A Bayesian neural network algorithm identifies patients in whom post-EVAR surveillance may be unnecessary

Lifelong surveillance is not cost-effective after endovascular aneurysm repair (EVAR), but is required to detect aortic complications which are fatal if untreated (type 1/3 endoleak, sac expansion, device migration). Aneurysm morphology determines the probability of aortic complications and therefore the need for surveillance, but existing analyses have proven incapable of identifying patients at sufficiently low risk to justify abandoning surveillance. This study aimed to improve the prediction of aortic complications, through the application of machine-learning techniques. Patients undergoing EVAR at 2 centres were studied from 2004–2010. Aneurysm morphology had previously been studied to derive the SGVI Score for predicting aortic complications. Bayesian Neural Networks were designed using the same data, to dichotomise patients into groups at low- or high-risk of aortic complications. Network training was performed only on patients treated at centre 1. External validation was performed by assessing network performance independently of network training, on patients treated at centre 2. Discrimination was assessed by Kaplan-Meier analysis to compare aortic complications in predicted low-risk versus predicted high-risk patients. 761 patients aged 75 +/− 7 years underwent EVAR in 2 centres. Mean follow-up was 36+/− 20 months. Neural networks were created incorporating neck angu- lation/length/diameter/volume; AAA diameter/area/volume/length/tortuosity; and common iliac tortuosity/diameter. A 19-feature network predicted aor- tic complications with excellent discrimination and external validation (5-year freedom from aortic complications in predicted low-risk vs predicted high-risk patients: 97.9% vs. 63%; p < 0.0001). A Bayesian Neural-Network algorithm can identify patients in whom it may be safe to abandon surveillance after EVAR. This proposal requires prospective study.

Pavement performance prediction using adaptive logic networks

Pavement performance is one of the most important components of the pavement management system. Prediction of the future performance of a pavement section is important in programming maintenance and rehabilitation needs. Models for predicting pavement performance have been developed on the basis of traffic and age. The purpose of this research is to extend the use of a relatively new approach to performance prediction in pavement performance modeling using adaptive logic networks (ALN). Adaptive logic networks have recently emerged as an effective alternative to artificial neural networks for machine learning tasks. ^ The ALN predictive methodology is applicable to a wide variety of contexts including prediction of roughness based indices, composite rating indices and/or individual pavement distresses. The ALN program requires key information about a pavement section, including the current distress indexes, pavement age, climate region, traffic and other variables to predict yearly performance values into the future. ^ This research investigates the effect of different learning rates of the ALN in pavement performance modeling. It can be used at both the network and project level for predicting the long term performance of a road network. Results indicate that the ALN approach is well suited for pavement performance prediction modeling and shows a significant improvement over the results obtained from other artificial intelligence approaches. ^

A novel neural network based system for assessing risks associated with information technology security breaches

Security remains a top priority for organizations as their information systems continue to be plagued by security breaches. This dissertation developed a unique approach to assess the security risks associated with information systems based on dynamic neural network architecture. The risks that are considered encompass the production computing environment and the client machine environment. The risks are established as metrics that define how susceptible each of the computing environments is to security breaches. ^ The merit of the approach developed in this dissertation is based on the design and implementation of Artificial Neural Networks to assess the risks in the computing and client machine environments. The datasets that were utilized in the implementation and validation of the model were obtained from business organizations using a web survey tool hosted by Microsoft. This site was designed as a host site for anonymous surveys that were devised specifically as part of this dissertation. Microsoft customers can login to the website and submit their responses to the questionnaire. ^ This work asserted that security in information systems is not dependent exclusively on technology but rather on the triumvirate people, process and technology. The questionnaire and consequently the developed neural network architecture accounted for all three key factors that impact information systems security. ^ As part of the study, a methodology on how to develop, train and validate such a predictive model was devised and successfully deployed. This methodology prescribed how to determine the optimal topology, activation function, and associated parameters for this security based scenario. The assessment of the effects of security breaches to the information systems has traditionally been post-mortem whereas this dissertation provided a predictive solution where organizations can determine how susceptible their environments are to security breaches in a proactive way. ^

