An automated method for tendon image segmentation on ultrasound using grey-level co-occurrence matrix features and hidden Gaussian Markov random fields.

BACKGROUND: Despite knowledge of qualitative changes that occur on ultrasound in tendinopathy, there is currently no objective and reliable means to quantify the severity or prognosis of tendinopathy on ultrasound. OBJECTIVE: The primary objective of this study is to produce a quantitative and automated means of inferring potential structural changes in tendinopathy by developing and implementing an algorithm which performs a texture based segmentation of tendon ultrasound (US) images. METHOD: A model-based segmentation approach is used which combines Gaussian mixture models, Markov random field theory and grey-level co-occurrence (GLCM) features. The algorithm is trained and tested on 49 longitudinal B-mode ultrasound images of the Achilles tendons which are labelled as tendinopathic (24) or healthy (25). Hyperparameters are tuned, using a training set of 25 images, to optimise a decision tree based classification of the images from texture class proportions. We segment and classify the remaining test images using the decision tree. RESULTS: Our approach successfully detects a difference in the texture profiles of tendinopathic and healthy tendons, with 22/24 of the test images accurately classified based on a simple texture proportion cut-off threshold. Results for the tendinopathic images are also collated to gain insight into the topology of structural changes that occur with tendinopathy. It is evident that distinct textures, which are predominantly present in tendinopathic tendons, appear most commonly near the transverse boundary of the tendon, though there was a large variability among diseased tendons. CONCLUSION: The GLCM based segmentation of tendons under ultrasound resulted in distinct segmentations between healthy and tendinopathic tendons and provides a potential tool to objectively quantify damage in tendinopathy.

An assessment of contamination fingerprinting techniques for determining the impact of domestic wastewater treatment systems on private well supplies.

Private wells in Ireland and elsewhere have been shown to be prone to microbial contamination with the main suspected sources being practices associated with agriculture and domestic wastewater treatment systems (DWWTS). While the microbial quality of private well water is commonly assessed using faecal indicator bacteria, such as Escherichia coli, such organisms are not usually source-specific, and hence cannot definitively conclude the exact origin of the contamination. This research assessed a range of different chemical contamination fingerprinting techniques (ionic ratios, artificial sweeteners, caffeine, fluorescent whitening compounds, faecal sterol profiles and pharmaceuticals) as to their use to apportion contamination of private wells between human wastewater and animal husbandry wastes in rural areas of Ireland. A one-off sampling and analysis campaign of 212 private wells found that 15% were contaminated with E. coli. More extensive monitoring of 24 selected wells found 58% to be contaminated with E. coli on at least one occasion over a 14-month period. The application of fingerprinting techniques to these monitored wells found that the use of chloride/bromide and potassium/sodium ratios is a useful low-cost fingerprinting technique capable of identifying impacts from human wastewater and organic agricultural contamination, respectively. The artificial sweetener acesulfame was detected on several occasions in a number of monitored wells, indicating its conservative nature and potential use as a fingerprinting technique for human wastewater. However, neither fluorescent whitening compounds nor caffeine were detected in any wells, and faecal sterol profiles proved inconclusive, suggesting limited suitability for the conditions investigated.

Efficacy of intra-articular injections of platelet-rich plasma as a symptom- and disease-modifying treatment for knee osteoarthritis - the RESTORE trial protocol.

BACKGROUND: Knee osteoarthritis (OA) causes substantial pain, physical dysfunction and impaired quality of life. There is no cure for knee OA, and for some people, the disease may involve progressive symptomatic and structural deterioration over time. Platelet-rich plasma (PRP) is a therapeutic agent that aims to address underlying biological processes responsible for OA pathogenesis. As such, it has the potential to improve both symptoms and joint structure. The aim of this clinical trial is to determine whether a series of injections of PRP into the knee joint will lead to a significantly greater reduction in knee pain, and less loss of medial tibial cartilage volume over 12 months when compared to a series of placebo saline injections in people with knee OA. METHODS: This will be a two-group, superiority, randomised, participant-, interventionist- and assessor-blinded, placebo-controlled trial. Two hundred and eighty-eight participants aged over 50 years with painful knee OA and mild to moderate structural change on x-ray (Kellgren and Lawrence grade 2 and 3) will be randomly allocated to receive either three PRP injections or three normal saline injections into the knee joint at weekly intervals. The primary outcomes will be 12-month change in average overall knee pain severity (numeric rating scale) and medial tibial cartilage volume (magnetic resonance imaging (MRI)). Secondary outcomes include additional measures of knee pain and other symptoms, function in daily living and sport and recreation, quality of life, participant-perceived global ratings of change, and other MRI structural outcomes including meniscal and cartilage morphology, synovitis, effusion, bone marrow lesions and cartilage defects. A range of additional measures will be recorded, and a separate health economic evaluation will be performed. DISCUSSION: The findings from this study will help determine whether PRP improves both clinical and structural knee OA outcomes over 12 months when compared to a series of placebo saline injections. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry reference: ACTRN12617000853347 . Prospectively registered 9th of June 2017.

A comparison between clinical assessment and magnetic resonance imaging of acute hamstring injuries.

BACKGROUND: Physicians evaluating hamstring strains in professional football players are increasingly turning to magnetic resonance imaging to support the clinical diagnosis and management of the injury. However, little information is available to assess how magnetic resonance imaging compares with the clinical evaluation in establishing the duration of rehabilitation required. HYPOTHESIS: Magnetic resonance imaging of hamstring strains can be useful in determining duration of rehabilitation. STUDY DESIGN: Cohort study (Diagnosis); Level of evidence, 1. METHODS: Fifty-eight professional football players with a diagnosis of hamstring injury made by the team physician were enrolled in the study. All players underwent magnetic resonance imaging and a clinical evaluation by an independent physical therapist within 3 days of the injury. Presence, type, and location of injury were recorded in each examination. The physical therapist estimated the time required until return to competition, and the radiologist used the length of the injury (coronal view) to establish rehabilitation duration. Both clinicians were blinded to the other modality. RESULTS: Clinical and magnetic resonance imaging assessments were in agreement in 38 of 58 cases (65.5%). In 18 cases (31.0%), a clinically positive diagnosis was made, but no abnormalities were evident on magnetic resonance imaging. In 2 cases (3.4%), magnetic resonance imaging detected an injury, whereas the clinical examination had negative or equivocal findings. Both clinical examination and magnetic resonance imaging findings were strongly correlated with the actual time required to return to competition (r = .69, P < .001 and r = .58, P < .001, respectively). The correlation coefficient between clinical predictions and magnetic resonance imaging findings was moderate (r = .36, P = .006). CONCLUSION: This study shows that magnetic resonance imaging is not required for estimating the duration of rehabilitation of an acute minor or moderate hamstring injury in professional football players.