Predicting fault-prone software modules with rank sum classification

The detection and correction of defects remains among the most time consuming and expensive aspects of software development. Extensive automated testing and code inspections may mitigate their effect, but some code fragments are necessarily more likely to be faulty than others, and automated identification of fault prone modules helps to focus testing and inspections, thus limiting wasted effort and potentially improving detection rates. However, software metrics data is often extremely noisy, with enormous imbalances in the size of the positive and negative classes. In this work, we present a new approach to predictive modelling of fault proneness in software modules, introducing a new feature representation to overcome some of these issues. This rank sum representation offers improved or at worst comparable performance to earlier approaches for standard data sets, and readily allows the user to choose an appropriate trade-off between precision and recall to optimise inspection effort to suit different testing environments. The method is evaluated using the NASA Metrics Data Program (MDP) data sets, and performance is compared with existing studies based on the Support Vector Machine (SVM) and Naïve Bayes (NB) Classifiers, and with our own comprehensive evaluation of these methods.

Mycobacterium abscessus isolated from municipal water : a potential source of human infection

Background Mycobacterium abscessus is a rapidly growing mycobacterium responsible for progressive pulmonary disease, soft tissue and wound infections. The incidence of disease due to M. abscessus has been increasing in Queensland. In a study of Brisbane drinking water, M. abscessus was isolated from ten different locations. The aim of this study was to compare genotypically the M. abscessus isolates obtained from water to those obtained from human clinical specimens. Methods Between 2007 and 2009, eleven isolates confirmed as M. abscessus were recovered from potable water, one strain was isolated from a rainwater tank and another from a swimming pool and two from domestic taps. Seventy-four clinical isolates referred during the same time period were available for comparison using rep-PCR strain typing (Diversilab). Results The drinking water isolates formed two clusters with ≥97% genetic similarity (Water patterns 1 and 2). The tankwater isolate (WP4), one municipal water isolate (WP3) and the pool isolate (WP5) were distinctly different. Patient isolates formed clusters with all of the water isolates except for WP3. Further patient isolates were unrelated to the water isolates. Conclusion The high degree of similarity between strains of M. abscessus from potable water and strains causing infection in humans from the same geographical area, strengthens the possibility that drinking water may be the source of infection in these patients.

Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia

Background Nontuberculous mycobacteria (NTM) are normal inhabitants of a variety of environmental reservoirs including natural and municipal water. The aim of this study was to document the variety of species of NTM in potable water in Brisbane, QLD, with a specific interest in the main pathogens responsible for disease in this region and to explore factors associated with the isolation of NTM. One-litre water samples were collected from 189 routine collection sites in summer and 195 sites in winter. Samples were split, with half decontaminated with CPC 0.005%, then concentrated by filtration and cultured on 7H11 plates in MGIT tubes (winter only). Results Mycobacteria were grown from 40.21% sites in Summer (76/189) and 82.05% sites in winter (160/195). The winter samples yielded the greatest number and variety of mycobacteria as there was a high degree of subculture overgrowth and contamination in summer. Of those samples that did yield mycobacteria in summer, the variety of species differed from those isolated in winter. The inclusion of liquid media increased the yield for some species of NTM. Species that have been documented to cause disease in humans residing in Brisbane that were also found in water include M. gordonae, M. kansasii, M. abscessus, M. chelonae, M. fortuitum complex, M. intracellulare, M. avium complex, M. flavescens, M. interjectum, M. lentiflavum, M. mucogenicum, M. simiae, M. szulgai, M. terrae. M. kansasii was frequently isolated, but M. avium and M. intracellulare (the main pathogens responsible for disease is QLD) were isolated infrequently. Distance of sampling site from treatment plant in summer was associated with isolation of NTM. Pathogenic NTM (defined as those known to cause disease in QLD) were more likely to be identified from sites with narrower diameter pipes, predominantly distribution sample points, and from sites with asbestos cement or modified PVC pipes. Conclusions NTM responsible for human disease can be found in large urban water distribution systems in Australia. Based on our findings, additional point chlorination, maintenance of more constant pressure gradients in the system, and the utilisation of particular pipe materials should be considered.

