Pitfalls in the use of kappa when interpreting agreement between multiple raters in reliability studies

OBJECTIVE To compare different reliability coefficients (exact agreement, and variations of the kappa (generalised, Cohen's and Prevalence Adjusted and Biased Adjusted (PABAK))) for four physiotherapists conducting visual assessments of scapulae. DESIGN Inter-therapist reliability study. SETTING Research laboratory. PARTICIPANTS 30 individuals with no history of neck or shoulder pain were recruited with no obvious significant postural abnormalities. MAIN OUTCOME MEASURES Ratings of scapular posture were recorded in multiple biomechanical planes under four test conditions (at rest, and while under three isometric conditions) by four physiotherapists. RESULTS The magnitude of discrepancy between the two therapist pairs was 0.04 to 0.76 for Cohen's kappa, and 0.00 to 0.86 for PABAK. In comparison, the generalised kappa provided a score between the two paired kappa coefficients. The difference between mean generalised kappa coefficients and mean Cohen's kappa (0.02) and between mean generalised kappa and PABAK (0.02) were negligible, but the magnitude of difference between the generalised kappa and paired kappa within each plane and condition was substantial; 0.02 to 0.57 for Cohen's kappa and 0.02 to 0.63 for PABAK, respectively. CONCLUSIONS Calculating coefficients for therapist pairs alone may result in inconsistent findings. In contrast, the generalised kappa provided a coefficient close to the mean of the paired kappa coefficients. These findings support an assertion that generalised kappa may lead to a better representation of reliability between three or more raters and that reliability studies only calculating agreement between two raters should be interpreted with caution. However, generalised kappa may mask more extreme cases of agreement (or disagreement) that paired comparisons may reveal.

Secure stream cipher initialisation processes

Stream ciphers are symmetric key cryptosystems that are used commonly to provide confidentiality for a wide range of applications; such as mobile phone, pay TV and Internet data transmissions. This research examines the features and properties of the initialisation processes of existing stream ciphers to identify flaws and weaknesses, then presents recommendations to improve the security of future cipher designs. This research investigates well-known stream ciphers: A5/1, Sfinks and the Common Scrambling Algorithm Stream Cipher (CSA-SC). This research focused on the security of the initialisation process. The recommendations given are based on both the results in the literature and the work in this thesis.

Burn healing is dependent on burn site : a quantitative analysis from a porcine burn model

This retrospective review examines healing in different sites on a porcine burn model; 24 pairs of burns on 18 pigs from other animal trials were selected for analysis. Each pair of burns was located on the either the cranial or the caudal part of the thoracic ribs region, on the same side of the animal. The burns were 40-50 cm(2) in size and of uniform deep-dermal partial thickness. Caudal burns healed significantly better than cranial burns, demonstrated by earlier closure of wounds, less scar formation and better cosmesis. To our knowledge, this is the first detailed study reporting that burn healing is affected by location on a porcine burn model. We recommend that similar symmetrical burns should be used for future comparative assessments of burn healing.

Improved rural health through a collaborative model of care – qualitative findings

INTRODUCTION Health disparity between urban and rural regions in Australia is well-documented. In the Wheatbelt catchments of Western Australia there is higher incidence and rate of avoidable hospitalisation for chronic diseases. Structured care approach to chronic illnesses is not new but the focus has been on single disease state. A recent ARC Discovery Project on general practice nurse-led chronic disease management of diabetes, hypertension and stable ischaemic heart disease reported improved communication and better medical administration.[1] In our study we investigated the sustainability of such a multi-morbidities general practice –led collaborative model of care in rural Australia. METHODS A QUAN(qual) design was utilised. Eight pairs of rural general practices were matched. Inclusion criteria used were >18 years and capable of giving informed consent, at least one identified risk factor or diagnosed with chronic conditions. Patients were excluded if deemed medically unsuitable. A comprehensive care plan was formulated by the respective general practice nurse in consultation with the treating General Practitioner (GP) and patient based on the individual’s readiness to change, and was informed by available local resource. A case management approach was utilised. Shediaz-Rizkallah and Lee’s conceptual framework on sustainability informed our evaluation.[2] Our primary outcome on measures of sustainability was reduction in avoidable hospitalisation. Secondary outcomes were patients and practitioners acceptance and satisfaction, and changes to pre-determined interim clinical and process outcomes. RESULTS The qualitative interviews highlighted the community preference for a ‘sustainable’ local hospital in addition to general practice. Costs, ease of access, low prioritisation of self chronic care, workforce turnover and perception of losing another local resource if underutilised influenced the respondents’ decision to present at local hospital for avoidable chronic diseases regardless. CONCLUSIONS Despite the pragmatic nature of rural general practice in Australia, the sustainability of chronic multi-morbidities management in general practice require efficient integration of primary-secondary health care and consideration of other social determinants of health. What this study adds: What is already known on this subject: Structured approach to chronic disease management is not new and has been shown to be effective for reducing hospitalisation. However, the focus has been on single disease state. What does this study add: Sustainability of collaborative model of multi-morbidities care require better primary-secondary integration and consideration of social determinants of health.