Multidimensional classification and diagnosis of leukemia in flow cytometry

This research is to establish new optimization methods for pattern recognition and classification of different white blood cells in actual patient data to enhance the process of diagnosis. Beckman-Coulter Corporation supplied flow cytometry data of numerous patients that are used as training sets to exploit the different physiological characteristics of the different samples provided. The methods of Support Vector Machines (SVM) and Artificial Neural Networks (ANN) were used as promising pattern classification techniques to identify different white blood cell samples and provide information to medical doctors in the form of diagnostic references for the specific disease states, leukemia. The obtained results prove that when a neural network classifier is well configured and trained with cross-validation, it can perform better than support vector classifiers alone for this type of data. Furthermore, a new unsupervised learning algorithm---Density based Adaptive Window Clustering algorithm (DAWC) was designed to process large volumes of data for finding location of high data cluster in real-time. It reduces the computational load to ∼O(N) number of computations, and thus making the algorithm more attractive and faster than current hierarchical algorithms.

A static traffic assignment model combined with an artificial neural network delay model

As traffic congestion continues to worsen in large urban areas, solutions are urgently sought. However, transportation planning models, which estimate traffic volumes on transportation network links, are often unable to realistically consider travel time delays at intersections. Introducing signal controls in models often result in significant and unstable changes in network attributes, which, in turn, leads to instability of models. Ignoring the effect of delays at intersections makes the model output inaccurate and unable to predict travel time. To represent traffic conditions in a network more accurately, planning models should be capable of arriving at a network solution based on travel costs that are consistent with the intersection delays due to signal controls. This research attempts to achieve this goal by optimizing signal controls and estimating intersection delays accordingly, which are then used in traffic assignment. Simultaneous optimization of traffic routing and signal controls has not been accomplished in real-world applications of traffic assignment. To this end, a delay model dealing with five major types of intersections has been developed using artificial neural networks (ANNs). An ANN architecture consists of interconnecting artificial neurons. The architecture may either be used to gain an understanding of biological neural networks, or for solving artificial intelligence problems without necessarily creating a model of a real biological system. The ANN delay model has been trained using extensive simulations based on TRANSYT-7F signal optimizations. The delay estimates by the ANN delay model have percentage root-mean-squared errors (%RMSE) that are less than 25.6%, which is satisfactory for planning purposes. Larger prediction errors are typically associated with severely oversaturated conditions. A combined system has also been developed that includes the artificial neural network (ANN) delay estimating model and a user-equilibrium (UE) traffic assignment model. The combined system employs the Frank-Wolfe method to achieve a convergent solution. Because the ANN delay model provides no derivatives of the delay function, a Mesh Adaptive Direct Search (MADS) method is applied to assist in and expedite the iterative process of the Frank-Wolfe method. The performance of the combined system confirms that the convergence of the solution is achieved, although the global optimum may not be guaranteed.

Neural network based off-line handwritten text recognition system

This dissertation introduces a new system for handwritten text recognition based on an improved neural network design. Most of the existing neural networks treat mean square error function as the standard error function. The system as proposed in this dissertation utilizes the mean quartic error function, where the third and fourth derivatives are non-zero. Consequently, many improvements on the training methods were achieved. The training results are carefully assessed before and after the update. To evaluate the performance of a training system, there are three essential factors to be considered, and they are from high to low importance priority: (1) error rate on testing set, (2) processing time needed to recognize a segmented character and (3) the total training time and subsequently the total testing time. It is observed that bounded training methods accelerate the training process, while semi-third order training methods, next-minimal training methods, and preprocessing operations reduce the error rate on the testing set. Empirical observations suggest that two combinations of training methods are needed for different case character recognition. Since character segmentation is required for word and sentence recognition, this dissertation provides also an effective rule-based segmentation method, which is different from the conventional adaptive segmentation methods. Dictionary-based correction is utilized to correct mistakes resulting from the recognition and segmentation phases. The integration of the segmentation methods with the handwritten character recognition algorithm yielded an accuracy of 92% for lower case characters and 97% for upper case characters. In the testing phase, the database consists of 20,000 handwritten characters, with 10,000 for each case. The testing phase on the recognition 10,000 handwritten characters required 8.5 seconds in processing time.