Isolation of nontuberculous mycobacteria (NTM) from household water and shower aerosols in patients with pulmonary disease caused by NTM

It has been postulated that susceptible individuals may acquire infection with nontuberculous mycobacteria (NTM) from water and aerosol exposure. This study examined household water and shower aerosols of patients with NTM pulmonary disease. The mycobacteria isolated from clinical samples from 20 patients included M. avium (5 patients), M. intracellulare (12 patients), M. abscessus (7 patients), M. gordonae (1 patient), M. lentiflavum (1 patient), M. fortuitum (1 patient), M. peregrinum (1 patient), M. chelonae (1 patient), M. triplex (1 patient), and M. kansasii (1 patient). One-liter water samples and swabs were collected from all taps, and swimming pools or rainwater tanks. Shower aerosols were sampled using Andersen six-stage cascade impactors. For a subgroup of patients, real-time PCR was performed and high-resolution melt profiles were compared to those of ATCC control strains. Pathogenic mycobacteria were isolated from 19 homes. Species identified in the home matched that found in the patient in seven (35%) cases: M. abscessus (3 cases), M. avium (1 case), M. gordonae (1 case), M. lentiflavum (1 case), and M. kansasii (1 case). In an additional patient with M. abscessus infection, this species was isolated from potable water supplying her home. NTM grown from aerosols included M. abscessus (3 homes), M. gordonae (2 homes), M. kansasii (1 home), M. fortuitum complex (4 homes), M. mucogenicum (1 home), and M. wolinskyi (1 home). NTM causing human disease can be isolated from household water and aerosols. The evidence appears strongest for M. avium, M. kansasii, M. lentiflavum, and M. abscessus. Despite a predominance of disease due to M. intracellulare, we found no evidence for acquisition of infection from household water for this species.

Local inter-session variability modelling for object classification

Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

The first x-ray diffraction peak of water as a function of temperature in the range 4 – 63 °C

X-ray diffraction structure functions for water flowing in a 1.5 mm diameter siphon in the temperature range 4 – 63 °C were obtained using a 20 keV beam at the Australian Synchrotron. These functions were compared with structure functions obtained at the Advanced Light Source for a 0.5 mm thick sample of water in the temperature range 1 – 77 °C irradiated with an 11 keV beam. The two sets of structure functions are similar, but there are subtle differences in the shape and relative position of the two functions suggesting a possible differences between the structure of bulk and siphon water. In addition, the first structural peak (Q0) for water in a siphon, showed evidence of a step-wise increase in Q0 with increasing temperature rather than a smoothly varying increase. More experiments are required to investigate this apparent difference.

Cell image classification using histograms, higher order statistics and adaboost

A cell classification algorithm that uses first, second and third order statistics of pixel intensity distributions over pre-defined regions is implemented and evaluated. A cell image is segmented into 6 regions extending from a boundary layer to an inner circle. First, second and third order statistical features are extracted from histograms of pixel intensities in these regions. Third order statistical features used are one-dimensional bispectral invariants. 108 features were considered as candidates for Adaboost based fusion. The best 10 stage fused classifier was selected for each class and a decision tree constructed for the 6-class problem. The classifier is robust, accurate and fast by design.

Real-time image classification for adaptive mission planning using an Autonomous Underwater Vehicle

Real-time image analysis and classification onboard robotic marine vehicles, such as AUVs, is a key step in the realisation of adaptive mission planning for large-scale habitat mapping in previously unexplored environments. This paper describes a novel technique to train, process, and classify images collected onboard an AUV used in relatively shallow waters with poor visibility and non-uniform lighting. The approach utilises Förstner feature detectors and Laws texture energy masks for image characterisation, and a bag of words approach for feature recognition. To improve classification performance we propose a usefulness gain to learn the importance of each histogram component for each class. Experimental results illustrate the performance of the system in characterisation of a variety of marine habitats and its ability to operate onboard an AUV's main processor suitable for real-time mission planning.

Experimental evaluation of an autonomous surface vehicle for water quality and greenhouse gas emission monitoring

This paper describes the experimental evaluation of a novel Autonomous Surface Vehicle capable of navigating complex inland water reservoirs and measuring a range of water quality properties and greenhouse gas emissions. The 16 ft long solar powered catamaran is capable of collecting water column profiles whilst in motion. It is also directly integrated with a reservoir scale floating sensor network to allow remote mission uploads, data download and adaptive sampling strategies. This paper describes the onboard vehicle navigation and control algorithms as well as obstacle avoidance strategies. Experimental results are shown demonstrating its ability to maintain track and avoid obstacles on a variety of large-scale missions and under differing weather conditions, as well as its ability to continuously collect various water quality parameters complimenting traditional manual monitoring campaigns.