Quantum direct secret sharing with efficient eavesdropping-check and authentication based on distributed fountain codes

We propose to use a simple and effective way to achieve secure quantum direct secret sharing. The proposed scheme uses the properties of fountain codes to allow a realization of the physical conditions necessary for the implementation of no-cloning principle for eavesdropping-check and authentication. In our scheme, to achieve a variety of security purposes, nonorthogonal state particles are inserted in the transmitted sequence carrying the secret shares to disorder it. However, the positions of the inserted nonorthogonal state particles are not announced directly, but are obtained by sending degrees and positions of a sequence that are pre-shared between Alice and each Bob. Moreover, they can confirm that whether there exists an eavesdropper without exchanging classical messages. Most importantly, without knowing the positions of the inserted nonorthogonal state particles and the sequence constituted by the first particles from every EPR pair, the proposed scheme is shown to be secure.

Security evaluation of Rakaposhi Stream Cipher

Rakaposhi is a synchronous stream cipher, which uses three main components: a non-linear feedback shift register (NLFSR), a dynamic linear feedback shift register (DLFSR) and a non-linear filtering function (NLF). NLFSR consists of 128 bits and is initialised by the secret key K. DLFSR holds 192 bits and is initialised by an initial vector (IV). NLF takes 8-bit inputs and returns a single output bit. The work identifies weaknesses and properties of the cipher. The main observation is that the initialisation procedure has the so-called sliding property. The property can be used to launch distinguishing and key recovery attacks. The distinguisher needs four observations of the related (K,IV) pairs. The key recovery algorithm allows to discover the secret key K after observing 29 pairs of (K,IV). Based on the proposed related-key attack, the number of related (K,IV) pairs is 2(128 + 192)/4 pairs. Further the cipher is studied when the registers enter short cycles. When NLFSR is set to all ones, then the cipher degenerates to a linear feedback shift register with a non-linear filter. Consequently, the initial state (and Secret Key and IV) can be recovered with complexity 263.87. If DLFSR is set to all zeros, then NLF reduces to a low non-linearity filter function. As the result, the cipher is insecure allowing the adversary to distinguish it from a random cipher after 217 observations of keystream bits. There is also the key recovery algorithm that allows to find the secret key with complexity 2 54.

Career mentoring : a career development-enhancing strategy for university students in transition to employment

In 1992 the Queensland University of Technology (QUT) Convocation initiated the Mentor Scheme to assist students to better prepare for the transition to employment. The scheme has developed over the past six years and in 1998 more than 130 pairs of employers and students from 12 different disciplines participated in it. Evaluations indicate the positive impact the scheme has made on both mentors and mentees, particularly in enhancing mentees' career development.

