Search

Breast cancer claims 11,400 lives on average every year in the UK, making it one of the deadliest diseases. Mammography is the gold standard for detecting early signs of breast cancer, which can help cure the disease during its early stages. However, incorrect mammography diagnoses are common and may harm patients through unnecessary treatments and operations (or a lack of treatment). Therefore, systems that can learn to detect breast cancer on their own could help reduce the number of incorrect interpretations and missed cases. Various deep learning techniques, which can be used to implement a system that learns how to detect instances of breast cancer in mammograms, are explored throughout this paper. Convolution Neural Networks (CNNs) are used as part of a pipeline based on deep learning techniques. A divide and conquer approach is followed to analyse the effects on performance and efficiency when utilising diverse deep learning techniques such as varying network architectures (VGG19, ResNet50, InceptionV3, DenseNet121, MobileNetV2), class weights, input sizes, image ratios, pre-processing techniques, transfer learning, dropout rates, and types of mammogram projections. This approach serves as a starting point for model development of mammography classification tasks. Practitioners can benefit from this work by using the divide and conquer results to select the most suitable deep learning techniques for their case out-of-the-box, thus reducing the need for extensive exploratory experimentation. Multiple techniques are found to provide accuracy gains relative to a general baseline (VGG19 model using uncropped 512 × 512 pixels input images with a dropout rate of 0.2 and a learning rate of 1 × 10−3) on the Curated Breast Imaging Subset of DDSM (CBIS-DDSM) dataset. These techniques involve transfer learning pre-trained ImagetNet weights to a MobileNetV2 architecture, with pre-trained weights from a binarised version of the mini Mammography Image Analysis Society (mini-MIAS) dataset applied to the fully connected layers of the model, coupled with using weights to alleviate class imbalance, and splitting CBIS-DDSM samples between images of masses and calcifications. Using these techniques, a 5.6% gain in accuracy over the baseline model was accomplished. Other deep learning techniques from the divide and conquer approach, such as larger image sizes, do not yield increased accuracies without the use of image pre-processing techniques such as Gaussian filtering, histogram equalisation and input cropping.

Funding: For all authors this work is supported by the Industrial Centre for AI Research in digital Diagnostics (iCAIRD) which is funded by Innovate UK (https://www.ukri.org/councils/innovate-uk/) on behalf of UK Research and Innovation (UKRI) [project number: 104690], and in part by Chief Scientist Office, Scotland. (https://www.cso.scot.nhs.uk/). In this study we use artificial intelligence (AI) to categorise endometrial biopsy whole slide images (WSI) from digital pathology as either “malignant”, “other or benign” or “insufficient”. An endometrial biopsy is a key step in diagnosis of endometrial cancer, biopsies are viewed and diagnosed by pathologists. Pathology is increasingly digitised, with slides viewed as images on screens rather than through the lens of a microscope. The availability of these images is driving automation via the application of AI. A model that classifies slides in the manner proposed would allow prioritisation of these slides for pathologist review and hence reduce time to diagnosis for patients with cancer. Previous studies using AI on endometrial biopsies have examined slightly different tasks, for example using images alongside genomic data to differentiate between cancer subtypes. We took 2909 slides with “malignant” and “other or benign” areas annotated by pathologists. A fully supervised convolutional neural network (CNN) model was trained to calculate the probability of a patch from the slide being “malignant” or “other or benign”. Heatmaps of all the patches on each slide were then produced to show malignant areas. These heatmaps were used to train a slide classification model to give the final slide categorisation as either “malignant”, “other or benign” or “insufficient”. The final model was able to accurately classify 90% of all slides correctly and 97% of slides in the malignant class; this accuracy is good enough to allow prioritisation of pathologists’ workload. Publisher PDF

The Internet of Things (IoT) and its relevant advances have attracted significant scholarly, governmental, and industrial attention in recent years. Since the IoT specifications are quite different from what the Internet can deliver today, many groundbreaking techniques, such as Mobile Ad hoc Networks (MANETs) and Wireless Sensor Networks (WSN), have gradually been integrated into IoT. The Routing Protocol for Low power and Lossy network (RPL) is the de-facto IoT routing protocol in such networks. Unfortunately, it is susceptible to numerous internal attacks. Many techniques, such as cryptography, Intrusion Detection System (IDS), and authorization have been used to counter this. The large computational overhead of these techniques limits their direct application to IoT nodes, especially due to their low power and lossy nature. Therefore, this paper proposes a Trust-based Hybrid Cooperative RPL protocol (THC-RPL) to detect malicious Sybil nodes in an RPL-based IoT network. The proposed technique is compared and evaluated with state-of-the-art and is found to outperform them. It detects more attacks while maintaining the packet loss ratio in the range of 15-25%. The average energy consumption of the nodes also remains in the ratio of 60-80 mj. There is approximately 40% more energy conservation at node level with an overall 50% increase in network lifetime. THC-RPL has 10% less message exchange and 0% storage costs.

