26199 research outputs found
Sort by
Optimized Bio-Inspired Thermal Image Analysis for Mastitis Detection Using YOLOv8 Probabilistic Spiking Networks
No full textOne of the most common and financially detrimental conditions affecting dairy cows is mastitis, which is often identified by visual inspection or somatic cell count (SCC) testing, both of which are labor-intensive, subjective, and time-consuming. Existing deep learning (DL)-based approaches, including CLE-UNet, DCYOLO, FS-YOLOv4, and YOLOv7-SVM, have improved detection accuracy but still face challenges in capturing temporal dynamics, handling noisy thermal data, and optimizing model parameters effectively. To address these limitations, this work proposes a novel YOLOv8 integrated with Probabilistic Spiking Network and Secretary Bird Optimization (YOLOv8PSN-SBO) framework for automated mastitis detection using thermal imaging. The system begins with thermal video acquisition and manual annotation of key anatomical regions, followed by adaptive iterative guided filtering (AIGF) to enhance image clarity. Multi-scale feature extraction is achieved using a Cascading Residual Graph Convolutional Network (CRGCN), while YOLOv8 performs precise spatial localization of cow eyes, udders, and head postures. These spatial outputs are then temporally processed using a Probabilistic Spiking Neural Network (PSNN) that leverages biological spiking behavior to infer thermal anomalies. The Secretary Bird Optimization Algorithm (SBOA) is employed to fine-tune parameters such as confidence thresholds, anchor box sizes, loss weights, and spiking neuron dynamics for optimal performance. The proposed YOLOv8PSN-SBO model, implemented in PyTorch and validated on 1200 labeled thermal images, significantly outperforms baseline methods, achieving a detection accuracy of 91.75%, recall of 92.31%, F1-score of 91.55%, and an [email protected] of 94.60%, with a real-time processing speed of 65 FPS, establishing it as a robust and efficient solution for automated mastitis detection in dairy farming
A Deep Q-Network (DQN) Framework for Joint Optimization of EV Charging Station Placement and Vehicle Routing
No full textRapid EV growth requires efficient charging infrastructure and routing for urban sustainability, vital for reducing congestion, wait times, and energy use while promoting renewables. This paper presents a two-stage Deep Q-Network (DQN) framework for optimal EV station placement and routing. A DQN placement model minimizes future route energy by selecting station locations; a second DQN guides vehicles by approximating cost-to-go functions. Our DQN approach outperforms baselines (Q-Learning, NN, DGNN, Random Walk) in energy consumption, travel time, and route success. DQN placement achieved 1.2387 kWh average energy and 6.34 km average distance. DQN routing delivered 1.2587 kWh average energy, 1.98 average hops, and 96.67% success, demonstrating our DRL's effectiveness for scalable, reliable, and energy-efficient EV infrastructure planning
Pulsation Blip in the roAp Star HD 60435 Detected by SALT and TESS * * Based on observations made with the Southern African Large Telescope (SALT).
Get PDFHD 60435, a member of the rapidly oscillating Ap class of pulsating stars, was recently reported to be the first star to be observed to entirely cease pulsating. This conclusion was drawn from photometric TESS observations that have less sensitivity to pulsational variability than spectroscopic time series data. To that end, we acquired three epochs of time resolved spectra, to coincide with TESS observations, with the Southern African Large Telescope to test for the continued cessation of pulsation. We found the star to show a pulsation blip at a single epoch, then continue to be non-variable at two further epochs. The same signal, and lack thereof, was also seen when analyzing the nearly contemporaneous TESS photometry
Coeliac disease on the rise: the COVID-19 vaccination hypothesis
Get PDFIntroduction
Coeliac disease (CD) is a common lifelong autoimmune condition, more frequently diagnosed in women, with a global prevalence of 1–2%. Its broad clinical spectrum necessitates careful suspicion for timely diagnosis. Over the past 2 years, we have observed a sharp rise in cases, coinciding with the post-pandemic era, prompting this study. Some research suggests that viral infections may trigger autoimmune diseases, such as CD. This study explores the potential link between the surge in cases and factors such as infections and vaccinations.
Method
In this retrospective analysis, the medical records of a cohort of 454 patients were categorised according to date, age, gender, clinical manifestations, CD serology and histopathological findings. Data on vaccine type and dosage were analysed to explore a potential association.
Results were classified as follows:
Positive CD diagnosis: confirmed by serology and/or histopathology;
Negative CD diagnosis: both serology and histopathology results were negative;
Borderline CD: serology results fell within the grey zone, and histopathology was insufficient for a definitive diagnosis according to the Marsh-Oberhuber classification (stages 0 & 1).