Causal behavioural profiles : efficient computation, applications, and evaluation

Analysis of behavioural consistency is an important aspect of software engineering. In process and service management, consistency verification of behavioural models has manifold applications. For instance, a business process model used as system specification and a corresponding workflow model used as implementation have to be consistent. Another example would be the analysis to what degree a process log of executed business operations is consistent with the corresponding normative process model. Typically, existing notions of behaviour equivalence, such as bisimulation and trace equivalence, are applied as consistency notions. Still, these notions are exponential in computation and yield a Boolean result. In many cases, however, a quantification of behavioural deviation is needed along with concepts to isolate the source of deviation. In this article, we propose causal behavioural profiles as the basis for a consistency notion. These profiles capture essential behavioural information, such as order, exclusiveness, and causality between pairs of activities of a process model. Consistency based on these profiles is weaker than trace equivalence, but can be computed efficiently for a broad class of models. In this article, we introduce techniques for the computation of causal behavioural profiles using structural decomposition techniques for sound free-choice workflow systems if unstructured net fragments are acyclic or can be traced back to S- or T-nets. We also elaborate on the findings of applying our technique to three industry model collections.

Efficient computation of causal behavioural profiles using structural decomposition

Identification of behavioural contradictions is an important aspect of software engineering, in particular for checking the consistency between a business process model used as system specification and a corresponding workflow model used as implementation. In this paper, we propose causal behavioural profiles as the basis for a consistency notion, which capture essential behavioural information, such as order, exclusiveness, and causality between pairs of activities. Existing notions of behavioural equivalence, such as bisimulation and trace equivalence, might also be applied as consistency notions. Still, they are exponential in computation. Our novel concept of causal behavioural profiles provides a weaker behavioural consistency notion that can be computed efficiently using structural decomposition techniques for sound free-choice workflow systems if unstructured net fragments are acyclic or can be traced back to S- or T-nets.

CoRE : a context-aware relation extraction method for relation completion

We identify relation completion (RC) as one recurring problem that is central to the success of novel big data applications such as Entity Reconstruction and Data Enrichment. Given a semantic relation, RC attempts at linking entity pairs between two entity lists under the relation. To accomplish the RC goals, we propose to formulate search queries for each query entity α based on some auxiliary information, so that to detect its target entity β from the set of retrieved documents. For instance, a pattern-based method (PaRE) uses extracted patterns as the auxiliary information in formulating search queries. However, high-quality patterns may decrease the probability of finding suitable target entities. As an alternative, we propose CoRE method that uses context terms learned surrounding the expression of a relation as the auxiliary information in formulating queries. The experimental results based on several real-world web data collections demonstrate that CoRE reaches a much higher accuracy than PaRE for the purpose of RC.

Genetic aspects of the organization of legumin genes in pea

We have compared physical and genetic maps of the region around the legJ gene in pea. In this vicinity there are four B-type legumin genes, arranged as two close pairs. The detection of a recombination event within this gene cluster allows the orientation of this group of genes within the surrounding linkage group to be determined. The relationship between physical and genetic distances in this region is discussed, as are the implications of this for relating physical and genetic maps elsewhere in the pea genome.

