Random Gabor based templates for facial expression recognition in images with facial occlusion

Robust facial expression recognition (FER) under occluded face conditions is challenging. It requires robust algorithms of feature extraction and investigations into the effects of different types of occlusion on the recognition performance to gain insight. Previous FER studies in this area have been limited. They have spanned recovery strategies for loss of local texture information and testing limited to only a few types of occlusion and predominantly a matched train-test strategy. This paper proposes a robust approach that employs a Monte Carlo algorithm to extract a set of Gabor based part-face templates from gallery images and converts these templates into template match distance features. The resulting feature vectors are robust to occlusion because occluded parts are covered by some but not all of the random templates. The method is evaluated using facial images with occluded regions around the eyes and the mouth, randomly placed occlusion patches of different sizes, and near-realistic occlusion of eyes with clear and solid glasses. Both matched and mis-matched train and test strategies are adopted to analyze the effects of such occlusion. Overall recognition performance and the performance for each facial expression are investigated. Experimental results on the Cohn-Kanade and JAFFE databases demonstrate the high robustness and fast processing speed of our approach, and provide useful insight into the effects of occlusion on FER. The results on the parameter sensitivity demonstrate a certain level of robustness of the approach to changes in the orientation and scale of Gabor filters, the size of templates, and occlusions ratios. Performance comparisons with previous approaches show that the proposed method is more robust to occlusion with lower reductions in accuracy from occlusion of eyes or mouth.

Comparison of high sensitivity troponin T and I assays in the diagnosis of non-ST elevation acute myocardial infarction in emergency patients with chest pain

Objectives: Concentrations of troponin measured with high sensitivity troponin assays are raised in a number of emergency department (ED) patients; however many are not diagnosed with acute myocardial infarction (AMI). Clinical comparisons between the early use (2 h after presentation) of high sensitivity cardiac troponin T (hs-cTnT) and I (hs-cTnI) assays for the diagnosis of AMI have not been reported. Design and methods: Early (0 h and 2 h) hs-cTnT and hs-cTnI assay results in 1571 ED patients with potential acute coronary syndrome (ACS) without ST elevation on electrocardiograph (ECG) were evaluated. The primary outcome was diagnosis of index MI adjudicated by cardiologists using the local cTnI assay results taken ≥6 h after presentation, ECGs and clinical information. Stored samples were later analysed with hs-cTnT and hs-cTnI assays. Results: The ROC analysis for AMI (204 patients; 13.0%) for hs-cTnT and hs-cTnI after 2 h was 0.95 (95% CI: 0.94–0.97) and 0.98 (95% CI: 0.97–0.99) respectively. The sensitivity, specificity, PLR, and NLR of hs-cTnT and hs-cTnI for AMI after 2 h were 94.1% (95% CI: 90.0–96.6) and 95.6% (95% CI: 91.8–97.7), 79.0% (95% CI: 76.8–81.1) and 92.5% (95% CI: 90.9–93.7), 4.48 (95% CI: 4.02–5.00) and 12.86 (95% CI: 10.51–15.31), and 0.07 (95% CI: 0.04–0.13) and 0.05 (95% CI:0.03–0.09) respectively. Conclusions: Exclusion of AMI 2 h after presentation in emergency patients with possible ACS can be achieved using hs-cTnT or hs-cTnI assays. Significant differences in specificity of these assays are relevant and if using the hs-cTnT assay, further clinical assessment in a larger proportion of patients would be required.