Radiation to the female pelvis as part of treatment for cancer predisposes young women to develop Premature Ovarian Insufficiency (POI). As the human female is born with their full complement of non-growing follicles which decline in an exponential fashion until the menopause, the age at which POI occurs is dependent on the age of the patient at treatment and the dose received by the ovary. A model that predicts the age at which POI occurs for a known dose at a known age will aid counselling patients on their fertility risk. Patients deemed to be at high risk of POI may be considered to be good candidates for established fertility preservation techniques. An updated and externally validated model of the age-related decline in human ovarian reserve was combined with the best available estimate of the median lethal dose LD50 for the human ovary. Using known age at diagnosis and posited radiotherapy treatment plan to estimate the dose to the least-affected ovary, we use an age-related model of the decline in ovarian reserve to generate a personalized age prediction of premature ovarian insufficiency. Our algorithm is available as an online calculator which graphs model outputs to inform discussions around survivor fertility. We report four example cases across different ages and diagnoses, each with two carefully designed photon and proton treatment plans. The treatment options are compared in terms of remaining fertile lifespan for the survivor. International oncology guidelines now mandate the consideration of later fertility when reviewing treatment options for children diagnosed with cancer. Our calculator (https://sites.cs.st-andrews.ac.uk/radiosensitivity), and the underlying algorithm and models, allow detailed predictions of the impact of various radiotherapy plans on fertility. These patient-specific data enhance pre-treatment discussions around post-treatment fertility and fertility preservation.

Background Mastectomy or breast conserving surgery, both with axillary lymph node dissection, are common treatments for early-stage breast cancer. Monopolar electrocautery is typically used for both procedures, despite evidence of improved clinical outcomes with HARMONIC FOCUS™+. This analysis evaluated the budget impact of adopting HARMONIC FOCUS™+ versus monopolar electrocautery for patients undergoing these procedures from an Italian hospital perspective. Methods Total costs for an annual caseload of 100 patients undergoing mastectomy or breast conserving surgery, with axillary lymph node dissection, with either the intervention or comparator were calculated. Italian clinical and cost input data were utilised. The analysis included costs for the device, operating room time, postoperative length of stay, treating seroma and managing postoperative chest wall drainage. Deterministic and probabilistic sensitivity analyses assessed uncertainty of model input values. Two scenario analyses investigated the impact of conservative estimates of postoperative length of stay reduction and daily hospital cost on the simulated cost difference. Results HARMONIC FOCUS™+ achieves annual savings of EUR 100,043 compared with monopolar electrocautery, derived from lower costs for operating room time, postoperative length of stay and seroma and postoperative chest wall drainage management, offsetting the incremental device cost increase (EUR 43,268). Cost savings are maintained in scenario analyses and across all variations in parameters in deterministic sensitivity analysis, with postoperative hospital stay costs being key drivers of budget impact. The mean (interquartile range) cost savings with HARMONIC FOCUS™+ versus monopolar electrocautery in probabilistic sensitivity analysis are EUR 101,637 (EUR 64,390–137,093) with a 98% probability of being cost saving. Conclusions The intervention demonstrates robust cost savings compared with monopolar electrocautery for mastectomy or breast conserving surgery, with axillary lymph node dissection, in an Italian hospital setting, and improved clinical and resource outcomes. These findings, with other clinical and cost analyses, support HARMONIC FOCUS™+ use in this setting.

Abnormalities and defects that can cause lumbar spinal stenosis often occur in the Intervertebral Disc (IVD) of the patient’s lumbar spine. Their automatic detection and classification require an application of an image analysis algorithm on suitable images, such as mid-sagittal images or traverse mid-height intervertebral disc slices, as inputs. Hence the process of selecting and separating these images from other medical images in the patient’s set of scans is necessary. However, the technological progress in making this process automated is still lagging behind other areas in medical image classification research. In this paper, we report the result of our investigation on the suitability and performance of different approaches of machine learning to automatically select the best traverse plane that cuts closest to the half-height of an IVD from a database of lumbar spine MRI images. This study considers images features extracted using eleven different pre-trained Deep Convolution Neural Network (DCNN) models. We investigate the effectiveness of three dimensionality-reduction techniques and three feature-selection techniques on the classification performance. We also investigate the performance of five different Machine Learning (ML) algorithms and three Fully Connected (FC) neural network learning optimizers which are used to train an image classifier with hyperparameter optimization using a wide range of hyperparameter options and values. The different combinations of methods are tested on a publicly available lumbar spine MRI dataset consisting of MRI studies of 515 patients with symptomatic back pain. Our experiment shows that applying the Support Vector Machine algorithm with a short Gaussian kernel on full-length image features extracted using a pre-trained DenseNet201 model is the best approach to use. This approach gives the minimum per-class classification performance of around 0.88 when measured using the precision and recall metrics. The median performance measured using the precision metric ranges from 0.95 to 0.99 whereas that using the recall metric ranges from 0.93 to 1.0. When only considering the L3/L4, L4/L5, and L5/S1 classes, the minimum F1-Scores range between 0.93 to 0.95, whereas the median F1-Scores range between 0.97 to 0.99.