Effects of small field output factors on IMRT optimisation and dose calculation

Introduction Given the known challenges of obtaining accurate measurements of small radiation fields, and the increasing use of small field segments in IMRT beams, this study examined the possible effects of referencing inaccurate field output factors in the planning of IMRT treatments. Methods This study used the Brainlab iPlan treatment planning system to devise IMRT treatment plans for delivery using the Brainlab m3 microMLC (Brainlab, Feldkirchen, Germany). Four pairs of sample IMRT treatments were planned using volumes, beams and prescriptions that were based on a set of test plans described in AAPM TG 119’s recommendations for the commissioning of IMRT treatment planning systems [1]: • C1, a set of three 4 cm volumes with different prescription doses, was modified to reduce the size of the PTV to 2 cm across and to include an OAR dose constraint for one of the other volumes. • C2, a prostate treatment, was planned as described by the TG 119 report [1]. • C3, a head-and-neck treatment with a PTV larger than 10 cm across, was excluded from the study. • C4, an 8 cm long C-shaped PTV surrounding a cylindrical OAR, was planned as described in the TG 119 report [1] and then replanned with the length of the PTV reduced to 4 cm. Both plans in each pair used the same beam angles, collimator angles, dose reference points, prescriptions and constraints. However, one of each pair of plans had its beam modulation optimisation and dose calculation completed with reference to existing iPlan beam data and the other had its beam modulation optimisation and dose calculation completed with reference to revised beam data. The beam data revisions consisted of increasing the field output factor for a 0.6 9 0.6 cm2 field by 17 % and increasing the field output factor for a 1.2 9 1.2 cm2 field by 3 %. Results The use of different beam data resulted in different optimisation results with different microMLC apertures and segment weightings between the two plans for each treatment, which led to large differences (up to 30 % with an average of 5 %) between reference point doses in each pair of plans. These point dose differences are more indicative of the modulation of the plans than of any clinically relevant changes to the overall PTV or OAR doses. By contrast, the maximum, minimum and mean doses to the PTVs and OARs were smaller (less than 1 %, for all beams in three out of four pairs of treatment plans) but are more clinically important. Of the four test cases, only the shortened (4 cm) version of TG 119’s C4 plan showed substantial differences between the overall doses calculated in the volumes of interest using the different sets of beam data and thereby suggested that treatment doses could be affected by changes to small field output factors. An analysis of the complexity of this pair of plans, using Crowe et al.’s TADA code [2], indicated that iPlan’s optimiser had produced IMRT segments comprised of larger numbers of small microMLC leaf separations than in the other three test cases. Conclusion: The use of altered small field output factors can result in substantially altered doses when large numbers of small leaf apertures are used to modulate the beams, even when treating relatively large volumes.

An instrument to assess the obesogenic environment of child care centers

OBJECTIVES To describe protocol and interobserver agreements of an instrument to evaluate nutrition and physical activity environments at child care. METHODS Interobserver data were collected from 9 child care centers, through direct observation and document review (17 observer pairs). RESULTS Mean agreement between observer pairs was 87.26% and 79.29% for the observation and document review, respectively. Items with lower agreement were primarily staff behavior, counting across the day/week, and policy classifications. CONCLUSIONS Although some revisions are required, the interobserver agreement for the environment and policy assessment and observation (EPAO instrument) appears to be quite good for assessing the nutrition and physical activity environment of child care centers.

Comparison of surveys used to measure physical activity

Objective To compare the level of agreement in results obtained from four physical activity (PA) measurement instruments that are in use in Australia and around the world. Methods 1,280 randomly selected participants answered two sets of PA questions by telephone. 428 answered the Active Australia (AA) and National Health Surveys, 427 answered the AA and CDC Behavioural Risk Factor Surveillance System surveys (BRFSS), and 425 answered the AA survey and the short International Physical Activity Questionnaire (IPAQ). Results Among the three pairs of survey items, the difference in mean total PA time was lowest when the AA and NHS items were asked (difference=24) (SE:17) minutes, compared with 144 (SE:21) mins for AA/BRFSS and 406 (SE:27) mins for AA/IPAQ). Correspondingly, prevalence estimates for 'sufficiently active' were similar for AA and NHS (56% and 55% respectively), but about 10% higher when BRFSS data were used, and about 26% higher when the IPAQ items were used, compared with estimates from the AA survey. Conclusions The findings clearly demonstrate that there are large differences in reported PA times and hence in prevalence estimates of 'sufficient activity' from these four measures. Implications It is important to consistently use the same survey for population monitoring purposes. As the AA survey has now been used three times in national surveys, its continued use for population surveys is recommended so that trend data ever a longer period of time can be established.

Detection of object onset and offset in naturalistic scenes

The present study was conducted to investigate whether ob- servers are equally prone to overlook any kinds of visual events in change blindness. Capitalizing on the finding from visual search studies that abrupt appearance of an object effectively captures observers' attention, the onset of a new object and the offset of an existing object were contrasted regarding their detectability when they occurred in a naturalistic scene. In an experiment, participants viewed a series of photograph pairs in which layouts of seven or eight objects were depicted. One object either appeared in or disappeared from the layout, and participants tried to detect this change. Results showed that onsets were detected more quickly than offsets, while they were detected with equivalent ac- curacy. This suggests that the primacy of onset over offset is a robust phenomenon that likely makes onsets more resistant to change blindness under natural viewing conditions.