Development of New Structural Health Monitoring Techniques

During the past two decades, many researchers have developed methods for the detection of structural defects at the early stages to operate the aerospace vehicles safely and to reduce the operating costs. The Surface Response to Excitation (SuRE) method is one of these approaches developed at FIU to reduce the cost and size of the equipment. The SuRE method excites the surface at a series of frequencies and monitors the propagation characteristics of the generated waves. The amplitude of the waves reaching to any point on the surface varies with frequency; however, it remains consistent as long as the integrity and strain distribution on the part is consistent. These spectral characteristics change when cracks develop or the strain distribution changes. The SHM methods may be used for many applications, from the detection of loose screws to the monitoring of manufacturing operations. A scanning laser vibrometer was used in this study to investigate the characteristics of the spectral changes at different points on the parts. The study started with detecting a load on a plate and estimating its location. The modifications on the part with manufacturing operations were detected and the Part-Based Manufacturing Process Performance Monitoring (PbPPM) method was developed. Hardware was prepared to demonstrate the feasibility of the proposed methods in real time. Using low-cost piezoelectric elements and the non-contact scanning laser vibrometer successfully, the data was collected for the SuRE and PbPPM methods. Locational force, loose bolts and material loss could be easily detected by comparing the spectral characteristics of the arriving waves. On-line methods used fast computational methods for estimating the spectrum and detecting the changing operational conditions from sum of the squares of the variations. Neural networks classified the spectrums when the desktop – DSP combination was used. The results demonstrated the feasibility of the SuRE and PbPPM methods.

Development of New Structural Health Monitoring Techniques

During the past two decades, many researchers have developed methods for the detection of structural defects at the early stages to operate the aerospace vehicles safely and to reduce the operating costs. The Surface Response to Excitation (SuRE) method is one of these approaches developed at FIU to reduce the cost and size of the equipment. The SuRE method excites the surface at a series of frequencies and monitors the propagation characteristics of the generated waves. The amplitude of the waves reaching to any point on the surface varies with frequency; however, it remains consistent as long as the integrity and strain distribution on the part is consistent. These spectral characteristics change when cracks develop or the strain distribution changes. The SHM methods may be used for many applications, from the detection of loose screws to the monitoring of manufacturing operations. A scanning laser vibrometer was used in this study to investigate the characteristics of the spectral changes at different points on the parts. The study started with detecting a load on a plate and estimating its location. The modifications on the part with manufacturing operations were detected and the Part-Based Manufacturing Process Performance Monitoring (PbPPM) method was developed. Hardware was prepared to demonstrate the feasibility of the proposed methods in real time. Using low-cost piezoelectric elements and the non-contact scanning laser vibrometer successfully, the data was collected for the SuRE and PbPPM methods. Locational force, loose bolts and material loss could be easily detected by comparing the spectral characteristics of the arriving waves. On-line methods used fast computational methods for estimating the spectrum and detecting the changing operational conditions from sum of the squares of the variations. Neural networks classified the spectrums when the desktop – DSP combination was used. The results demonstrated the feasibility of the SuRE and PbPPM methods.

Flexible pavement layer moduli determination : an adaptive artificial neural network approach

The estimation of pavement layer moduli through the use of an artificial neural network is a new concept which provides a less strenuous strategy for backcalculation procedures. Artificial Neural Networks are biologically inspired models of the human nervous system. They are specifically designed to carry out a mapping characteristic. This study demonstrates how an artificial neural network uses non-destructive pavement test data in determining flexible pavement layer moduli. The input parameters include plate loadings, corresponding sensor deflections, temperature of pavement surface, pavement layer thicknesses and independently deduced pavement layer moduli.

A Static Traffic Assignment Model Combined with an Artificial Neural Network Delay Model