One of the significant challenges in the Internet of Things (IoT) is the provisioning of guaranteed security and privacy, considering the fact that IoT devices are resource-limited. Oftentimes, in IoT applications, remote users need to obtain real-time data, with guaranteed security and privacy, from resource-limited network nodes through the public Internet. For this purpose, the users need to establish a secure link with the network nodes. Though the IPv6 over low-power wireless personal area networks (6LoWPAN) adaptation layer standard offers IPv6 compatibility for resource-limited wireless networks, the fundamental 6LoWPAN structure ignores security and privacy characteristics. Thus, there is a pressing need to design a resource-efficient authenticated key exchange (AKE) scheme for ensuring secure communication in 6LoWPAN-based resource-limited networks. This paper proposes a resource-efficient secure remote user authentication scheme for 6LoWPAN-based IoT networks, called SRUA-IoT. SRUA-IoT achieves the authentication of remote users and enables the users and network entities to establish private session keys between themselves for indecipherable communication. To this end, SRUA-IoT uses a secure hash algorithm, exclusive-OR operation, and symmetric encryption primitive. We prove through informal security analysis that SRUA-IoT is secured against a variety of malicious attacks. We also prove the security strength of SRUA-IoT through formal security analysis conducted by employing the random oracle model. Additionally, we prove through Scyther-based validation that SRUA-IoT is resilient against various attacks. Likewise, we demonstrate that SRUA-IoT reduces the computational cost of the nodes and communication overheads of the network.

Fractures of the wrist are common in Emergency Departments, where some patients are treated with a procedure called Manipulation under Anaesthesia. In some cases, this procedure is unsuccessful and patients need to revisit the hospital where they undergo surgery to treat the fracture. This work describes a geometric semi-automatic image analysis algorithm to analyse and compare the x-rays of healthy controls and patients with dorsally displaced wrist fractures (Colles’ fractures) who were treated with Manipulation under Anaesthesia. A series of 161 posterior-anterior radiographs from healthy controls and patients with Colles’ fractures were acquired and analysed. The patients’ group was further subdivided according to the outcome of the procedure (successful/unsuccessful) and pre- or post-intervention creating five groups in total (healthy, pre-successful, pre-unsuccessful, post-successful, post-unsuccessful). The semi-automatic analysis consisted of manual location of three landmarks (finger, lunate and radial styloid) and automatic processing to generate 32 geometric and texture measurements, which may be related to conditions such as osteoporosis and swelling of the wrist. Statistical differences were found between patients and controls, as well as between pre- and post-intervention, but not between the procedures. The most distinct measurements were those of texture. Although the study includes a relatively low number of cases and measurements, the statistical differences are encouraging.

Lumbar Spinal Stenosis causes low back pain through pressures exerted on the spinal nerves. This can be verified by measuring the anteroposterior diameter and foraminal widths of the patient’s lumbar spine. Our goal is to develop a novel strategy for assessing the extent of Lumbar Spinal Stenosis by automatically calculating these distances from the patient’s lumbar spine MRI. Our method starts with a semantic segmentation of T1- and T2-weighted composite axial MRI images using SegNet that partitions the image into six regions of interest. They consist of three main regions-of-interest, namely the Intervertebral Disc, Posterior Element, and Thecal Sac, and three auxiliary regions-of-interest that includes the Area between Anterior and Posterior elements. A novel contour evolution algorithm is then applied to improve the accuracy of the segmentation results along important region boundaries. Nine anatomical landmarks on the image are located by delineating the region boundaries found in the segmented image before the anteroposterior diameter and foraminal widths can be measured. The performance of the proposed algorithm was evaluated through a set of experiments on the Lumbar Spine MRI dataset containing MRI studies of 515 patients. These experiments compare the performance of our contour evolution algorithm with the Geodesic Active Contour and Chan-Vese methods over 22 different setups. We found that our method works best when our contour evolution algorithm is applied to improve the accuracy of both the label images used to train the SegNet model and the automatically segmented image. The average error of the calculated right and left foraminal distances relative to their expert-measured distances are 0.28 mm (p = 0.92) and 0.29 mm (p = 0.97), respectively. The average error of the calculated anteroposterior diameter relative to their expert-measured diameter is 0.90 mm (p = 0.92). The method also achieves 96.7% agreement with an expert opinion on determining the severity of the Intervertebral Disc herniations.

