Decision trees for detection of activity intensity in youth with cerebral palsy

PURPOSE To develop and test decision tree (DT) models to classify physical activity (PA) intensity from accelerometer output and Gross Motor Function Classification System (GMFCS) classification level in ambulatory youth with cerebral palsy (CP); and 2) compare the classification accuracy of the new DT models to that achieved by previously published cut-points for youth with CP. METHODS Youth with CP (GMFCS Levels I - III) (N=51) completed seven activity trials with increasing PA intensity while wearing a portable metabolic system and ActiGraph GT3X accelerometers. DT models were used to identify vertical axis (VA) and vector magnitude (VM) count thresholds corresponding to sedentary (SED) (<1.5 METs), light PA (LPA) (>/=1.5 and <3 METs) and moderate-to-vigorous PA (MVPA) (>/=3 METs). Models were trained and cross-validated using the 'rpart' and 'caret' packages within R. RESULTS For the VA (VA_DT) and VM decision trees (VM_DT), a single threshold differentiated LPA from SED, while the threshold for differentiating MVPA from LPA decreased as the level of impairment increased. The average cross-validation accuracy for the VC_DT was 81.1%, 76.7%, and 82.9% for GMFCS levels I, II, and III, respectively. The corresponding cross-validation accuracy for the VM_DT was 80.5%, 75.6%, and 84.2%, respectively. Within each GMFCS level, the decision tree models achieved better PA intensity recognition than previously published cut-points. The accuracy differential was greatest among GMFCS level III participants, in whom the previously published cut-points misclassified 40% of the MVPA activity trials. CONCLUSION GMFCS-specific cut-points provide more accurate assessments of MVPA levels in youth with CP across the full spectrum of ambulatory ability.

Quick abnormal-bid-detection method for construction contract auctions

Non-competitive bids have recently become a major concern in both Public and Private sector construction contract auctions. Consequently, several models have been developed to help identify bidders potentially involved in collusive practices. However, most of these models require complex calculations and extensive information that is difficult to obtain. The aim of this paper is to utilize recent developments for detecting abnormal bids in capped auctions (auctions with an upper bid limit set by the auctioner) and extend them to the more conventional uncapped auctions (where no such limits are set). To accomplish this, a new method is developed for estimating the values of bid distribution supports by using the solution to what has become known as the German tank problem. The model is then demonstrated and tested on a sample of real construction bid data and shown to detect cover bids with high accuracy. This work contributes to an improved understanding of abnormal bid behavior as an aid to detecting and monitoring potential collusive bid practices.

Infrared thermal imaging for automated detection of diabetic foot complications

Background Although thermal imaging can be a valuable technology in the prevention and management of diabetic foot disease, it is not yet widely used in clinical practice. Technological advancement in infrared imaging increases its application range. The aim was to explore the first steps in the applicability of high-resolution infrared thermal imaging for noninvasive automated detection of signs of diabetic foot disease. Methods The plantar foot surfaces of 15 diabetes patients were imaged with an infrared camera (resolution, 1.2 mm/pixel): 5 patients had no visible signs of foot complications, 5 patients had local complications (e.g., abundant callus or neuropathic ulcer), and 5 patients had difuse complications (e.g., Charcot foot, infected ulcer, or critical ischemia). Foot temperature was calculated as mean temperature across pixels for the whole foot and for specified regions of interest (ROIs). Results No diferences in mean temperature >1.5 °C between the ipsilateral and the contralateral foot were found in patients without complications. In patients with local complications, mean temperatures of the ipsilateral and the contralateral foot were similar, but temperature at the ROI was >2 °C higher compared with the corresponding region in the contralateral foot and to the mean of the whole ipsilateral foot. In patients with difuse complications, mean temperature diferences of >3 °C between ipsilateral and contralateral foot were found. Conclusions With an algorithm based on parameters that can be captured and analyzed with a high-resolution infrared camera and a computer, it is possible to detect signs of diabetic foot disease and to discriminate between no, local, or difuse diabetic foot complications. As such, an intelligent telemedicine monitoring system for noninvasive automated detection of signs of diabetic foot disease is one step closer. Future studies are essential to confirm and extend these promising early findings.

Automatic detection of diabetic foot complications with infrared thermography by asymmetric analysis

Early identification of diabetic foot complications and their precursors is essential in preventing their devastating consequences, such as foot infection and amputation. Frequent, automatic risk assessment by an intelligent telemedicine system might be feasible and cost effective. Infrared thermography is a promising modality for such a system. The temperature differences between corresponding areas on contralateral feet are the clinically significant parameters. This asymmetric analysis is hindered by (1) foot segmentation errors, especially when the foot temperature and the ambient temperature are comparable, and by (2) different shapes and sizes between contralateral feet due to deformities or minor amputations. To circumvent the first problem, we used a color image and a thermal image acquired synchronously. Foot regions, detected in the color image, were rigidly registered to the thermal image. This resulted in 97.8% ± 1.1% sensitivity and 98.4% ± 0.5% specificity over 76 high-risk diabetic patients with manual annotation as a reference. Nonrigid landmark-based registration with Bsplines solved the second problem. Corresponding points in the two feet could be found regardless of the shapes and sizes of the feet. With that, the temperature difference of the left and right feet could be obtained.