Subdivisions: each diagnostic category was further subdivided based on vaccination status (vaccinated, unvaccinated, or unknown).
Gender analysis: results were also evaluated by patient sex.
Results
The results showed that 47.4% of the cohort tested positive or borderline for CD, including 32.8% of vaccinated patients who experienced symptoms during and after the coronavirus 2019 (COVID-19) pandemic. Notably, positivity increased among men, deviating from typical patterns. Many files lacked infection status, and genetic studies were not conducted because of cost, highlighting the need for further research.
Figure 1 visually represents CD diagnoses based on vaccination status using serology/histopathology results. It categorises patients into vaccinated (orange), unvaccinated (blue) and unknown status (green). Among vaccinated patients, 204 tested negative, while 149 were positive/borderline for CD. In the unvaccinated group, 33 were negative, and 37 were positive/borderline for CD. In the unknown group, 20 tested negative, and 11 tested positive/borderline for CD.
This breakdown suggests a notable CD prevalence, raising clinical suspicion about a link to COVID-19 vaccination/infection. The Chi-square value (6.18) supports this, but the p value (0.186) exceeds the statistical significance threshold (<0.05). While not statistically significant, the post-pandemic case surge may hold clinical significance.
Conclusion
To conclude, the data point toward an increasing trend in CD diagnosis, putting us at the forefront of managing a serious autoimmune systemic illness that carries a significant burden on healthcare. There are a few potential factors contributing to this trend, including:
• Enhanced awareness and increased availability of diagnostic tools;
• Suboptimal processing techniques of gluten-containing foods may increase immunogenicity;
• The possibility of an immunomodulating effect of the COVID-19 virus and its vaccine on the immune system.
Case reports suggest COVID may contribute to autoimmune diseases, such as rheumatoid arthritis. Despite lacking statistical significance, the data do not justify accepting the null hypothesis. The potential link between CD and COVID-19 vaccination/infection requires further study through larger prospective and retrospective studies
Does the Eccentric Strength Profile of the Hamstrings Determine the Severity of Injury Sustained?
Get PDFBackground and Purpose
Understanding how strength profiles of footballers effect the extent or location of injury sustained is limited. The purpose of this study was to investigate the effect of eccentric hamstrings strength (EHS), on the resultant injury sustained, graded utilizing the British Athletic Muscle Injury Classification, diagnosed via MRI.
Methods
EHS of 95 professional footballers were assessed on match-day minus two (MD-2) and plus two (MD+2). Participants completed three maximal repetitions of the bilateral Nordic hamstring exercises. Metrics included Peak Force (PF), Peak Torque (PkT), Average Force (AvF), Average Torque (AvT), Maximum Impulse (MI), Impulse Symmetry (IS), PF symmetry (PFS), PkT symmetry (PkTS) and relative strength (RS). Profiles were identified in relation to the date the injury was sustained.
Results
EHS across all metrics with the exemption of PkT, had no effect on injury grade, site or playing level. Tendon injuries were the most frequently occurring injury location, specifically grade 3c. Athlete strength had no effect on whether an injury was sustained or grade of injury.
Conclusion
The occurrence of tendon related injury is high within this study population and may be a result of the physical strength of the muscle tissue, combined with game play demands, exposing the tendon to greater forces, thus heightening the extent of injury sustained
A comparison of equine hind limb muscle activation and joint motion between forward and backward walking
No full textBackward walking (BW) is commonly employed as a physiotherapeutic exercise for horses based on anecdotal evidence for improving hindlimb strength, coordination, and range of motion. However, limited scientific evidence supports these assumed benefits. This study aimed to measure and compare equine hindlimb muscle activity and movement during BW and forward walking (FW). Three-dimensional kinematic and surface electromyography (sEMG) data were synchronously collected from unilateral (left) hindlimb and hip extensor (biceps femoris and gluteus medius) and flexor (tensor fasciae latae) muscles of ten horses during FW and BW. Normalised average rectified value (ARV), peak amplitude (PA), and muscle activity duration were calculated from sEMG data. Spatiotemporal and angular parameters were calculated from kinematic data. Wilcoxon signed rank tests or paired t-tests assessed differences between FW and BW. Compared to FW, significant (p<0.001) increases in hip extensor and decreases in hip flexor muscle ARV and PA were observed during BW. Muscle activity duration was significantly (p<0.001) longer for biceps femoris, and shorter for gluteus medius and tensor fasciae latae during BW. Hindlimb pro/retraction and sagittal plane hip and stifle joint movement cycles were time-reversed between BW and FW. Hindlimb joints were significantly less flexed and extended (p<0.001) during BW, except for peak stifle and hock flexion during swing, and MTPJ extension during stance, which were significantly greater (p<0.001) during BW. Findings support BW as a physiotherapeutic exercise to target increased hip extensor activity during weightbearing and to facilitate increased stifle and hock joint flexion during non-weightbearing phases of the gait cycle
252 - Barriers and facilitators to exercise engagement in people over 60 at risk of falls: an exploratory study
No full textPurpose: A study protocol abstract.