As traffic congestion continues to worsen in large urban areas, solutions are urgently sought. However, transportation planning models, which estimate traffic volumes on transportation network links, are often unable to realistically consider travel time delays at intersections. Introducing signal controls in models often result in significant and unstable changes in network attributes, which, in turn, leads to instability of models. Ignoring the effect of delays at intersections makes the model output inaccurate and unable to predict travel time. To represent traffic conditions in a network more accurately, planning models should be capable of arriving at a network solution based on travel costs that are consistent with the intersection delays due to signal controls. This research attempts to achieve this goal by optimizing signal controls and estimating intersection delays accordingly, which are then used in traffic assignment. Simultaneous optimization of traffic routing and signal controls has not been accomplished in real-world applications of traffic assignment. To this end, a delay model dealing with five major types of intersections has been developed using artificial neural networks (ANNs). An ANN architecture consists of interconnecting artificial neurons. The architecture may either be used to gain an understanding of biological neural networks, or for solving artificial intelligence problems without necessarily creating a model of a real biological system. The ANN delay model has been trained using extensive simulations based on TRANSYT-7F signal optimizations. The delay estimates by the ANN delay model have percentage root-mean-squared errors (%RMSE) that are less than 25.6%, which is satisfactory for planning purposes. Larger prediction errors are typically associated with severely oversaturated conditions. A combined system has also been developed that includes the artificial neural network (ANN) delay estimating model and a user-equilibrium (UE) traffic assignment model. The combined system employs the Frank-Wolfe method to achieve a convergent solution. Because the ANN delay model provides no derivatives of the delay function, a Mesh Adaptive Direct Search (MADS) method is applied to assist in and expedite the iterative process of the Frank-Wolfe method. The performance of the combined system confirms that the convergence of the solution is achieved, although the global optimum may not be guaranteed.

The Neural Basis of Involuntary Episodic Memories

Involuntary episodic memories are memories that come into consciousness without preceding retrieval effort. These memories are commonplace and are relevant to multiple mental disorders. However, they are vastly understudied. We use a novel paradigm to elicit involuntary memories in the laboratory so that we can study their neural basis. In session one, an encoding session, sounds are presented with picture pairs or alone. In session two, in the scanner, sounds-picture pairs and unpaired sounds are reencoded. Immediately following, participants are split into two groups: a voluntary and an involuntary group. Both groups perform a sound localization task in which they hear the sounds and indicate the side from which they are coming. The voluntary group additionally tries to remember the pictures that were paired with the sounds. Looking at neural activity, we find a main effect of condition (paired vs. unpaired sounds) showing similar activity in both groups for voluntary and involuntary memories in regions typically associated with retrieval. There is also a main effect of group (voluntary vs. involuntary) in the dorsolateral prefrontal cortex, a region typically associated with cognitive control. Turning to connectivity similarities and differences between groups again, there is a main effect of condition showing paired > unpaired sounds are associated with a recollection network. In addition, three group differences were found: (1) increased connectivity between the pulvinar nucleus of the thalamus and the recollection network for the voluntary group, (2) a higher association between the voluntary group and a network that includes regions typically found in frontoparietal and cingulo-opercular networks, and (3) shorter path length for about half of the nodes in these networks for the voluntary group. Finally, we use the same paradigm to compare involuntary memories in people with posttraumatic stress disorder (PTSD) to trauma-controls. This study also included the addition of emotional pictures. There were two main findings. (1) A similar pattern of activity was found for paired > unpaired sounds for both groups but this activity was delayed in the PTSD group. (2) A similar pattern of activity was found for high > low emotion stimuli but it occurred early in the PTSD group compared to the control group. Our results suggest that involuntary and voluntary memories share the same neural representation but that voluntary memories are associated with additional cognitive control processes. They also suggest that disorders associated with cognitive deficits, like PTSD, can affect the processing of involuntary memories.

(Table 1) Age control and temporal resolution of paleoceanographic records of the North Atlantic

Short-term changes in sea surface conditions controlling the thermohaline circulation in the northern North Atlantic are expected to be especially efficient in perturbing global climate stability. Here we assess past variability of sea surface temperature (SST) in the northeast Atlantic and Norwegian Sea during Marine Isotope Stage (MIS) 2 and, in particular, during the Last Glacial Maximum (LGM). Five high-resolution SST records were established on a meridional transect (53°N-72°N) to trace centennial-scale oscillations in SST and sea-ice cover. We used three independent computational techniques (SIMMAX modern analogue technique, Artificial Neural Networks (ANN), and Revised Analog Method (RAM)) to reconstruct SST from planktonic foraminifer census counts. SIMMAX and ANN reproduced short-term SST oscillations of similar magnitude and absolute levels, while RAM, owing to a restrictive analog selection, appears less suitable for reconstructing "cold end" SST. The SIMMAX and ANN SST reconstructions support the existence of a weak paleo-Norwegian Current during Dansgaard-Oeschger (DO) interstadials number 4, 3, 2, and 1. During the LGM, two warm incursions of 7°C water to occurred in the northern North Atlantic but ended north of the Iceland Faroe Ridge. A rough numerical estimate shows that the near-surface poleward heat transfer from 53° across the Iceland-Faroe Ridge up to to 72° N dropped to less than 60% of the modern value during DO interstadials and to almost zero during DO stadials. Summer sea ice was generally confined to the area north of 70°N and only rarely expanded southward along the margins of continental ice sheets. Internal LGM variability of North Atlantic (>40°N) SST in the GLAMAP 2000 compilation (Sarnthein et al., 2003, doi:10.1029/2002PA000771; Pflaumann et al., 2003, doi:10.1029/2002PA000774) indicates maximum instability in the glacial subpolar gyre and at the Iberian Margin, while in the Nordic Seas, SST was continuously low.

