predictive ability was found to be greater in basal-like than non basal-like patients (P = 0.03 and P = 0.06). Samples from patients with co-activated Myc and E2F represented the cohort with the lowest percentage (8%) of responders. Using mRNA signatures of sensitivity to other cytotoxic agents, we predict that TFAC non-responders are more likely to be sensitive to docetaxel (P = 0.04), representing a viable alternative therapy. CONCLUSIONS: Our results suggest that the optimal strategy for chemotherapy sensitivity prediction integrates molecular variables such as ER and HER2 status with corresponding microRNA and mRNA expression profiles. Importantly, we also present evidence to support the concept that analysis of molecular variables can present a rational strategy to identifying alternative therapeutic opportunities.

Impacts of land use change and climate variations on annual inflow into Miyun Reservoir,Beijing, China

The Miyun Reservoir, the only surface water source for Beijing city, has experienced water supply decline in recent decades. Previous studies suggest that both land use change and climate contribute to the changes of water supply in this critical watershed. However, the specific causes of the decline in the Miyun Reservoir are debatable under a non-stationary climate in the past 4 decades. The central objective of this study was to quantify the separate and collective contributions of land use change and climate variability to the decreasing inflow into the Miyun Reservoir during 1961–2008. Different from previous studies on this watershed, we used a comprehensive approach to quantify the timing of changes in hydrology and associated environmental variables using the long-term historical hydrometeorology and remote-sensing-based land use records. To effectively quantify the different impacts of the climate variation and land use change on streamflow during different sub-periods, an annual water balance model (AWB), the climate elasticity model (CEM), and a rainfall–runoff model (RRM) were employed to conduct attribution analysis synthetically. We found a significant (p  <  0.01) decrease in annual streamflow, a significant positive trend in annual potential evapotranspiration (p  <  0.01), and an insignificant (p  >  0.1) negative trend in annual precipitation during 1961–2008. We identified two streamflow breakpoints, 1983 and 1999, by the sequential Mann–Kendall test and double-mass curve. Climate variability alone did not explain the decrease in inflow to the Miyun Reservoir. Reduction of water yield was closely related to increase in actual evapotranspiration due to the expansion of forestland and reduction in cropland and grassland, and was likely exacerbated by increased water consumption for domestic and industrial uses in the basin. The contribution to the observed streamflow decline from land use change fell from 64–92 % during 1984–1999 to 36–58 % during 2000–2008, whereas the contribution from climate variation climbed from 8–36 % during the 1984–1999 to 42–64 % during 2000–2008. Model uncertainty analysis further demonstrated that climate warming played a dominant role in streamflow reduction in the most recent decade (i.e., 2000s). We conclude that future climate change and variability will further challenge the water supply capacity of the Miyun Reservoir to meet water demand. A comprehensive watershed management strategy needs to consider the climate variations besides vegetation management in the study basin.

The prediction of fire propagation in enclosure fires

In this paper we present some early work concerned with the development of a simple solid fuel combustion model incorporated within a Computational Fluid Dynamics (CFD) framework. The model is intended for use in engineering applications of fire field modelling and represents an extension of this technique to situations involving the combustion of solid cellulosic hels A simple solid &el combustion model consisting of a thermal pyrolysis model, a six flux radiation model and an eddydissipation model for gaseous combustion have been developed and implemented within the CFD code CFDS-FLOW3D The model is briefly described and demonstrated through two applications involving fire spread in a compartment with a plywood lined ceiling. The two scenarios considered involve a fire in an open and closed compartment The model is shown to be able to qualitatively predict behaviours similar to flashover - in the case of the open room - and backdrafl - in the case of the initially closed room.

The FUELCON meta-architecture, II: alternatives for parameter prediction

We continue the discussion of the decision points in the FUELCON metaarchitecture. Having discussed the relation of the original expert system to its sequel projects in terms of an AND/OR tree, we consider one further domain for a neural component: parameter prediction downstream of the core reload candidate pattern generator, thus, a replacement for the NOXER simulator currently in use in the project.