The main aim is to explore the barriers and facilitators to exercise engagement in participants over the age of 60 at risk of falls.
30% of over 65's and 50% of over 80's will fall each year, costing the NHS 2.3 billion. Locally, A and E attendances due to falls and hip fractures are above national average. Falling reduces patient confidence and further mobility, which can impact on developing long term conditions.
Musculoskeletal (MSK) Physiotherapists receive referrals of injuries or fractures post falls into the service. 30% of our caseload are over the age of 60, and health professionals should be screening for falls risk. Research has shown that functional exercises help improve the strength and balance in over 60's.
Patients may return back into the service following another injury, or fail to attend. Exploring patient views will help us to understand engagement, improve our services and create proactive falls signposting pathways.
Research has shown 78% of over 70's had not completed home exercises. There is also a lack of research studying engagement in patients over 60 in the north west and those with health inequalities.
Methods: This study is a qualitative study, using semi structured interviews to explore 10 patients experiences.
The inclusion criteria includes patients over the age of 60, identified at risk of falls and referred for MSK Physiotherapy in Rossendale.
Physiotherapy staff have filled in an electronic template. ‘Risk of falls’ is defined as: Self-reported unsteadiness/concern, one or more falls in the last 6 months, or clinician identified a slow/unsteady gait pattern.
Falls risk will already have been discussed with the physiotherapist and an exercise programme provided.
PPI engagement groups identified falls and exercise as important topics, and reviewed the patient facing documents prior to ethics submission and study commencement.
Interviews will be recorded using dictaphone, following confidential and governance principles as detailed in protocol.
Results: At the time of writing, three patients have been interviewed. It is estimated 10 patients will have been interviewed by Summer 2024. Patient's have been randomly stratified. Anonymised qualitative data will be transcribed by an NIHR transcriber. Data will be studied using NVivo to identify common patient themes, including peer review of analysed data and feedback from public patient group. It is estimated around 12 months from participant recruitment, to analysis of findings.
The template has been rolled out across Pendle, and ongoing meetings re: implementation of the fragility pathway.
Conclusion(s): Current findings are themes around participant beliefs such as habits, motivations and influences regarding behaviour change and physiotherapy engagement.
Also identifying participant preferences regarding reminders of exercises, leaflet/literacy and staff communication.
Developing next stage of research: questioning staff and patients around health inequalities such as age, fraility and socioeconomic underserved populations.
Impact: Short term: Apply themes to improve local and community service delivery, including exercise engagement, format and staff communication
Medium term: Contributing to MSK fragility pathway roll out, including support for wider trust workforce
Long term: Reduce re-referrals, reduce cost on NHS services and improve quality of life
Funding acknowledgements: Research was part of an internship programme funded by NIHR ARC 2022-23 and a scholarship programme funded by NIHR CRN 2023-2025. This provides one day a week towards the research project, training, scholarship programme work and any additional research activities such as ARCFEST and research festivals, poster conferences. NIHR ARC will also fund towards participant vouchers and the use of their professional transcriber and dictaphone. NIHR public advisors will be reimbursed for their time at the research meetings through the NIHR public advisors group
Case Study Challenge in Advanced Bladder Cancer: Tailoring Treatment to Individual Patients
No full textAll's Eco‐Friendly That Ends Eco‐Friendly (If Remembered as Such): Memory Processes in Retrospective Judgment of Environmentally Significant Sequences
Get PDFRetrospective judgments of environmentally significant sequences are biased by recency: sequences ending with an environmentally friendly item are rated as more eco‐friendly than otherwise identical sequences with the same item earlier in the list. A corresponding primacy effect is typically absent. This may have applied consequences for how consumers perceive the environmental friendliness of their purchase decisions, for example. The aim of the present investigation was to reach a better understanding of why the recency but not the primacy effect manifests in eco‐judgments. We found that the recency effect is just as large when continual distraction takes place between item presentations as when it does not. Moreover, memory for recently presented items was better than that for older items, but a filled retention interval reduced the recency effect in both memory and retrospective judgments. These findings support a memory‐based explanation of the recency effect in retrospective judgments and suggest that poor memory of items early in the sequence is the reason why the primacy effect in judgments does not manifest