External knowledge and query strategies in active learning: A study in clinical information extraction

This paper presents a new active learning query strategy for information extraction, called Domain Knowledge Informativeness (DKI). Active learning is often used to reduce the amount of annotation effort required to obtain training data for machine learning algorithms. A key component of an active learning approach is the query strategy, which is used to iteratively select samples for annotation. Knowledge resources have been used in information extraction as a means to derive additional features for sample representation. DKI is, however, the first query strategy that exploits such resources to inform sample selection. To evaluate the merits of DKI, in particular with respect to the reduction in annotation effort that the new query strategy allows to achieve, we conduct a comprehensive empirical comparison of active learning query strategies for information extraction within the clinical domain. The clinical domain was chosen for this work because of the availability of extensive structured knowledge resources which have often been exploited for feature generation. In addition, the clinical domain offers a compelling use case for active learning because of the necessary high costs and hurdles associated with obtaining annotations in this domain. Our experimental findings demonstrated that 1) amongst existing query strategies, the ones based on the classification model’s confidence are a better choice for clinical data as they perform equally well with a much lighter computational load, and 2) significant reductions in annotation effort are achievable by exploiting knowledge resources within active learning query strategies, with up to 14% less tokens and concepts to manually annotate than with state-of-the-art query strategies.

Post-quantum key exchange for the TLS protocol from the ring learning with errors problem

Lattice-based cryptographic primitives are believed to offer resilience against attacks by quantum computers. We demonstrate the practicality of post-quantum key exchange by constructing cipher suites for the Transport Layer Security (TLS) protocol that provide key exchange based on the ring learning with errors (R-LWE) problem, we accompany these cipher suites with a rigorous proof of security. Our approach ties lattice-based key exchange together with traditional authentication using RSA or elliptic curve digital signatures: the post-quantum key exchange provides forward secrecy against future quantum attackers, while authentication can be provided using RSA keys that are issued by today's commercial certificate authorities, smoothing the path to adoption. Our cryptographically secure implementation, aimed at the 128-bit security level, reveals that the performance price when switching from non-quantum-safe key exchange is not too high. With our R-LWE cipher suites integrated into the Open SSL library and using the Apache web server on a 2-core desktop computer, we could serve 506 RLWE-ECDSA-AES128-GCM-SHA256 HTTPS connections per second for a 10 KiB payload. Compared to elliptic curve Diffie-Hellman, this means an 8 KiB increased handshake size and a reduction in throughput of only 21%. This demonstrates that provably secure post-quantum key-exchange can already be considered practical.

The impact of orbital errors on the estimation of satellite clock errors and PPP

Precise satellite orbit and clocks are essential for providing high accuracy real-time PPP (Precise Point Positioning) service. However, by treating the predicted orbits as fixed, the orbital errors may be partially assimilated by the estimated satellite clock and hence impact the positioning solutions. This paper presents the impact analysis of errors in radial and tangential orbital components on the estimation of satellite clocks and PPP through theoretical study and experimental evaluation. The relationship between the compensation of the orbital errors by the satellite clocks and the satellite-station geometry is discussed in details. Based on the satellite clocks estimated with regional station networks of different sizes (∼100, ∼300, ∼500 and ∼700 km in radius), results indicated that the orbital errors compensated by the satellite clock estimates reduce as the size of the network increases. An interesting regional PPP mode based on the broadcast ephemeris and the corresponding estimated satellite clocks is proposed and evaluated through the numerical study. The impact of orbital errors in the broadcast ephemeris has shown to be negligible for PPP users in a regional network of a radius of ∼300 km, with positioning RMS of about 1.4, 1.4 and 3.7 cm for east, north and up component in the post-mission kinematic mode, comparable with 1.3, 1.3 and 3.6 cm using the precise orbits and the corresponding estimated clocks. Compared with the DGPS and RTK positioning, only the estimated satellite clocks are needed to be disseminated to PPP users for this approach. It can significantly alleviate the communication burdens and therefore can be beneficial to the real time applications.