The numerical simulation of noncharring thermal degradation and its application to the prediction of compartment fire development

In this paper we present some work concerned with the development and testing of a simple solid fuel combustion model incorporated within a Computational Fluid Dynamics (CFD) framework. The model is intended for use in engineering applications of fire field modeling and represents an extension of this technique to situations involving the combustion of solid fuels. The CFD model is coupled with a simple thermal pyrolysis model for combustible solid noncharring fuels, a six-flux radiation model and an eddy-dissipation model for gaseous combustion. The model is then used to simulate a series of small-scale room fire experiments in which the target solid fuel is polymethylmethacrylate. The numerical predictions produced by this coupled model are found to be in very good agreement with experimental data. Furthermore, numerical predictions of the relationship between the air entrained into the fire compartment and the ventilation factor produce a characteristic linear correlation with constant of proportionality 0.38 kg/sm5/12. The simulation results also suggest that the model is capable of predicting the onset of "flashover" type behavior within the fire compartment.

A computational model for defect prediction in shape castings based on the interaction of free surface flow, heat transfer, and solidification phenomena

High-integrity castings require sophisticated design and manufacturing procedures to ensure they are essentially macrodefect free. Unfortunately, an important class of such defects—macroporosity, misruns, and pipe shrinkage—are all functions of the interactions of free surface flow, heat transfer, and solidication in complex geometries. Because these defects arise as an interaction of the preceding continuum phenomena, genuinely predictive models of these defects must represent these interactions explicitly. This work describes an attempt to model the formation of macrodefects explicitly as a function of the interacting continuum phenomena in arbitrarily complex three-dimensional geometries. The computational approach exploits a compatible set of finite volume procedures extended to unstructured meshes. The implementation of the model is described together with its testing and a measure of validation. The model demonstrates the potential to predict reliably shrinkage macroporosity, misruns, and pipe shrinkage directly as a result of interactions among free-surface fluid flow, heat transfer, and solidification.

Computational model for prediction of particle degradation during dilute phase pneumatic conveying: the use of a laboratory scale degradation tester for the determination of degradation propensity

The overall objective of this work is to develop a computational model of particle degradation during dilute-phasepneumatic conveying. A key feature of such a model is the prediction of particle breakage due to particle–wall collisions in pipeline bends. This paper presents a method for calculating particle impact degradation propensity under a range of particle velocities and particle sizes. It is based on interpolation on impact data obtained in a new laboratory-scale degradation tester. The method is tested and validated against experimental results for degradation at 90± impact angle of a full-size distribution sample of granulated sugar. In a subsequent work, the calculation of degradation propensity is coupled with a ow model of the solids and gas phases in the pipeline.

Computational model for prediction of particle degradation during dilute-phase pneumatic conveying: modeling of dilute-phase pneumatic conveying

A complete model of particle impact degradation during dilute-phase pneumatic conveying is developed, which combines a degradation model, based on the experimental determination of breakage matrices, and a physical model of solids and gas flow in the pipeline. The solids flow in a straight pipe element is represented by a model consisting of two zones: a strand-type flow zone immediately downstream of a bend, followed by a fully suspended flow region after dispersion of the strand. The breakage matrices constructed from data on 90° angle single-impact tests are shown to give a good representation of the degradation occurring in a pipe bend of 90° angle. Numerical results are presented for degradation of granulated sugar in a large scale pneumatic conveyor.

Solder life prediction in a thermal analysis software environment

This paper describes recent developments made to the stress analysis module within FLOTHERM, extending its capability to handle viscoplastic behavior. It also presents the validation of this approach and results obtained for an SMT resistor as an illustrative example. Lifetime predictions are made using the creep strain energy based models of Darveaux. Comment is made about the applicability of the damage model to the geometry of the joint under study.