Quantitative sensory measures distinguish office workers with varying levels of neck pain and disability

This study was undertaken to investigate any relationship between sensory features and neck pain in female office workers using quantitative sensory measures to better understand neck pain in this group. Office workers who used a visual display monitor for more than four hours per day with varying levels of neck pain and disability were eligible for inclusion. There were 85 participants categorized according to their scores on the neck disability index (NDI): 33 with no pain (NDI < 8); 38 with mild levels of pain and disability (NDI 9–29); 14 with moderate levels of pain (NDI ⩾ 30). A fourth group of women without neck pain (n = 22) who did not work formed the control group. Measures included: thermal pain thresholds over the posterior cervical spine; pressure pain thresholds over the posterior neck, trapezius, levator scapulae and tibialis anterior muscles, and the median nerve trunk; sensitivity to vibrotactile stimulus over areas of the hand innervated by the median, ulnar and radial nerves; sympathetic vasoconstrictor response. All tests were conducted bilaterally. ANCOVA models were used to determine group differences between the means for each sensory measure. Office workers with greater self-reported neck pain demonstrated hyperalgesia to thermal stimuli over the neck, hyperalgesia to pressure stimulation over several sites tested; hypoaesthesia to vibration stimulation but no changes in the sympathetic vasoconstrictor response. There is evidence of multiple peripheral nerve dysfunction with widespread sensitivity most likely due to altered central nociceptive processing initiated and sustained by nociceptive input from the periphery.

Alterations in cervical muscle activity in functional and stressful tasks in female office workers with neck pain

This study determined differences between computer workers with varying levels of neck pain in terms of work stressors, employee strain, electromyography (EMG) amplitude and heart rate response to various tasks. Participants included 85 workers (33, no pain; 38, mild pain; 14, moderate pain) and 22 non-working controls. Work stressors evaluated were job demands, decision authority, and social support. Heart rate was recorded during three tasks: copy-typing, typing with superimposed stress and a colour word task. Measures included electromyography signals from the sternocleidomastoid (SCM), anterior scalene (AS), cervical extensor (CE) and upper trapezius (UT) muscles bilaterally. Results showed no difference between groups in work stressors or employee strain measures. Workers with and without pain had higher measured levels of EMG amplitude in SCM, AS and CE muscles during the tasks than controls (all P < 0.02). In workers with neck pain, the UT had difficulty in switching off on completion of tasks compared with controls and workers without pain. There was an increase in heart rate, perceived tension and pain and decrease in accuracy for all groups during the stressful tasks with symptomatic workers producing more typing errors than controls and workers without pain. These findings suggest an altered muscle recruitment pattern in the neck flexor and extensor muscles. Whether this is a consequence or source of the musculoskeletal disorder cannot be determined from this study. It is possible that workers currently without symptoms may be at risk of developing a musculoskeletal disorder.

Neck movement and muscle activity characteristics in female office workers with neck pain

Study Design Cross-sectional study. Objective To explore aspects of cervical musculoskeletal function in female office workers with neck pain. Summary of Background Data Evidence of physical characteristics that differentiate computer workers with and without neck pain is sparse. Patients with chronic neck pain demonstrate reduced motion and altered patterns of muscle control in the cervical flexor and upper trapezius (UT) muscles during specific tasks. Understanding cervical musculoskeletal function in office workers will better direct intervention and prevention strategies. Methods Measures included neck range of motion; superficial neck flexor muscle activity during a clinical test, the craniocerivcal flexion test; and a motor task, a unilateral muscle coordination task, to assess the activity of both the anterior and posterior neck muscles. Office workers with and without neck pain were formed into 3 groups based on their scores on the Neck Disability Index. Nonworking women without neck pain formed the control group. Surface electromyographic activity was recorded bilaterally from the sternocleidomastoid, anterior scalene (AS), cervical extensor (CE) and UT muscles. Results Workers with neck pain had reduced rotation range and increased activity of the superficial cervical flexors during the craniocervical flexion test. During the coordination task, workers with pain demonstrated greater activity in the CE muscles bilaterally. On completion of the task, the UT and dominant CE and AS muscles demonstrated an inability to relax in workers with pain. In general, there was a linear relationship between the workers’ self-reported levels of pain and disability and the movement and muscle changes. Conclusion These results are consistent with those found in other cervical musculoskeletal disorders and may represent an altered muscle recruitment strategy to stabilize the head and neck. An exercise program including motor reeducation may assist in the management of neck pain in office workers.

