Automatic classification of free-text radiology reports to identify limb fractures using machine learning and the SNOMED CT ontology

Objective To develop and evaluate machine learning techniques that identify limb fractures and other abnormalities (e.g. dislocations) from radiology reports. Materials and Methods 99 free-text reports of limb radiology examinations were acquired from an Australian public hospital. Two clinicians were employed to identify fractures and abnormalities from the reports; a third senior clinician resolved disagreements. These assessors found that, of the 99 reports, 48 referred to fractures or abnormalities of limb structures. Automated methods were then used to extract features from these reports that could be useful for their automatic classification. The Naive Bayes classification algorithm and two implementations of the support vector machine algorithm were formally evaluated using cross-fold validation over the 99 reports. Result Results show that the Naive Bayes classifier accurately identifies fractures and other abnormalities from the radiology reports. These results were achieved when extracting stemmed token bigram and negation features, as well as using these features in combination with SNOMED CT concepts related to abnormalities and disorders. The latter feature has not been used in previous works that attempted classifying free-text radiology reports. Discussion Automated classification methods have proven effective at identifying fractures and other abnormalities from radiology reports (F-Measure up to 92.31%). Key to the success of these techniques are features such as stemmed token bigrams, negations, and SNOMED CT concepts associated with morphologic abnormalities and disorders. Conclusion This investigation shows early promising results and future work will further validate and strengthen the proposed approaches.

Classification of cancer-related death certificates using machine learning

Background Cancer monitoring and prevention relies on the critical aspect of timely notification of cancer cases. However, the abstraction and classification of cancer from the free-text of pathology reports and other relevant documents, such as death certificates, exist as complex and time-consuming activities. Aims In this paper, approaches for the automatic detection of notifiable cancer cases as the cause of death from free-text death certificates supplied to Cancer Registries are investigated. Method A number of machine learning classifiers were studied. Features were extracted using natural language techniques and the Medtex toolkit. The numerous features encompassed stemmed words, bi-grams, and concepts from the SNOMED CT medical terminology. The baseline consisted of a keyword spotter using keywords extracted from the long description of ICD-10 cancer related codes. Results Death certificates with notifiable cancer listed as the cause of death can be effectively identified with the methods studied in this paper. A Support Vector Machine (SVM) classifier achieved best performance with an overall F-measure of 0.9866 when evaluated on a set of 5,000 free-text death certificates using the token stem feature set. The SNOMED CT concept plus token stem feature set reached the lowest variance (0.0032) and false negative rate (0.0297) while achieving an F-measure of 0.9864. The SVM classifier accounts for the first 18 of the top 40 evaluated runs, and entails the most robust classifier with a variance of 0.001141, half the variance of the other classifiers. Conclusion The selection of features significantly produced the most influences on the performance of the classifiers, although the type of classifier employed also affects performance. In contrast, the feature weighting schema created a negligible effect on performance. Specifically, it is found that stemmed tokens with or without SNOMED CT concepts create the most effective feature when combined with an SVM classifier.

Application of machine learning technique in wind turbine fault diagnosis

In this study, a machine learning technique called anomaly detection is employed for wind turbine bearing fault detection. Basically, the anomaly detection algorithm is used to recognize the presence of unusual and potentially faulty data in a dataset, which contains two phases: a training phase and a testing phase. Two bearing datasets were used to validate the proposed technique, fault-seeded bearing from a test rig located at Case Western Reserve University to validate the accuracy of the anomaly detection method, and a test to failure data of bearings from the NSF I/UCR Center for Intelligent Maintenance Systems (IMS). The latter data set was used to compare anomaly detection with SVM, a previously well-known applied method, in rapidly finding the incipient faults.

Fault diagnosis of low speed bearing based on acoustic emission signal and multi-class relevance vector machine

This study presents an acoustic emission (AE) based fault diagnosis for low speed bearing using multi-class relevance vector machine (RVM). A low speed test rig was developed to simulate the various defects with shaft speeds as low as 10 rpm under several loading conditions. The data was acquired using anAEsensor with the test bearing operating at a constant loading (5 kN) andwith a speed range from20 to 80 rpm. This study is aimed at finding a reliable method/tool for low speed machines fault diagnosis based on AE signal. In the present study, component analysis was performed to extract the bearing feature and to reduce the dimensionality of original data feature. The result shows that multi-class RVM offers a promising approach for fault diagnosis of low speed machines.