Quarantine and surveillance strategies for plant pathogen detection and control

Incursions of plant pests and diseases pose serious threats to food security, agricultural productivity and the natural environment. One of the challenges in confidently delimiting and eradicating incursions is how to choose from an arsenal of surveillance and quarantine approaches in order to best control multiple dispersal pathways. Anthropogenic spread (propagules carried on humans or transported on produce or equipment) can be controlled with quarantine measures, which in turn can vary in intensity. In contrast, environmental spread processes are more difficult to control, but often have a temporal signal (e.g. seasonality) which can introduce both challenges and opportunities for surveillance and control. This leads to complex decisions regarding when, where and how to search. Recent modelling investigations of surveillance performance have optimised the output of simulation models, and found that a risk-weighted randomised search can perform close to optimally. However, exactly how quarantine and surveillance strategies should change to reflect different dispersal modes remains largely unaddressed. Here we develop a spatial simulation model of a plant fungal-pathogen incursion into an agricultural region, and its subsequent surveillance and control. We include structural differences in dispersal via the interplay of biological, environmental and anthropogenic connectivity between host sites (farms). Our objective was to gain broad insights into the relative roles played by different spread modes in propagating an invasion, and how incorporating knowledge of these spread risks may improve approaches to quarantine restrictions and surveillance. We find that broad heuristic rules for quarantine restrictions fail to contain the pathogen due to residual connectivity between sites, but surveillance measures enable early detection and successfully lead to suppression of the pathogen in all farms. Alternative surveillance strategies attain similar levels of performance by incorporating environmental or anthropogenic dispersal risk in the prioritisation of sites. Our model provides the basis to develop essential insights into the effectiveness of different surveillance and quarantine decisions for fungal pathogen control. Parameterised for authentic settings it will aid our understanding of how the extent and resolution of interventions should suitably reflect the spatial structure of dispersal processes.

vSpeak: Edge detection based feature extraction for sign to text conversion

This paper presents 'vSpeak', the first initiative taken in Pakistan for ICT enabled conversion of dynamic Sign Urdu gestures into natural language sentences. To realize this, vSpeak has adopted a novel approach for feature extraction using edge detection and image compression which gives input to the Artificial Neural Network that recognizes the gesture. This technique caters for the blurred images as well. The training and testing is currently being performed on a dataset of 200 patterns of 20 words from Sign Urdu with target accuracy of 90% and above.

R1SVM: A randomised nonlinear approach to large-scale anomaly detection

The problem of unsupervised anomaly detection arises in a wide variety of practical applications. While one-class support vector machines have demonstrated their effectiveness as an anomaly detection technique, their ability to model large datasets is limited due to their memory and time complexity for training. To address this issue for supervised learning of kernel machines, there has been growing interest in random projection methods as an alternative to the computationally expensive problems of kernel matrix construction and sup-port vector optimisation. In this paper we leverage the theory of nonlinear random projections and propose the Randomised One-class SVM (R1SVM), which is an efficient and scalable anomaly detection technique that can be trained on large-scale datasets. Our empirical analysis on several real-life and synthetic datasets shows that our randomised 1SVM algorithm achieves comparable or better accuracy to deep auto encoder and traditional kernelised approaches for anomaly detection, while being approximately 100 times faster in training and testing.

Visual detection of occluded crop: For automated harvesting

This paper presents a novel crop detection system applied to the challenging task of field sweet pepper (capsicum) detection. The field-grown sweet pepper crop presents several challenges for robotic systems such as the high degree of occlusion and the fact that the crop can have a similar colour to the background (green on green). To overcome these issues, we propose a two-stage system that performs per-pixel segmentation followed by region detection. The output of the segmentation is used to search for highly probable regions and declares these to be sweet pepper. We propose the novel use of the local binary pattern (LBP) to perform crop segmentation. This feature improves the accuracy of crop segmentation from an AUC of 0.10, for previously proposed features, to 0.56. Using the LBP feature as the basis for our two-stage algorithm, we are able to detect 69.2% of field grown sweet peppers in three sites. This is an impressive result given that the average detection accuracy of people viewing the same colour imagery is 66.8%.

Automated feature engineering for HTTP tunnel detection

Generating discriminative input features is a key requirement for achieving highly accurate classifiers. The process of generating features from raw data is known as feature engineering and it can take significant manual effort. In this paper we propose automated feature engineering to derive a suite of additional features from a given set of basic features with the aim of both improving classifier accuracy through discriminative features, and to assist data scientists through automation. Our implementation is specific to HTTP computer network traffic. To measure the effectiveness of our proposal, we compare the performance of a supervised machine learning classifier built with automated feature engineering versus one using human-guided features. The classifier addresses a problem in computer network security, namely the detection of HTTP tunnels. We use Bro to process network traffic into base features and then apply automated feature engineering to calculate a larger set of derived features. The derived features are calculated without favour to any base feature and include entropy, length and N-grams for all string features, and counts and averages over time for all numeric features. Feature selection is then used to find the most relevant subset of these features. Testing showed that both classifiers achieved a detection rate above 99.93% at a false positive rate below 0.01%. For our datasets, we conclude that automated feature engineering can provide the advantages of increasing classifier development speed and reducing development technical difficulties through the removal of manual feature engineering. These are achieved while also maintaining classification accuracy.