The aim was to objectively identify key performance indicators in professional soccer that influence outfield players' league status using an artificial neural network. Mean technical performance data were collected from 966 outfield players' (mean SD; age: 25 ± 4 yr, 1.81 ±) 90-minute performances in the English Football League. ProZone's MatchViewer system and online databases were used to collect data on 347 indicators assessing the total number, accuracy and consistency of passes, tackles, possessions regained, clearances and shots. Players were assigned to one of three categories based on where they went on to complete most of their match time in the following season: group 0 (n = 209 players) went on to play in a lower soccer league, group 1 (n = 637 players) remained in the Football League Championship, and group 2 (n = 120 players) consisted of players who moved up to the English Premier League. The models created correctly predicted between 61.5% and 78.8% of the players' league status. The model with the highest average test performance was for group 0 v 2 (U21 international caps, international caps, median tackles, percentage of first time passes unsuccessful upper quartile, maximum dribbles and possessions gained minimum) which correctly predicted 78.8% of the players' league status with a test error of 8.3%. To date, there has not been a published example of an objective method of predicting career trajectory in soccer. This is a significant development as it highlights the potential for machine learning to be used in the scouting and recruitment process in a professional soccer environment.

Graph drawing, or the automatic layout of graphs, is a challenging problem. There are several search based methods for graph drawing which are based on optimizing an objective function which is formed from a weighted sum of multiple criteria. In this paper, we propose a new neighbourhood search method which uses a tabu search coupled with path relinking to optimize such objective functions for general graph layouts with undirected straight lines. To our knowledge, before our work, neither of these methods have been previously used in general multi-criteria graph drawing. Tabu search uses a memory list to speed up searching by avoiding previously tested solutions, while the path relinking method generates new solutions by exploring paths that connect high quality solutions. We use path relinking periodically within the tabu search procedure to speed up the identification of good solutions. We have evaluated our new method against the commonly used neighbourhood search optimization techniques: hill climbing and simulated annealing. Our evaluation examines the quality of the graph layout (objective function's value) and the speed of layout in terms of the number of evaluated solutions required to draw a graph. We also examine the relative scalability of each method. Our experimental results were applied to both random graphs and a real-world dataset. We show that our method outperforms both hill climbing and simulated annealing by producing a better layout in a lower number of evaluated solutions. In addition, we demonstrate that our method has greater scalability as it can layout larger graphs than the state-of-the-art neighbourhood search methods. Finally, we show that similar results can be produced in a real world setting by testing our method against a standard public graph dataset.

The study of facial growth is explored in many fields of science, including anatomy, genetics, and forensics. In the field of forensics, it acts as a valuable tool for combating child pornography. The present research proposes a new method, based on relative measurements and fixed references of the human face - specifically considering measurements of the diameter of the iris (iris ratio) - for the analysis of facial growth in association with age in children and sub-adults. The experimental sample consisted of digital photographs of 1000 Brazilian subjects, aged between 6 and 22 years, distributed equally by sex and divided into five specific age groups (6, 10, 14, 18, and 22 year olds ± one month). The software package SAFF-2D® (Forensic Facial Analysis System, Brazilian Federal Police, Brazil) was used for positioning 11 landmarks on the images. Ten measurements were calculated and used as fixed references to evaluate the growth of the other measurements for each age group, as well the accumulated growth (6-22 years old). The Intraclass Correlation Coefficient (ICC) was applied for the evaluation of intra-examiner and inter-examiner reliability within a specific set of images. Pearson's Correlation Coefficient was used to assess the association between each measurement taken and the respective age groups. ANOVA and Post-hoc Tukey tests were used to search for statistical differences between the age groups. The outcomes indicated that facial structures grow with different timing in children and adolescents. Moreover, the growth allometry expressed in this study may be used to understand what structures have more or less proportional variation in function for the age ranges studied. The diameter of the iris was found to be the most stable measurement compared to the others and represented the best cephalometric measurement as a fixed reference for facial growth ratios (or indices). The method described shows promising potential for forensic applications, especially as part of the armamentarium against crimes involving child pornography and child abuse.