An autonomous surface vehicle for water quality monitoring

This paper describes a novel Autonomous Surface Vehicle capable of navigating throughout complex inland water storages and measuring a range of water quality properties and greenhouse gas emissions. The 16 ft long solar powered catamaran can collect this information throughout the water column whilst the vehicle is moving. A unique feature of this ASV is its integration into a storage scale floating sensor network to allow remote mission uploads, data download and adaptive sampling strategies. This paper provides an overview of the vehicle design and operation including control, laser-based obstacle avoidance, and vision-based inspection capabilities. Experimental results are shown illustrating its ability to continuously collect key water quality parameters and compliment intensive manual monitoring campaigns.

A comparison between bagasse and water efficient alternatives using KOH/AQ pulping

Numerous crops grow in sugar regions that have the potential to increase the amount of biomass available to a small bagasse-based pulp factory. Arundo donax and Sorghum offer unique advantages to farmers compared to other agricultural crops. Sorghum bicolour requires only 1/3 of the water of sugarcane. Arundo donax is a very high yield crop, it can also grow with little water but it has the further advantage in that it is also highly stress tolerant, making it suitable for land which is unsuited to other crops. Pulps produced from these crops were benchmarked against sugarcane bagasse pulp. Arundo, sorghum and bagasse were pulped using KOH and anthraquinone to 20 Kappa number so as to produce a bleachable pulp which is suitable for making photocopier paper and tissue products. The unbleached sorghum pulp has better tensile strength properties than the unbleached Arundo pulp (43.8 Nm/g compared to 21.4 Nm/g) and the bleached sorghum pulp tensile strength was similar to bagasse (28.4 Nm/g). At 20 Kappa number, sorghum pulp had acceptable yield for a non-wood fibre (45% c.f. 55% for bagasse), Arundo donax pulp had low tensile strength, and relatively low yield (38.7%), even for an agricultural fibre and required severe cooking conditions to achieve similar delignification to sugarcane bagasse or sorghum. Sorghum and Arundo donax produced thicker handsheets than bagasse (>160 µm c.f. 122 µm for bagasse). In preliminary experiments sorghum and bagasse responded slightly better to Totally Chlorine Free peroxide bleaching (QPP), although none achieved a satisfactory brightness level and further improvement would be required to produce a bleached pulp.

The impact of OCR accuracy on automated cancer classification of pathology reports

Objective To evaluate the effects of Optical Character Recognition (OCR) on the automatic cancer classification of pathology reports. Method Scanned images of pathology reports were converted to electronic free-text using a commercial OCR system. A state-of-the-art cancer classification system, the Medical Text Extraction (MEDTEX) system, was used to automatically classify the OCR reports. Classifications produced by MEDTEX on the OCR versions of the reports were compared with the classification from a human amended version of the OCR reports. Results The employed OCR system was found to recognise scanned pathology reports with up to 99.12% character accuracy and up to 98.95% word accuracy. Errors in the OCR processing were found to minimally impact on the automatic classification of scanned pathology reports into notifiable groups. However, the impact of OCR errors is not negligible when considering the extraction of cancer notification items, such as primary site, histological type, etc. Conclusions The automatic cancer classification system used in this work, MEDTEX, has proven to be robust to errors produced by the acquisition of freetext pathology reports from scanned images through OCR software. However, issues emerge when considering the extraction of cancer notification items.

Classification of pathology reports for cancer registry notifications

Objective: To develop a system for the automatic classification of pathology reports for Cancer Registry notifications. Method: A two pass approach is proposed to classify whether pathology reports are cancer notifiable or not. The first pass queries pathology HL7 messages for known report types that are received by the Queensland Cancer Registry (QCR), while the second pass aims to analyse the free text reports and identify those that are cancer notifiable. Cancer Registry business rules, natural language processing and symbolic reasoning using the SNOMED CT ontology were adopted in the system. Results: The system was developed on a corpus of 500 histology and cytology reports (with 47% notifiable reports) and evaluated on an independent set of 479 reports (with 52% notifiable reports). Results show that the system can reliably classify cancer notifiable reports with a sensitivity, specificity, and positive predicted value (PPV) of 0.99, 0.95, and 0.95, respectively for the development set, and 0.98, 0.96, and 0.96 for the evaluation set. High sensitivity can be achieved at a slight expense in specificity and PPV. Conclusion: The system demonstrates how medical free-text processing enables the classification of cancer notifiable pathology reports with high reliability for potential use by Cancer Registries and pathology laboratories.