Errors in the knee joint forces and moments during gait depending on the foot and knee prosthetic components

Background Previously studies showed that inverse dynamics based on motion analysis and force-plate is inaccurate compared to direct measurements for individuals with transfemoral amputation (TFA). Indeed, direct measurements can appropriately take into account the absorption at the prosthetic foot and the resistance at the prosthetic knee. [1-3] However, these studies involved only a passive prosthetic knee. Aim The objective of the present study was to investigate if different types of prosthetic feet and knees can exhibit different levels of error in the knee joint forces and moments. Method Three trials of walking at self-selected speed were analysed for 9 TFAs (7 males and 2 females, 47±9 years old, 1.76±0.1 m 79±17 kg) with a motion analysis system (Qualisys, Goteborg, Sweden), force plates (Kitsler, Winterthur, Switzerland) and a multi-axial transducer (JR3, Woodland, USA) mounted above the prosthetic knee [1-17]. TFAs were all fitted with an osseointegrated implant system. The prostheses included different type of foot (N=5) and knee (N=3) components. The root mean square errors (RMSE) between direct measurements and the knee joint forces and moments estimated by inverse dynamics were computed for stance and swing phases of gait and expressed as a percentage of the measured amplitudes. A one-way Kruskal-Wallis ANOVA was performed (Statgraphics, Levallois-Perret, France) to analyse the effects of the prosthetic components on the RMSEs. Cross-effects and post-hoc tests were not analysed in this study. Results A significant effect (*) was found for the type of prosthetic foot on anterior-posterior force during swing (p=0.016), lateral-medial force during stance (p=0.009), adduction-abduction moment during stance (p=0.038), internal-external rotation moment during stance (p=0.014) and during swing (p=0.006), and flexion-extension moment during stance (p = 0.035). A significant effect (#) was found for the type of prosthetic knee on anterior-posterior force during swing (p=0.018) and adduction-abduction moment during stance (p=0.035). Discussion & Conclusion The RMSEs were larger during swing than during stance. It is because the errors on accelerations (as derived from motion analysis) become substantial with respect to the external loads. Thus, inverse dynamics during swing should be analysed with caution because the mean RMSEs are close to 50%. Conversely, there were fewer effects of the prosthetic components on RMSE during swing than during stance and, accordingly, fewer effects due to knees than feet. Thus, inverse dynamics during stance should be used with caution for comparison of different prosthetic components.

Standard errors and covariance matrices for smoothed rank estimators

A 'pseudo-Bayesian' interpretation of standard errors yields a natural induced smoothing of statistical estimating functions. When applied to rank estimation, the lack of smoothness which prevents standard error estimation is remedied. Efficiency and robustness are preserved, while the smoothed estimation has excellent computational properties. In particular, convergence of the iterative equation for standard error is fast, and standard error calculation becomes asymptotically a one-step procedure. This property also extends to covariance matrix calculation for rank estimates in multi-parameter problems. Examples, and some simple explanations, are given.

The impact of inter-fraction set-up errors on the probability of pulmonary and cardiac complication in left-sided breast cancer patients

Purpose This study evaluated the impact of patient set-up errors on the probability of pulmonary and cardiac complications in the irradiation of left-sided breast cancer. Methods and Materials Using the CMS XiO Version 4.6 (CMS Inc., St Louis, MO) radiotherapy planning system's NTCP algorithm and the Lyman -Kutcher-Burman (LKB) model, we calculated the DVH indices for the ipsilateral lung and heart and the resultant normal tissue complication probabilities (NTCP) for radiation-induced pneumonitis and excess cardiac mortality in 12 left-sided breast cancer patients. Results Isocenter shifts in the posterior direction had the greatest effect on the lung V20, heart V25, mean and maximum doses to the lung and the heart. Dose volume histograms (DVH) results show that the ipsilateral lung V20 tolerance was exceeded in 58% of the patients after 1cm posterior shifts. Similarly, the heart V25 tolerance was exceeded after 1cm antero-posterior and left-right isocentric shifts in 70% of the patients. The baseline NTCPs for radiation-induced pneumonitis ranged from 0.73% - 3.4% with a mean value of 1.7%. The maximum reported NTCP for radiation-induced pneumonitis was 5.8% (mean 2.6%) after 1cm posterior isocentric shift. The NTCP for excess cardiac mortality were 0 % in 100% of the patients (n=12) before and after setup error simulations. Conclusions Set-up errors in left sided breast cancer patients have a statistically significant impact on the Lung NTCPs and DVH indices. However, with a central lung distance of 3cm or less (CLD <3cm), and a maximum heart distance of 1.5cm or less (MHD<1.5cm), the treatment plans could tolerate set-up errors of up to 1cm without any change in the NTCP to the heart.