Diffusion processes in social networks often cause the emergence of global phenomena from individual behavior within a society. The study of those global phenomena and the simulation of those diffusion processes frequently require a good model of the global network. However, survey data and data from online sources are often restricted to single social groups or features, such as age groups, single schools, companies, or interest groups. Hence, a modeling approach is required that extrapolates the locally restricted data to a global network model. We tackle this Missing Data Problem using Link-Prediction techniques from social network research, network generation techniques from the area of Social Simulation, as well as a combination of both. We found that techniques employing less information may be more adequate to solve this problem, especially when data granularity is an issue. We validated the network models created with our techniques on a number of real-world networks, investigating degree distributions as well as the likelihood of links given the geographical distance between two nodes.

To achieve realistic Augmented Reality (AR), shadows play an important role in creating a 3D impression of a scene. Casting virtual shadows on real and virtual objects is one of the topics of research being conducted in this area. In this paper, we propose a new method for creating complex AR indoor scenes using real time depth detection to exert virtual shadows on virtual and real environments. A Kinect camera was used to produce a depth map for the physical scene mixing into a single real-time transparent tacit surface. Once this is created, the camera’s position can be tracked from the reconstructed 3D scene. Real objects are represented by virtual object phantoms in the AR scene enabling users holding a webcam and a standard Kinect camera to capture and reconstruct environments simultaneously. The tracking capability of the algorithm is shown and the findings are assessed drawing upon qualitative and quantitative methods making comparisons with previous AR phantom generation applications. The results demonstrate the robustness of the technique for realistic indoor rendering in AR systems. Refereed/Peer-reviewed

In this paper, we present a new method to recognise the leaf type and identify plant species using phenetic parts of the leaf; lobes, apex and base detection. Most of the research in this area focuses on the popular features such as the shape, colour, vein, and texture, which consumes large amounts of computational processing and are not efficient, especially in the Acer database with a high complexity structure of the leaves. This paper is focused on phenetic parts of the leaf which increases accuracy. Detecting the local maxima and local minima are done based on Centroid Contour Distance for Every Boundary Point, using north and south region to recognise the apex and base. Digital morphology is used to measure the leaf shape and the leaf margin. Centroid Contour Gradient is presented to extract the curvature of leaf apex and base. We analyse 32 leaf images of tropical plants and evaluated with two different datasets, Flavia, and Acer. The best accuracy obtained is 94.76% and 82.6% respectively. Experimental results show the effectiveness of the proposed technique without considering the commonly used features with high computational cost.

The comparative analysis of world music cultures has been the focus of several ethnomusicological studies in the last century. With the advances of Music Information Retrieval and the increased accessibility of sound archives, large-scale analysis of world music with computational tools is today feasible. We investigate music similarity in a corpus of 8200 recordings of folk and traditional music from 137 countries around the world. In particular, we aim to identify music recordings that are most distinct compared to the rest of our corpus. We refer to these recordings as 'outliers'. We use signal processing tools to extract music information from audio recordings, data mining to quantify similarity and detect outliers, and spatial statistics to account for geographical correlation. Our findings suggest that Botswana is the country with the most distinct recordings in the corpus and China is the country with the most distinct recordings when considering spatial correlation. Our analysis includes a comparison of musical attributes and styles that contribute to the 'uniqueness' of the music of each country.

In this paper we compare two academic networking platforms, HASTAC and Hypotheses, to show the distinct ways in which they serve specific communities in the Digital Humanities (DH) in different national and disciplinary contexts. After providing background information on both platforms, we apply co-word analysis and topic modeling to show thematic similarities and differences between the two sites, focusing particularly on how they frame DH as a new paradigm in humanities research. We encounter a much higher ratio of posts using humanities-related terms compared to their digital counterparts, suggesting a one-way dependency of digital humanities-related terms on the corresponding unprefixed labels. The results also show that the terms digital archive, digital literacy, and digital pedagogy are relatively independent from the respective unprefixed terms, and that digital publishing, digital libraries, and digital media show considerable cross-pollination between the specialization and the general noun. The topic modeling reproduces these findings and reveals further differences between the two platforms. Our findings also indicate local differences in how the emerging field of DH is conceptualized and show dynamic topical shifts inside these respective contexts.