Associations between individual and workplace risk factors for self-reported neck pain and disability among female office workers

A cross-sectional survey of female office workers (n=333) was undertaken to determine the level of neck pain and disability (Neck Disability Index—NDI) and to explore the relationship between individual and workplace risk factors with the NDI score and the presence of pain. Workers reported nil (32%), mild (53%), moderate (14%) and severe (1%) neck pain. There were more risk factors associated with the NDI score than the presence of neck pain. The presence of neck pain was associated with a history of neck trauma (OR: 4.8), using a graduated lens (OR: 4.6), and negative affectivity (OR: 2.7) in the multiple regression model. Factors associated with higher NDI score were using the computer mouse for more than 6 h per day, higher negative affectivity, older age and an uncomfortable workstation. These results suggest that measuring the level of neck pain and disability rather than just the presence of neck pain provides more specific directives for the prevention and management of this disorder.

Contribution of individual, workplace, psychosocial and physiological factors to neck pain in female office workers

This study investigated the relative contribution of individual, workplace, psychosocial and physiological features associated with neck pain in female office workers towards developing appropriate intervention programs. Workers without disability (Neck Disability Index (NDI) score≤8, n=33); workers with neck pain and disability (NDI≥9/100, n=52) and 22 controls (women who did not work and without neck pain) participated in this study. Two logistic regression models were constructed to test the association between various measures in (1) workers with and without disability, and (2) workers without disability and controls. Measures included those found to be significantly associated with higher NDI in our previous studies: psychosocial domains; individual factors; task demands; quantitative sensory measures and measures of motor function. In the final model, higher score on negative affectivity scale (OR=4.47), greater activity in the neck flexors during cranio-cervical flexion (OR=1.44), cold hyperalgesia (OR=1.27) and longer duration of symptoms (OR=1.19) remained significantly associated with neck pain in workers. Workers without disability and controls could only be differentiated by greater muscle activity in the cervical flexors and extensors during a typing task. No psychosocial domains remained in either regression model. These results suggest that impairments in the sensory and motor system should be considered in any assessment of the office worker with neck pain and may have stronger influences on the presenting symptoms than workplace and psychosocial features.

Interactive effects from self-reported physical and psychosocial factors in the workplace on neck pain and disability in female office workers

This study explored the interaction between physical and psychosocial factors in the workplace on neck pain and disability in female computer users. A self-report survey was used to collect data on physical risk factors (monitor location, duration of time spent using the keyboard and mouse) and psychosocial domains (as assessed by the Job Content Questionnaire). The neck disability index was the outcome measure. Interactions among the physical and psychosocial factors were examined in analysis of covariance. High supervisor support, decision authority and skill discretion protect against the negative impact of (1) time spent on computer-based tasks, (2) non-optimal placement of the computer monitor, and; (3) long duration of mouse use. Office workers with greater neck pain experience a combination of high physical and low psychosocial stressors at work. Prevention and intervention strategies that target both sets of risk factors are likely to be more successful than single intervention programmes. Statement of Relevance The results of this study demonstrate that the interaction of physical and psychosocial factors in the workplace has a stronger association with neck pain and disability than the presence of either factor alone. This finding has important implications for strategies aimed at the prevention of musculoskeletal problems in office workers.