Proteases and proteomics : cutting to the core of human skin pathologies

Preserving the integrity of the skin's outermost layer (the epidermis) is vital for humans to thrive in hostile surroundings. Covering the entire body, the epidermis forms a thin but impenetrable cellular cordon that repels external assaults and blocks escape of water and electrolytes from within. This structure exists in a perpetual state of regeneration where the production of new cellular subunits at the base of the epidermis is offset by the release of terminally differentiated corneocytes from the surface. It is becoming increasingly clear that proteases hold vital roles in assembling and maintaining the epidermal barrier. More than 30 proteases are expressed by keratinocytes or infiltrating immune cells and the activity of each must be maintained within narrow limits and confined to the correct time and place. Accordingly, over- or under-exertion of proteolytic activity is a common factor in a multitude of skin disorders that range in severity from relatively mild to life-threatening. This review explores the current state of knowledge on the involvement of proteases in skin diseases and the latest findings from proteomic and transcriptomic studies focused on uncovering novel (patho)physiological roles for these enzymes.

Towards the development of 1-to-n human machine interfaces for unmanned aerial vehicles

Multi-touch interfaces across a wide range of hardware platforms are becoming pervasive. This is due to the adoption of smart phones and tablets in both the consumer and corporate market place. This paper proposes a human-machine interface to interact with unmanned aerial systems based on the philosophy of multi-touch hardware-independent high-level interaction with multiple systems simultaneously. Our approach incorporates emerging development methods for multi-touch interfaces on mobile platforms. A framework is defined for supporting multiple protocols. An open source solution is presented that demonstrates: architecture supporting different communications hardware; an extensible approach for supporting multiple protocols; and the ability to monitor and interact with multiple UAVs from multiple clients simultaneously. Validation tests were conducted to assess the performance, scalability and impact on packet latency under different client configurations.

Modeling and sensorless direct torque and flux control of a dual-airgap axial flux permanent-magnet machine with field-weakening operation

This paper presents the modeling and motion-sensorless direct torque and flux control of a novel dual-airgap axial-flux permanent-magnet machine optimized for use in flywheel energy storage system (FESS) applications. Independent closed-loop torque and stator flux regulation are performed in the stator flux ( x-y) reference frame via two PI controllers. This facilitates fast torque dynamics, which is critical as far as energy charging/discharging in the FESS is concerned. As FESS applications demand high-speed operation, a new field-weakening algorithm is proposed in this paper. Flux weakening is achieved autonomously once the y-axis voltage exceeds the available inverter voltage. An inherently speed sensorless stator flux observer immune to stator resistance variations and dc-offset effects is also proposed for accurate flux and speed estimation. The proposed observer eliminates the rotary encoder, which in turn reduces the overall weight and cost of the system while improving its reliability. The effectiveness of the proposed control scheme has been verified by simulations and experiments on a machine prototype.

Modeling and position-sensorless control of a dual-airgap axial flux permanent magnet machine for flywheel energy storage systems

This paper presents the modeling and position-sensorless vector control of a dual-airgap axial flux permanent magnet (AFPM) machine optimized for use in flywheel energy storage system (FESS) applications. The proposed AFPM machine has two sets of three-phase stator windings but requires only a single power converter to control both the electromagnetic torque and the axial levitation force. The proper controllability of the latter is crucial as it can be utilized to minimize the vertical bearing stress to improve the efficiency of the FESS. The method for controlling both the speed and axial displacement of the machine is discussed. An inherent speed sensorless observer is also proposed for speed estimation. The proposed observer eliminates the rotary encoder, which in turn reduces the overall weight and cost of the system while improving its reliability. The effectiveness of the proposed control scheme has been verified by simulations and experiments on a prototype machine.

Plasma-aided nanofabrication: where is the cutting edge?