Although inequalities in health and socioeconomic status have an important influence on childhood educational performance, the interactions between these multiple factors relating to variation in educational outcomes at micro-level is unknown, and how to evaluate the many possible interactions of these factors is not well established. This paper aims to examine multi-dimensional deprivation factors and their impact on childhood educational outcomes at micro-level, focusing on geographic areas having widely different disparity patterns, in which each area is characterised by six deprivation domains (Income, Health, Geographical Access to Services, Housing, Physical Environment, and Community Safety). Traditional health statistical studies tend to use one global model to describe the whole population for macro-analysis. In this paper, we combine linked educational and deprivation data across small areas (median population of 1500), then use a local modelling technique, the Takagi-Sugeno fuzzy system, to predict area educational outcomes at ages 7 and 11. We define two new metrics, "Micro-impact of Domain" and "Contribution of Domain", to quantify the variations of local impacts of multidimensional factors on educational outcomes across small areas. The two metrics highlight differing priorities. Our study reveals complex multi-way interactions between the deprivation domains, which could not be provided by traditional health statistical methods based on single global model. We demonstrate that although Income has an expected central role, all domains contribute, and in some areas Health, Environment, Access to Services, Housing and Community Safety each could be the dominant factor. Thus the relative importance of health and socioeconomic factors varies considerably for different areas, depending on the levels of each of the other factors, and therefore each component of deprivation must be considered as part of a wider system. Childhood educational achievement could benefit from policies and intervention strategies that are tailored to the local geographic areas' profiles.

Due to the demand for performance improvement and the existence of prior information, semi-supervised community detection with pairwise constraints becomes a hot topic. Most existing methods have been successfully encoding the must-link constraints, but neglect the opposite ones, i.e., the cannot-link constraints, which can force the exclusion between nodes. In this paper, we are interested in understanding the role of cannot-link constraints and effectively encoding pairwise constraints. Towards these goals, we define an integral generative process jointly considering the network topology, must-link and cannot-link constraints. We propose to characterize this process as a Multi-variance Mixed Gaussian Generative (MMGG) Model to address diverse degrees of confidences that exist in network topology and pairwise constraints and formulate it as a weighted nonnegative matrix factorization problem. The experiments on artificial and real-world networks not only illustrate the superiority of our proposed MMGG, but also, most importantly, reveal the roles of pairwise constraints. That is, though the must-link is more important than cannot-link when either of them is available, both must-link and cannot-link are equally important when both of them are available. To the best of our knowledge, this is the first work on discovering and exploring the importance of cannot-link constraints in semi-supervised community detection.

We address the problem of feature selection for classifying a diverse set of chemicals using an array of metal oxide sensors. Our aim is to evaluate a filter approach to feature selection with reference to previous work, which used a wrapper approach on the same data set, and established best features and upper bounds on classification performance. We selected feature sets that exhibit the maximal mutual information with the identity of the chemicals. The selected features closely match those found to perform well in the previous study using a wrapper approach to conduct an exhaustive search of all permitted feature combinations. By comparing the classification performance of support vector machines (using features selected by mutual information) with the performance observed in the previous study, we found that while our approach does not always give the maximum possible classification performance, it always selects features that achieve classification performance approaching the optimum obtained by exhaustive search. We performed further classification using the selected feature set with some common classifiers and found that, for the selected features, Bayesian Networks gave the best performance. Finally, we compared the observed classification performances with the performance of classifiers using randomly selected features. We found that the selected features consistently outperformed randomly selected features for all tested classifiers. The mutual information filter approach is therefore a computationally efficient method for selecting near optimal features for chemical sensor arrays.

Forecasting naturally occurring phenomena is a common problem in many domains of science, and this has been addressed and investigated by many scientists. The importance of time series prediction stems from the fact that it has wide range of applications, including control systems, engineering processes, environmental systems and economics. From the knowledge of some aspects of the previous behaviour of the system, the aim of the prediction process is to determine or predict its future behaviour. In this paper, we consider a novel application of a higher order polynomial neural network architecture called Dynamic Ridge Polynomial Neural Network that combines the properties of higher order and recurrent neural networks for the prediction of physical time series. In this study, four types of signals have been used, which are; The Lorenz attractor, mean value of the AE index, sunspot number, and heat wave temperature. The simulation results showed good improvements in terms of the signal to noise ratio in comparison to a number of higher order and feedforward neural networks in comparison to the benchmarked techniques.

One major drawback of deception detection is its vulnerability to countermeasures, whereby participants wilfully modulate their physiological or neurophysiological response to critical guilt-determining stimuli. One reason for this vulnerability is that stimuli are usually presented slowly. This allows enough time to consciously apply countermeasures, once the role of stimuli is determined. However, by increasing presentation speed, stimuli can be placed on the fringe of awareness, rendering it hard to perceive those that have not been previously identified, hindering the possibility to employ countermeasures. We tested an identity deception detector by presenting first names in Rapid Serial Visual Presentation and instructing participants to lie about their own identity. We also instructed participants to apply a series of countermeasures. The method proved resilient, remaining effective at detecting deception under all countermeasures.