Automatic wildfire detection and simulation using optical information from unmanned aerial systems

In many parts of the world, uncontrolled fires in sparsely populated areas are a major concern as they can quickly grow into large and destructive conflagrations in short time spans. Detecting these fires has traditionally been a job for trained humans on the ground, or in the air. In many cases, these manned solutions are simply not able to survey the amount of area necessary to maintain sufficient vigilance and coverage. This paper investigates the use of unmanned aerial systems (UAS) for automated wildfire detection. The proposed system uses low-cost, consumer-grade electronics and sensors combined with various airframes to create a system suitable for automatic detection of wildfires. The system employs automatic image processing techniques to analyze captured images and autonomously detect fire-related features such as fire lines, burnt regions, and flammable material. This image recognition algorithm is designed to cope with environmental occlusions such as shadows, smoke and obstructions. Once the fire is identified and classified, it is used to initialize a spatial/temporal fire simulation. This simulation is based on occupancy maps whose fidelity can be varied to include stochastic elements, various types of vegetation, weather conditions, and unique terrain. The simulations can be used to predict the effects of optimized firefighting methods to prevent the future propagation of the fires and greatly reduce time to detection of wildfires, thereby greatly minimizing the ensuing damage. This paper also documents experimental flight tests using a SenseFly Swinglet UAS conducted in Brisbane, Australia as well as modifications for custom UAS.

Business busts and U.S. auditors' interest in fraud detection: Evidence from the 20th century

We find evidence that U.S. auditors increased their attention to fraud detection during or immediately after the economic contractions of the 20th century, based on a content analysis of the 12 volumes of the 20th-century auditing reference series Montgomery’s Auditing. Contractions, however, do not seem to have affected auditors’ attention to the formal goal of fraud detection. The study suggests that auditors’ aversion to the heightened risks of fraud during economic downturns leads them to focus more on fraud detection at those times regardless of the particular guidance in formal audit standards. This study is the first to find some evidence of a recession-influenced difference between fraud detection practices and formal fraud detection goals.

Using multi-label classification for acoustic pattern detection and assisting bird species surveys

Acoustics is a rich source of environmental information that can reflect the ecological dynamics. To deal with the escalating acoustic data, a variety of automated classification techniques have been used for acoustic patterns or scene recognition, including urban soundscapes such as streets and restaurants; and natural soundscapes such as raining and thundering. It is common to classify acoustic patterns under the assumption that a single type of soundscapes present in an audio clip. This assumption is reasonable for some carefully selected audios. However, only few experiments have been focused on classifying simultaneous acoustic patterns in long-duration recordings. This paper proposes a binary relevance based multi-label classification approach to recognise simultaneous acoustic patterns in one-minute audio clips. By utilising acoustic indices as global features and multilayer perceptron as a base classifier, we achieve good classification performance on in-the-field data. Compared with single-label classification, multi-label classification approach provides more detailed information about the distributions of various acoustic patterns in long-duration recordings. These results will merit further biodiversity investigations, such as bird species surveys.

Stationary processes for object detection and non-rigid structure-from-motion

Stationary processes are random variables whose value is a signal and whose distribution is invariant to translation in the domain of the signal. They are intimately connected to convolution, and therefore to the Fourier transform, since the covariance matrix of a stationary process is a Toeplitz matrix, and Toeplitz matrices are the expression of convolution as a linear operator. This thesis utilises this connection in the study of i) efficient training algorithms for object detection and ii) trajectory-based non-rigid structure-from-motion.

When and why threats go undetected: Impacts of event rate and shift length on threat detection accuracy during airport baggage screening

Objective: We aimed to assess the impact of task demands and individual characteristics on threat detection in baggage screeners. Background: Airport security staff work under time constraints to ensure optimal threat detection. Understanding the impact of individual characteristics and task demands on performance is vital to ensure accurate threat detection. Method: We examined threat detection in baggage screeners as a function of event rate (i.e., number of bags per minute) and time on task across 4 months. We measured performance in terms of the accuracy of detection of Fictitious Threat Items (FTIs) randomly superimposed on X-ray images of real passenger bags. Results: Analyses of the percentage of correct FTI identifications (hits) show that longer shifts with high baggage throughput result in worse threat detection. Importantly, these significant performance decrements emerge within the first 10 min of these busy screening shifts only. Conclusion: Longer shift lengths, especially when combined with high baggage throughput, increase the likelihood that threats go undetected. Application: Shorter shift rotations, although perhaps difficult to implement during busy screening periods, would ensure more consistently high vigilance in baggage screeners and, therefore, optimal threat detection and passenger safety.