PROGNOSTIC FACTORS PREDICTING FUNCTIONAL OUTCOME AT FOUR MONTHS FOLLOWING ACUTE ANKLE SPRAIN

PROGNOSTIC FACTORS PREDICTING FUNCTIONAL OUTCOME AT FOUR MONTHS FOLLOWING ACUTE ANKLE SPRAINBleakley C.M.1, O'Connor S.R.1, Tully M.A.2, Rocke L.G.3, MacAuley D.C.1, Bradbury I.4, Keegan S.4, McDonough S.M.11University of Ulster, Health & Rehabilitation Sciences Research Institute, Newtownabbey, United Kingdom, 2Queen's University, UKCRC Centre of Excellence for Public Health (NI), Belfast, United Kingdom, 3Royal Victoria Hospital, Department of Emergency Medicine, Belfast, United Kingdom, 4Frontier Science (Scotland), Kincraig, Inverness-shire, United KingdomPurpose: To identify clinically relevant factors assessed following acute ankle sprain that predict functional recovery at four months post-injury.Relevance: Ankle sprains are one of the most common musculoskeletal injuries with an estimated 5000 new cases occurring each day in the United Kingdom. In the acute phase, ankle sprains may be associated with pain and loss of function. In the longer-term there is a risk of residual problems including chronic pain or reinjury. Few studies have sought to examine factors associated with a poor long-term prognosis.Participants: 101 patients (Age: Mean (SD) 25.9 (7.9) years; Body Mass Index (BMI): 25.3 (3.5) kg/m2) with an acute grade 1 or 2 ankle sprain attending an accident and emergency department or sports injury clinic. Exclusion criteria included complete (grade 3) rupture of the ankle ligament complex, bony ankle injury or multiple injuries.Methods: Participants were allocated as part of a randomised controlled trial to an accelerated intervention incorporating intermittent ice and early therapeutic exercise or a standard protection, rest, ice, compression, and elevation intervention for one week. Treatment was then standardised in both groups and consisted of ankle rehabilitation exercises focusing on muscle strengthening, neuromuscular training, and sports specific functional exercises for a period of approximately four to six weeks. On initial assessment age, gender, mechanism of injury, presence of an audible pop or snap and the presence of contact during the injury were recorded. The following factors were also recorded at baseline and at one and four weeks post-injury: weight-bearing dorsi-flexion test, lateral hop test, presence of medial pain on palpation and a positive impingement sign. Functional status was assessed using the Karlsson score at baseline, at week four and at four months. Reinjury rates were recorded throughout the intervention phase and at four months.Analysis: A mixed between-within subjects analysis of variance (ANOVA) was used to determine the effect of each factor on functional status at week four and at four months. Significance was set at a Bonferroni adjusted level of 0.0125 (0.05/4).Results: Eighty-five participants (84%) were available at final follow-up assessment. Pain on weight-bearing dorsi-flexion and lateral hop tests at week four were both associated with a lower functional score at four months post-injury (P = 0.011 and P = 0.001). No other significant interactions were observed at any other timepoint (baseline or week one). There were only two reinjuries within the four month follow-up period with a further two reported at approximately six months post-injury. We were therefore unable to determine whether any factors were associated with an increased risk of reinjury.Conclusions: Potential prognostic factors on initial or early examination after acute ankle sprain did not help predict functional recovery at four months post-injury. However, pain on weight-bearing dorsi-flexion and lateral hop tests observed at four weeks were associated with a slower rate of recovery.Implications: Some clinical tests may help identify patients at risk of poor functional recovery after acute ankle sprain. However, further work is required to examine factors which may be predictive on initial assessment.Key-words: 1. Prognostic factors 2. Recovery 3. Ankle sprainFunding acknowledgements: Physiotherapy Research Foundation, Chartered Society of Physiotherapy, Strategic Priority Fund; Department of Employment and Learning, Northern Ireland.Ethics approval: Office for Research Ethics Committee (UK).