Realistic rendering techniques of outdoor Augmented Reality (AR) has been an attractive topic since the last two decades considering the sizeable amount of publications in computer graphics. Realistic virtual objects in outdoor rendering AR systems require sophisticated effects such as: shadows, daylight and interactions between sky colours and virtual as well as real objects. A few realistic rendering techniques have been designed to overcome this obstacle, most of which are related to non real-time rendering. However, the problem still remains, especially in outdoor rendering. This paper proposed a much newer, unique technique to achieve realistic real-time outdoor rendering, while taking into account the interaction between sky colours and objects in AR systems with respect to shadows in any specific location, date and time. This approach involves three main phases, which cover different outdoor AR rendering requirements. Firstly, sky colour was generated with respect to the position of the sun. Second step involves the shadow generation algorithm, Z-Partitioning: Gaussian and Fog Shadow Maps (Z-GaF Shadow Maps). Lastly, a technique to integrate sky colours and shadows through its effects on virtual objects in the AR system, is introduced. The experimental results reveal that the proposed technique has significantly improved the realism of real-time outdoor AR rendering, thus solving the problem of realistic AR systems.

There has been some improvement in the treatment of preterm infants, which has helped to increase their chance of survival. However, the rate of premature births is still globally increasing. As a result, this group of infants are most at risk of developing severe medical conditions that can affect the respiratory, gastrointestinal, immune, central nervous, auditory and visual systems. In extreme cases, this can also lead to long-term conditions, such as cerebral palsy, mental retardation, learning difficulties, including poor health and growth. In the US alone, the societal and economic cost of preterm births, in 2005, was estimated to be $26.2 billion, per annum. In the UK, this value was close to £2.95 billion, in 2009. Many believe that a better understanding of why preterm births occur, and a strategic focus on prevention, will help to improve the health of children and reduce healthcare costs. At present, most methods of preterm birth prediction are subjective. However, a strong body of evidence suggests the analysis of uterine electrical signals (Electrohysterography), could provide a viable way of diagnosing true labour and predict preterm deliveries. Most Electrohysterography studies focus on true labour detection during the final seven days, before labour. The challenge is to utilise Electrohysterography techniques to predict preterm delivery earlier in the pregnancy. This paper explores this idea further and presents a supervised machine learning approach that classifies term and preterm records, using an open source dataset containing 300 records (38 preterm and 262 term). The synthetic minority oversampling technique is used to oversample the minority preterm class, and cross validation techniques, are used to evaluate the dataset against other similar studies. Our approach shows an improvement on existing studies with 96% sensitivity, 90% specificity, and a 95% area under the curve value with 8% global error using the polynomial classifier.

BackgroundRecent focus on earlier detection of pathogen introduction in human and animal populations has led to the development of surveillance systems based on automated monitoring of health data. Real- or near real-time monitoring of pre-diagnostic data requires automated classification of records into syndromes--syndromic surveillance--using algorithms that incorporate medical knowledge in a reliable and efficient way, while remaining comprehensible to end users.MethodsThis paper describes the application of two of machine learning (Naïve Bayes and Decision Trees) and rule-based methods to extract syndromic information from laboratory test requests submitted to a veterinary diagnostic laboratory.ResultsHigh performance (F1-macro = 0.9995) was achieved through the use of a rule-based syndrome classifier, based on rule induction followed by manual modification during the construction phase, which also resulted in clear interpretability of the resulting classification process. An unmodified rule induction algorithm achieved an F(1-micro) score of 0.979 though this fell to 0.677 when performance for individual classes was averaged in an unweighted manner (F(1-macro)), due to the fact that the algorithm failed to learn 3 of the 16 classes from the training set. Decision Trees showed equal interpretability to the rule-based approaches, but achieved an F(1-micro) score of 0.923 (falling to 0.311 when classes are given equal weight). A Naïve Bayes classifier learned all classes and achieved high performance (F(1-micro)= 0.994 and F(1-macro) = .955), however the classification process is not transparent to the domain experts.ConclusionThe use of a manually customised rule set allowed for the development of a system for classification of laboratory tests into syndromic groups with very high performance, and high interpretability by the domain experts. Further research is required to develop internal validation rules in order to establish automated methods to update model rules without user input.

Objective assessments of lip movement can be beneficial in many disciplines including visual speech recognition, for surgical outcome assessment in patients with cleft lip and for the rehabilitation of patients with facial nerve impairments. The aim of this study was to develop an outcome measure for lip shape during speech using statistical shape analysis techniques. Lip movements during speech were captured from a sample of adult subjects considered as average using a three-dimensional motion capture system. Geometric Morphometrics was employed to extract three-dimensional coordinate data for lip shape during four spoken words decomposed into seven visemes (which included the resting lip shape). Canonical variate analysis was carried out in an attempt to statistically discriminate the seven visemes. The results showed that the second canonical variate discriminated the resting lip shape from articulation of the utterances and accounted for 17.2% of the total variance of the model. The first canonical variate was significant in discriminating between the utterances and accounted for 72.8% of the total variance of the model. The outcome measure was created using the 95% confidence intervals of the canonical variate scores for each subject plotted as ellipses for each viseme. The method and outcome model is proposed as reference to compare lip movement during speech in similar population groups.