Look-alike, sound-alike medication errors: A novel case concerning a Slow-Na, Slow-K prescribing error

A 59-year-old man was mistakenly prescribed Slow-Na instead of Slow-K due to incorrect selection from a drop-down list in the prescribing software. This error was identified by a pharmacist during a home medicine review (HMR) before the patient began taking the supplement. The reported error emphasizes the need for vigilance due to the emergence of novel look-alike, sound-alike (LASA) drug pairings. This case highlights the important role of pharmacists in medication safety.

Medication errors in residential aged care facilities: A distributed cognition analysis of the information exchange process

Background Medication safety is a pressing concern for residential aged care facilities (RACFs). Retrospective studies in RACF settings identify inadequate communication between RACFs, doctors, hospitals and community pharmacies as the major cause of medication errors. Existing literature offers limited insight about the gaps in the existing information exchange process that may lead to medication errors. The aim of this research was to explicate the cognitive distribution that underlies RACF medication ordering and delivery to identify gaps in medication-related information exchange which lead to medication errors in RACFs. Methods The study was undertaken in three RACFs in Sydney, Australia. Data were generated through ethnographic field work over a period of five months (May–September 2011). Triangulated analysis of data primarily focused on examining the transformation and exchange of information between different media across the process. Results The findings of this study highlight the extensive scope and intense nature of information exchange in RACF medication ordering and delivery. Rather than attributing error to individual care providers, the explication of distributed cognition processes enabled the identification of gaps in three information exchange dimensions which potentially contribute to the occurrence of medication errors namely: (1) design of medication charts which complicates order processing and record keeping (2) lack of coordination mechanisms between participants which results in misalignment of local practices (3) reliance on restricted communication bandwidth channels mainly telephone and fax which complicates the information processing requirements. The study demonstrates how the identification of these gaps enhances understanding of medication errors in RACFs. Conclusions Application of the theoretical lens of distributed cognition can assist in enhancing our understanding of medication errors in RACFs through identification of gaps in information exchange. Understanding the dynamics of the cognitive process can inform the design of interventions to manage errors and improve residents’ safety.

A study of errors that can occur when weighing infants

When infants are weighed at well baby or infant welfare clinics, the weight change from one visit to the next is used as a guide to the welfare of the child. Infant welfare clinic nurses are expert clinicians who use weight measurements as a rough indicator of well-being only, as it is well known by them that these measurements are fraught with error. This paper calculates the amount of error which was found in repeated tests of weights of infants, and in the weight changes brought about by biological variation. As a result, it is recommended that babies under nine months of age be weighed at clinic visits no less than a fortnight apart, and older infants, at least one month apart. If they are weighed more often, then the weight changes detected will be less than the amount of error which affects the measurements.

Clinical reasoning: The relative contribution of identification, interpretation and hypothesis errors to misdiagnosis

The aim of this study was to identify and describe the types of errors in clinical reasoning that contribute to poor diagnostic performance at different levels of medical training and experience. Three cohorts of subjects, second- and fourth- (final) year medical students and a group of general practitioners, completed a set of clinical reasoning problems. The responses of those whose scores fell below the 25th centile were analysed to establish the stage of the clinical reasoning process - identification of relevant information, interpretation or hypothesis generation - at which most errors occurred and whether this was dependent on problem difficulty and level of medical experience. Results indicate that hypothesis errors decrease as expertise increases but that identification and interpretation errors increase. This may be due to inappropriate use of pattern recognition or to failure of the knowledge base. Furthermore, although hypothesis errors increased in line with problem difficulty, identification and interpretation errors decreased. A possible explanation is that as problem difficulty increases, subjects at all levels of expertise are less able to differentiate between relevant and irrelevant clinical features and so give equal consideration to all information contained within a case. It is concluded that the development of clinical reasoning in medical students throughout the course of their pre-clinical and clinical education may be enhanced by both an analysis of the clinical reasoning process and a specific focus on each of the stages at which errors commonly occur.