Plasma-aided nanofabrication is a rapidly expanding area of research spanning disciplines ranging from physics and chemistry of plasmas and gas discharges to solid state physics, materials science, surface science, nanoscience and nanotechnology and related engineering subjects. The current status of the research field is discussed and examples of superior performance and competitive advantage of plasma processes and techniques are given. These examples are selected to represent a range of applications of two major types of plasmas suitable for nanoscale synthesis and processing, namely thermally non-equilibrium and thermal plasmas. Major concepts and terminology used in the field are introduced. The paper also pinpoints the major challenges facing plasma-aided nanofabrication and identifies some emerging topics for future research. © 2007 IOP Publishing Ltd.

Brain decoding based on functional magnetic resonance imaging using machine learning : a comparative study

Brain decoding of functional Magnetic Resonance Imaging data is a pattern analysis task that links brain activity patterns to the experimental conditions. Classifiers predict the neural states from the spatial and temporal pattern of brain activity extracted from multiple voxels in the functional images in a certain period of time. The prediction results offer insight into the nature of neural representations and cognitive mechanisms and the classification accuracy determines our confidence in understanding the relationship between brain activity and stimuli. In this paper, we compared the efficacy of three machine learning algorithms: neural network, support vector machines, and conditional random field to decode the visual stimuli or neural cognitive states from functional Magnetic Resonance data. Leave-one-out cross validation was performed to quantify the generalization accuracy of each algorithm on unseen data. The results indicated support vector machine and conditional random field have comparable performance and the potential of the latter is worthy of further investigation.

Cutting the Gordian knot of futility : a case for law reform on unilateral withholding and withdrawal of potentially life-sustaining treatment

In this paper, we propose law reform with respect to the unilateral withholding or withdrawal of potentially life-sustaining treatment in Australia and New Zealand. That is, where a doctor withholds or withdraws potentially life-sustaining treatment without consent from a patient or a patient’s substitute decision-maker (where the patient lacks capacity), or authorisation from a court or tribunal, or by operation of a statute or justifiable government or institutional policy. Our proposal is grounded in the core values that do (or should) underpin a regulatory framework on an issue such as this; these values are drawn from existing commitments made by Australia and New Zealand through legislation, the common law, and conventions and treaties. It is also grounded in a critical review of the law on unilateral withholding and withdrawal as well as the legal context within which this issue sits in Australasia. We argue that the current law is inconsistent with the core values and develop a proposal for a legal response to this issue that more closely aligns with the core values it is supposed to serve.

Female genital cutting : An overview of Australian legal principles, ethical controversies, and medical, legal and social challenges

Female genital cutting (also often called female genital mutilation, or female circumcision) is a cultural practice that originated thousands of years ago. Female genital cutting has various forms, some of which are more invasive than others, but all of which produce health, legal and social consequences for those involved. Due to patterns of immigration in Australia, especially since the 1990s, there are women in Australia who have experienced female genital cutting. There may be some families, or some parents, who still hold a cultural commitment to female genital cutting. As a result, female genital cutting presents complex legal, ethical, medical and social challenges in contemporary Australian society. Medical practitioners and other health and welfare workers may encounter women who have experienced genital cutting and who require treatment for its sequelae. Currently, legislative frameworks for female genital cutting vary across states and territories, including the penalties for conducting it, and for removing a child for the purpose of conducting it outside Australia. This presentation provides an overview of the history, nature and consequences of the various forms of female genital cutting, and of the major Australian legal principles, ethical controversies, and medical, legal and social challenges in this field.

Higher order spectra based support vector machine for arrhythmia classification

Heart rate variability (HRV) refers to the regulation of the sinoatrial node, the natural pacemaker of the heart by the sympathetic and parasympathetic branches of the autonomic nervous system. HRV analysis is an important tool to observe the heart’s ability to respond to normal regulatory impulses that affect its rhythm. Like many bio-signals, HRV signals are non-linear in nature. Higher order spectral analysis (HOS) is known to be a good tool for the analysis of non-linear systems and provides good noise immunity. A computer-based arrhythmia detection system of cardiac states is very useful in diagnostics and disease management. In this work, we studied the identification of the HRV signals using features derived from HOS. These features were fed to the support vector machine (SVM) for classification. Our proposed system can classify the normal and other four classes of arrhythmia with an average accuracy of more than 85%.