Creativity is a complex, multi-faceted concept encompassing a variety of related aspects, abilities, properties and behaviours. If we wish to study creativity scientifically, then a tractable and well-articulated model of creativity is required. Such a model would be of great value to researchers investigating the nature of creativity and in particular, those concerned with the evaluation of creative practice. This paper describes a unique approach to developing a suitable model of how creative behaviour emerges that is based on the words people use to describe the concept. Using techniques from the field of statistical natural language processing, we identify a collection of fourteen key components of creativity through an analysis of a corpus of academic papers on the topic. Words are identified which appear significantly often in connection with discussions of the concept. Using a measure of lexical similarity to help cluster these words, a number of distinct themes emerge, which collectively contribute to a comprehensive and multi-perspective model of creativity. The components provide an ontology of creativity: a set of building blocks which can be used to model creative practice in a variety of domains. The components have been employed in two case studies to evaluate the creativity of computational systems and have proven useful in articulating achievements of this work and directions for further research., Comment: Submitted to PLOS ONE; currently under review. Figures not included

Transabdominal pelvic ultrasound and/or pelvic Magnetic Resonance Imaging are safe, accurate and non - invasive means of determining the size and configuration of the internal female genitalia. The assessment of uterine size and volume is helpful in the assessment of many conditions including disorders of sex development, precocious or delayed puberty, infertility and menstrual disorders. Using our own data from the assessment of MRI scans in healthy young females and data extracted from four studies that assessed uterine volume using transabdominal ultrasound in healthy females we have derived and validated a normative model of uterine volume from birth to age 40 years. This shows that uterine volume increases across childhood, with a faster increase in adolescence reflecting the influence of puberty, followed by a slow but progressive rise during adult life. The model suggests that around 84% of the variation in uterine volumes in the healthy population up to age 40 is due to age alone . The derivation of a validated normative model for uterine volume from birth to age 40 years has important clinical applications by providing age-related reference values for uterine volume. Publisher PDF

In this paper a novel application of a particular type of spiking neural network, a Polychronous Spiking Network, was used for financial time series prediction. It is argued that the inherent temporal capabilities of this type of network are suited to non-stationary data such as this. The performance of the spiking neural network was benchmarked against three systems: two "traditional", rate-encoded, neural networks; a Multi-Layer Perceptron neural network and a Dynamic Ridge Polynomial neural network, and a standard Linear Predictor Coefficients model. For this comparison three non-stationary and noisy time series were used: IBM stock data; US/Euro exchange rate data, and the price of Brent crude oil. The experiments demonstrated favourable prediction results for the Spiking Neural Network in terms of Annualised Return and prediction error for 5-Step ahead predictions. These results were also supported by other relevant metrics such as Maximum Drawdown and Signal-To-Noise ratio. This work demonstrated the applicability of the Polychronous Spiking Network to financial data forecasting and this in turn indicates the potential of using such networks over traditional systems in difficult to manage non-stationary environments. © 2014 Reid et al.

Global climate change is known to result in the emergence or re-emergence of some infectious diseases. Reliable methods to identify the infectious diseases of humans and animals and that are most likely to be influenced by climate are therefore required. Since different priorities will affect the decision to address a particular pathogen threat, decision makers need a standardised method of prioritisation. Ranking methods and Multi-Criteria Decision approaches provide such a standardised method and were employed here to design two different pathogen prioritisation tools. The opinion of 64 experts was elicited to assess the importance of 40 criteria that could be used to prioritise emerging infectious diseases of humans and animals in Canada. A weight was calculated for each criterion according to the expert opinion. Attributes were defined for each criterion as a transparent and repeatable method of measurement. Two different Multi-Criteria Decision Analysis tools were tested, both of which used an additive aggregation approach. These were an Excel spreadsheet tool and a tool developed in software 'M-MACBETH'. The tools were trialed on nine 'test' pathogens. Two different methods of criteria weighting were compared, one using fixed weighting values, the other using probability distributions to account for uncertainty and variation in expert opinion. The ranking of the nine pathogens varied according to the weighting method that was used. In both tools, using both weighting methods, the diseases that tended to rank the highest were West Nile virus, Giardiasis and Chagas, while Coccidioidomycosis tended to rank the lowest. Both tools are a simple and user friendly approach to prioritising pathogens according to climate change by including explicit scoring of 40 criteria and incorporating weighting methods based on expert opinion. They provide a dynamic interactive method that can help to identify pathogens for which a full risk assessment should be pursued.