Toward Self-Organising Service Communities

This paper discusses a framework in which catalog service communities are built, linked for interaction, and constantly monitored and adapted over time. A catalog service community (represented as a peer node in a peer-to-peer network) in our system can be viewed as domain specific data integration mediators representing the domain knowledge and the registry information. The query routing among communities is performed to identify a set of data sources that are relevant to answering a given query. The system monitors the interactions between the communities to discover patterns that may lead to restructuring of the network (e.g., irrelevant peers removed, new relationships created, etc.).

Effective Cell-Centred Time-Domain Maxwell's Equations Numerical Solvers

This research work analyses techniques for implementing a cell-centred finite-volume time-domain (ccFV-TD) computational methodology for the purpose of studying microwave heating. Various state-of-the-art spatial and temporal discretisation methods employed to solve Maxwell's equations on multidimensional structured grid networks are investigated, and the dispersive and dissipative errors inherent in those techniques examined. Both staggered and unstaggered grid approaches are considered. Upwind schemes using a Riemann solver and intensity vector splitting are studied and evaluated. Staggered and unstaggered Leapfrog and Runge-Kutta time integration methods are analysed in terms of phase and amplitude error to identify which method is the most accurate and efficient for simulating microwave heating processes. The implementation and migration of typical electromagnetic boundary conditions. from staggered in space to cell-centred approaches also is deliberated. In particular, an existing perfectly matched layer absorbing boundary methodology is adapted to formulate a new cell-centred boundary implementation for the ccFV-TD solvers. Finally for microwave heating purposes, a comparison of analytical and numerical results for standard case studies in rectangular waveguides allows the accuracy of the developed methods to be assessed.

Brucite Microbialites in Living Coral Skeletons: Indicators of Extreme Microenvironments in Shallow-Marine Settings

Brucite [Mg(OH)2] microbialites occur in vacated interseptal spaces of living scleractinian coral colonies (Acropora, Pocillopora, Porites) from subtidal and intertidal settings in the Great Barrier Reef, Australia, and subtidal Montastraea from the Florida Keys, United States. Brucite encrusts microbial filaments of endobionts (i.e., fungi, green algae, cyanobacteria) growing under organic biofilms; the brucite distribution is patchy both within interseptal spaces and within coralla. Although brucite is undersaturated in seawater, its precipitation was apparently induced in the corals by lowered pCO2 and increased pH within microenvironments protected by microbial biofilms. The occurrence of brucite in shallow-marine settings highlights the importance of microenvironments in the formation and early diagenesis of marine carbonates. Significantly, the brucite precipitates discovered in microenvironments in these corals show that early diagenetic products do not necessarily reflect ambient seawater chemistry. Errors in environmental interpretation may arise where unidentified precipitates occur in microenvironments in skeletal carbonates that are subsequently utilized as geochemical seawater proxies.

Automated Feedback for 'Fill in the Gap' Programming Exercises

Timely feedback is a vital component in the learning process. It is especially important for beginner students in Information Technology since many have not yet formed an effective internal model of a computer that they can use to construct viable knowledge. Research has shown that learning efficiency is increased if immediate feedback is provided for students. Automatic analysis of student programs has the potential to provide immediate feedback for students and to assist teaching staff in the marking process. This paper describes a “fill in the gap” programming analysis framework which tests students’ solutions and gives feedback on their correctness, detects logic errors and provides hints on how to fix these errors. Currently, the framework is being used with the Environment for Learning to Programming (ELP) system at Queensland University of Technology (QUT); however, the framework can be integrated into any existing online learning environment or programming Integrated Development Environment (IDE)