Spiking neuron models of the medial and lateral superior olive for sound localisation

Sound localisation is defined as the ability to identify the position of a sound source. The brain employs two cues to achieve this functionality for the horizontal plane, interaural time difference (ITD) by means of neurons in the medial superior olive (MSO) and interaural intensity difference (IID) by neurons of the lateral superior olive (LSO), both located in the superior olivary complex of the auditory pathway. This paper presents spiking neuron architectures of the MSO and LSO. An implementation of the Jeffress model using spiking neurons is presented as a representation of the MSO, while a spiking neuron architecture showing how neurons of the medial nucleus of the trapezoid body interact with LSO neurons to determine the azimuthal angle is discussed. Experimental results to support this work are presented.

Leveraging Multiple Features for Image Retrieval and Matching

The goal of image retrieval and matching is to find and locate object instances in images from a large-scale image database. While visual features are abundant, how to combine them to improve performance by individual features remains a challenging task. In this work, we focus on leveraging multiple features for accurate and efficient image retrieval and matching. We first propose two graph-based approaches to rerank initially retrieved images for generic image retrieval. In the graph, vertices are images while edges are similarities between image pairs. Our first approach employs a mixture Markov model based on a random walk model on multiple graphs to fuse graphs. We introduce a probabilistic model to compute the importance of each feature for graph fusion under a naive Bayesian formulation, which requires statistics of similarities from a manually labeled dataset containing irrelevant images. To reduce human labeling, we further propose a fully unsupervised reranking algorithm based on a submodular objective function that can be efficiently optimized by greedy algorithm. By maximizing an information gain term over the graph, our submodular function favors a subset of database images that are similar to query images and resemble each other. The function also exploits the rank relationships of images from multiple ranked lists obtained by different features. We then study a more well-defined application, person re-identification, where the database contains labeled images of human bodies captured by multiple cameras. Re-identifications from multiple cameras are regarded as related tasks to exploit shared information. We apply a novel multi-task learning algorithm using both low level features and attributes. A low rank attribute embedding is joint learned within the multi-task learning formulation to embed original binary attributes to a continuous attribute space, where incorrect and incomplete attributes are rectified and recovered. To locate objects in images, we design an object detector based on object proposals and deep convolutional neural networks (CNN) in view of the emergence of deep networks. We improve a Fast RCNN framework and investigate two new strategies to detect objects accurately and efficiently: scale-dependent pooling (SDP) and cascaded rejection classifiers (CRC). The SDP improves detection accuracy by exploiting appropriate convolutional features depending on the scale of input object proposals. The CRC effectively utilizes convolutional features and greatly eliminates negative proposals in a cascaded manner, while maintaining a high recall for true objects. The two strategies together improve the detection accuracy and reduce the computational cost.

Combined Machine Learning Techniques for Decision Making Support in Medicine

Computational intelligent support for decision making is becoming increasingly popular and essential among medical professionals. Also, with the modern medical devices being capable to communicate with ICT, created models can easily find practical translation into software. Machine learning solutions for medicine range from the robust but opaque paradigms of support vector machines and neural networks to the also performant, yet more comprehensible, decision trees and rule-based models. So how can such different techniques be combined such that the professional obtains the whole spectrum of their particular advantages? The presented approaches have been conceived for various medical problems, while permanently bearing in mind the balance between good accuracy and understandable interpretation of the decision in order to truly establish a trustworthy ‘artificial’ second opinion for the medical expert.