Inpatient facial expression recognition with SMS alert

Facial expression recognition (FER) has been dramatically developed in recent years, thanks to the advancements in related fields, especially machine learning, image processing and human recognition. Accordingly, the impact and potential usage of automatic FER have been growing in a wide range of applications, including human-computer interaction, robot control and driver state surveillance. However, to date, robust recognition of facial expressions from images and videos is still a challenging task due to the difficulty in accurately extracting the useful emotional features. These features are often represented in different forms, such as static, dynamic, point-based geometric or region-based appearance. Facial movement features, which include feature position and shape changes, are generally caused by the movements of facial elements and muscles during the course of emotional expression. The facial elements, especially key elements, will constantly change their positions when subjects are expressing emotions. As a consequence, the same feature in different images usually has different positions. In some cases, the shape of the feature may also be distorted due to the subtle facial muscle movements. Therefore, for any feature representing a certain emotion, the geometric-based position and appearance-based shape normally changes from one image to another image in image databases, as well as in videos. This kind of movement features represents a rich pool of both static and dynamic characteristics of expressions, which playa critical role for FER. The vast majority of the past work on FER does not take the dynamics of facial expressions into account. Some efforts have been made on capturing and utilizing facial movement features, and almost all of them are static based. These efforts try to adopt either geometric features of the tracked facial points, or appearance difference between holistic facial regions in consequent frames or texture and motion changes in loca- facial regions. Although achieved promising results, these approaches often require accurate location and tracking of facial points, which remains problematic.

From prototype to exploitation : mobile services for patients with chronic lower back pain

Many research and development projects that are carried out by firms and research institutes are technology-oriented. There is a large gap between research results, for instance in the form of prototypes, and the actual service offerings to customers. This becomes problematic when an organization wants to bring the results from such a project to the market, which will be particularly troublesome when the research results do not readily fit traditional offerings, roles and capabilities in the industry, nor the financial arrangements. In this chapter, we discuss the design of a business model for a mobile health service, starting with a research prototype that was developed for patients with chronic lower back pain, using the STOF model and method. In a number of design sessions, an initial business model was developed that identifies critical design issues that play a role in moving from prototype toward market deployment. The business model serves as a starting-point to identify and commit relevant stakeholders, and to draw up a business plan and case. This chapter is structured as follows. We begin by discussing the need for mobile health business models. Next, the research and development project on mobile health and the prototype for chronic lower back pain patients are introduced, after which the approach used to develop the business model is described, followed by a discussion of the developed mobile health business model for each of the STOF domains. We conclude with a discussion regarding the lessons that were learned with respect to the development of a business model on the basis of a prototype.

Teaching basic finance with Excel, and without pain

The five quantities of interest in elementary finance problems are present value, future value, amount of periodic payment, number of periods and the rate of compound interest per period. A recursive approach to computing each of these five quantities in a modern version of Excel, for the case of ordinary annuities, is described. The aim is to increase student understanding and build confidence in the answer obtained, and this may be achieved with only linear relationships and in cases where student knowledge of algebra is essentially zero. Annuity problems may be solved without use of logarithms and black-box intrinsic functions; these being used only as check mechanisms. The author has had success with the method at Bond University and surrounding high schools in Queensland, Australia.

Correlates of foot pain severity in adults with hallux valgus : a cross-sectional study

Background Hallux valgus (HV) is highly prevalent and associated with progressive first metatarsophalangeal joint subluxation and osteoarthritis. The link between structural HV deformity and foot pain is unclear. This study investigated possible explanatory factors surrounding foot pain in HV, including radiographic HV angle and signs of joint degeneration. Methods Participants were 60 adults (53 female) with HV aged 20 to 75 years. Participant demographics and a range of radiographic, clinical and functional measures were considered potential correlates of foot pain. Self-reported foot pain (visual analogue scales and a dichotomous definition) was considered the dependent variable. Multivariate modelling was used to determine which characteristics and measures explained pain, with univariate analyses first used to screen potential variables. Results Approximately 20 to 30% of the variance in foot pain associated with HV could be explained by patient characteristics such as poorer general health status, lower educational attainment and increased occupational physical activity levels, in combination with some dynamic physical characteristics such as hallux plantarflexion weakness and reduced force-time integral under the second metatarsal during gait. Neither increasing lateral deviation of the hallux (HV angle) nor presence of first metatarsophalangeal joint osteoarthritis was associated with foot pain. Conclusions This study shows that passive structural factors, including HV angle, do not appear to be significant correlates of foot pain intensity in HV. Our data demonstrate the importance of considering patient characteristics such as general health and physical activity levels when assessing foot pain associated with HV.