A Generalized Fuzzy Linguistic Model for Predicting Component Concentrations in an Optical Gas Sensing System

Motivated by environmental protection concerns, monitoring the flue gas of thermal power plant is now often mandatory due to the need to ensure that emission levels stay within safe limits. Optical based gas sensing systems are increasingly employed for this purpose, with regression techniques used to relate gas optical absorption spectra to the concentrations of specific gas components of interest (NOx, SO2 etc.). Accurately predicting gas concentrations from absorption spectra remains a challenging problem due to the presence of nonlinearities in the relationships and the high-dimensional and correlated nature of the spectral data. This article proposes a generalized fuzzy linguistic model (GFLM) to address this challenge. The GFLM is made up of a series of “If-Then” fuzzy rules. The absorption spectra are input variables in the rule antecedent. The rule consequent is a general nonlinear polynomial function of the absorption spectra. Model parameters are estimated using least squares and gradient descent optimization algorithms. The performance of GFLM is compared with other traditional prediction models, such as partial least squares, support vector machines, multilayer perceptron neural networks and radial basis function networks, for two real flue gas spectral datasets: one from a coal-fired power plant and one from a gas-fired power plant. The experimental results show that the generalized fuzzy linguistic model has good predictive ability, and is competitive with alternative approaches, while having the added advantage of providing an interpretable model.

Predicting polarization signatures for double-detonation and delayed-detonation models of Type Ia supernovae

Calculations of synthetic spectropolarimetry are one means to test multidimensional explosion models for Type Ia supernovae. In a recent paper, we demonstrated that the violent merger of a 1.1 and 0.9 M⊙ white dwarf binary system is too asymmetric to explain the low polarization levels commonly observed in normal Type Ia supernovae. Here, we present polarization simulations for two alternative scenarios: the sub-Chandrasekhar mass double-detonation and the Chandrasekhar mass delayed-detonation model. Specifically, we study a 2D double-detonation model and a 3D delayed-detonation model, and calculate polarization spectra for multiple observer orientations in both cases. We find modest polarization levels (<1 per cent) for both explosion models. Polarization in the continuum peaks at ∼0.1–0.3 per cent and decreases after maximum light, in excellent agreement with spectropolarimetric data of normal Type Ia supernovae. Higher degrees of polarization are found across individual spectral lines. In particular, the synthetic Si II λ6355 profiles are polarized at levels that match remarkably well the values observed in normal Type Ia supernovae, while the low degrees of polarization predicted across the O I λ7774 region are consistent with the non-detection of this feature in current data. We conclude that our models can reproduce many of the characteristics of both flux and polarization spectra for well-studied Type Ia supernovae, such as SN 2001el and SN 2012fr. However, the two models considered here cannot account for the unusually high level of polarization observed in extreme cases such as SN 2004dt.

A generalized fuzzy linguistic model for predicting component concentrations in an optical gas sensing system

Motivated by environmental protection concerns, monitoring the flue gas of thermal power plant is now often mandatory due to the need to ensure that emission levels stay within safe limits. Optical based gas sensing systems are increasingly employed for this purpose, with regression techniques used to relate gas optical absorption spectra to the concentrations of specific gas components of interest (NO_x, SO₂ etc.). Accurately predicting gas concentrations from absorption spectra remains a challenging problem due to the presence of nonlinearities in the relationships and the high-dimensional and correlated nature of the spectral data. This article proposes a generalized fuzzy linguistic model (GFLM) to address this challenge. The GFLM is made up of a series of “If-Then” fuzzy rules. The absorption spectra are input variables in the rule antecedent. The rule consequent is a general nonlinear polynomial function of the absorption spectra. Model parameters are estimated using least squares and gradient descent optimization algorithms. The performance of GFLM is compared with other traditional prediction models, such as partial least squares, support vector machines, multilayer perceptron neural networks and radial basis function networks, for two real flue gas spectral datasets: one from a coal-fired power plant and one from a gas-fired power plant. The experimental results show that the generalized fuzzy linguistic model has good predictive ability, and is competitive with alternative approaches, while having the added advantage of providing an interpretable model.