A pilot study on affective classification of facial images for emerging news topics

The proliferation of news reports published in online websites and news information sharing among social media users necessitates effective techniques for analysing the image, text and video data related to news topics. This paper presents the first study to classify affective facial images on emerging news topics. The proposed system dynamically monitors and selects the current hot (of great interest) news topics with strong affective interestingness using textual keywords in news articles and social media discussions. Images from the selected hot topics are extracted and classified into three categorized emotions, positive, neutral and negative, based on facial expressions of subjects in the images. Performance evaluations on two facial image datasets collected from real-world resources demonstrate the applicability and effectiveness of the proposed system in affective classification of facial images in news reports. Facial expression shows high consistency with the affective textual content in news reports for positive emotion, while only low correlation has been observed for neutral and negative. The system can be directly used for applications, such as assisting editors in choosing photos with a proper affective semantic for a certain topic during news report preparation.

The effect of hip muscle strengthening on pain and disability for individuals with non-specific low back pain : a randomized controlled trial

Introduction Clinical guidelines for the treatment of chronic low back pain suggest the use of supervised exercise. Motor control (MC) based exercise is widely used within clinical practice but its efficacy is equivalent to general exercise therapy. MC exercise targets the trunk musculature. Considering the mechanical links between the hip, pelvis, and lumbar spine, surprisingly little focus has been on investigating the contribution of the hip musculature to lumbopelvic support. The purpose of this study is to compare the efficacy of two exercise programs for the treatment of non-specific low back pain (NSLBP). Methods Eighty individuals aged 18-65 years of age were randomized into two groups to participate in this trial. The primary outcome measures included self-reported pain intensity (0-100mm VAS) and percent disability (Oswestry Disability Index V2). Bilateral measures of hip strength (N/kg) and two dimensional frontal plane mechanics (º) were the secondary outcomes. Outcomes were measured at baseline and following a six-week home based exercise program including weekly sessions of real-time ultrasound imaging. Results Within group comparisons revealed clinically meaningful reductions in pain for both groups. The MC exercise only (N= 40, xˉ =-20.9mm, 95%CI -25.7, -16.1) and the combined MC and hip exercise (N= 40, xˉ = -24.9mm, 95%CI -30.8, -19.0). There was no statistical difference in the change of pain (xˉ =-4.0mm, t= -1.07, p=0.29, 95%CI -11.5, 3.5) or disability (xˉ =-0.3%, t=-0.19, p=0.85, 95%CI -11.5, 3.5) between groups. Conclusion Both exercise programs had similar and positive effects on NSLBP which support the use of the home based exercise programs with weekly supervised visits. However, the addition of specific hip strengthening exercises to a MC based exercise program did not result in significantly greater reductions in pain or disability. Trial Registration NCTO1567566 Funding: Worker’s Compensation Board Alberta Research Grant.

Facial motion engages predictive visual mechanisms

We employed a novel cuing paradigm to assess whether dynamically versus statically presented facial expressions differentially engaged predictive visual mechanisms. Participants were presented with a cueing stimulus that was either the static depiction of a low intensity expressed emotion; or a dynamic sequence evolving from a neutral expression to the low intensity expressed emotion. Following this cue and a backwards mask, participants were presented with a probe face that displayed either the same emotion (congruent) or a different emotion (incongruent) with respect to that displayed by the cue although expressed at a high intensity. The probe face had either the same or different identity from the cued face. The participants' task was to indicate whether or not the probe face showed the same emotion as the cue. Dynamic cues and same identity cues both led to a greater tendency towards congruent responding, although these factors did not interact. Facial motion also led to faster responding when the probe face was emotionally congruent to the cue. We interpret these results as indicating that dynamic facial displays preferentially invoke predictive visual mechanisms, and suggest that motoric simulation may provide an important basis for the generation of predictions in the visual system.