Browse Articles

Gaussian random fields as an abstract representation of patient metadata for multimodal medical image segmentation

Abstract Growing rates of chronic wound occurrence, especially in patients with diabetes, has become a recent concerning trend. Chronic wounds are difficult and costly to treat, and have become a serious burden on health care systems worldwide. Innovative deep learning methods for the detection and monitoring of such wounds have the potential to reduce the impact to patients and clinicians. We present a novel multimodal segmentation method which allows for the introduction of patient metadata into the training workflow whereby the patient data are expressed as Gaussian random fields. Our results indicate that the proposed method improved performance when utilising multiple models, each trained on different metadata categories. Using the Diabetic Foot Ulcer Challenge 2022 test set, when compared to the baseline results (intersection over union = 0.4670, Dice similarity coefficient = 0.5908) we demonstrate improvements of +0.0220 and +0.0229 for intersection over union and Dice similarity coefficient respectively. This paper presents the first study to focus on integrating patient data into a chronic wound segmentation workflow. Our results show significant performance gains when training individual models using specific metadata categories, followed by average merging of prediction masks using distance transforms. All source code for this study is available at: https://github.com/mmu-dermatology-research/multimodal-grf

Science’s ‘Gollum effect’: PhDs bear brunt of territorial behaviour

Understanding public perception of health examination services: Key factors influencing satisfaction in Türkiye

The aim of this study is to evaluate the level of satisfaction with health examination services in Türkiye. It is thought that the findings will contribute to the more effective management of the health service process and offer potential solutions to identified problems. Notably, a significant portion of the problems encountered in healthcare services tends to arise during the examination phase. Therefore, this research was conducted to address these problems by thoroughly analyzing public satisfaction, with the expectation that such an approach could provide actionable insights for resolving these problems. In the study, the micro data set of the 2023 Life Satisfaction Survey conducted by the Turkish Statistical Institute was used. The analysis process was carried out with a two-stage method. In the first stage, Pearson’s χ² test was used to evaluate whether the independent variables had a statistically significant relationship with satisfaction with health examination services. In the second stage, a considering the binary categorical structure of the dependent variable, a logit regression model was applied to estimate the relationship between satisfaction with health examination services and the independent variables. The findings revealed that 61.85% of Turkish citizens were satisfied with health examination services. Furthermore, this level of satisfaction was significantly affected by a wide range of sociodemographic, individual, and institution-related factors. The study’s findings suggest that aligning individuals’ demands in the health service process with guidance from field experts and developing targeted policies could lead to improved satisfaction with health examination services. In addition, it is foreseen that the concept of trust is important in the satisfaction that constitutes the main subject of the study in health services and in the negative situations experienced in different subjects. Based on these insights, initiatives can be taken to increase trust in the health system through health policies to be designed. Furthermore, the results highlight the growing importance of digitalization and digital hospitals in healthcare. Further progress in this direction will increase the satisfaction with health examination and contribute to positive results in health services.

New healthcare payment models: risk scores aren’t enough to guide resource allocation

Abstract Around the world, aging populations compel healthcare delivery systems to improve how they allocate increasingly scarce resources. In parallel, economic pressures motivate healthcare payors and policy makers to adopt global budgeting and accountable payment models based on actuarial risk. We investigated whether these risk-based approaches could apply to healthcare resource allocation. Because a significant portion of healthcare resources for older adults is currently associated with potentially avoidable hospital admissions, we focused our investigation on allocating care coordination resources targeted toward those most likely to be admitted. Using a computational risk-based analysis of claims data, we found the 20% highest expected hospital utilization segment had an average hospitalization rate of over 68% per year, compared to 27% for the overall study population. However, only half of all hospitalizations in the study population were accounted for in the top 20% risk segment. Additionally, 63% of beneficiaries in the top 20% risk segment experienced zero hospitalizations. Our results indicate that risk-based resource allocation may fail to target some high hospital utilizers while allocating resources to many who are never hospitalized. These results further indicate that risk-based expectation of hospital utilization may be insufficient as a basis for effective allocation of care coordination resources.

The impact of autonomy-supportive organizational environments on employees’ emotions and creative performance: A self-determination theory perspective

The ongoing debate over whether positive or negative emotions foster creative performance remains a pivotal issue in understanding the interplay between emotions and creativity. Emerging research suggests that both positive and certain negative emotions, such as fear and guilt, can enhance creativity under specific conditions. Grounded in Self-Determination Theory (SDT), this study examines how autonomy-supportive organizational environments contribute to the satisfaction of employees’ basic psychological needs. It further explores how these needs influence work-related emotions and ultimately foster creative performance. Data were collected from 283 leaders and employees across various enterprises in Mainland China. Descriptive statistics were analyzed using SPSS 26.0, and structural equation modeling (SEM) with latent variables was conducted using AMOS 26.0 to test the hypothesized relationships and mediating effects. The results demonstrate that autonomy support positively influences the satisfaction of basic psychological needs, which subsequently promotes positive emotions and enhances creative performance. Conversely, autonomy support negatively affects the frustration of basic psychological needs, thereby mitigating negative emotions. Mediation analyses reveal that basic psychological needs mediate the relationship between autonomy support and emotions, while both positive and negative emotions mediate the relationship between autonomy support and creative performance. These findings provide valuable insights into the mechanisms linking autonomy-supportive environments, psychological needs, emotions, and creativity. Beyond its theoretical contributions to SDT, this study offers practical guidance for organizations aiming to cultivate employee creativity and well-being by fostering supportive and autonomy-oriented workplace climates.

First-principles theory for cerium predicts three distinct face-centered cubic phases

Neurological outcomes and predictive factors in traumatic spinal cord injury patients in the intensive care unit

Background This study aims to analyze the demographic characteristics, clinical features, and neurological outcomes of patients with traumatic spinal cord injury (TSCI) admitted to the Intensive Care Unit (ICU), in order to provide scientific evidence for the prevention and management of TSCI. Methods A retrospective analysis was conducted on data from TSCI patients admitted to the ICU between January 2018 and December 2022. Demographic information, neurological injury characteristics, complications during hospitalization, treatment interventions, and prognosis were comprehensively collected. Based on changes in neurological function before and after treatment, patients undergoing surgery were classified into improvement and non-improvement groups. Neurological recovery was assessed using the American Spinal Injury Association (ASIA) impairment scale. Univariate and multivariate logistic regression analyses were performed to identify key factors influencing neurological recovery. Results A total of 341 TSCI patients were included, with a mean age of 55.2 ± 13.4 years and a male-to-female ratio of 6.3:1. The leading cause of TSCI were high falls (47.5%), traffic accidents (35.8%) and low falls (9.1%). Cervical spinal cord was most common, followed by thoracic and lumbar cord. Among surgical patients, the neurological improvement rate was 14.8%, compared to 12.5% in non-surgical patients, highlighting the potential benefits of surgical intervention. Multivariate analysis revealed that early targeted blood pressure management (MAP ≥ 85 mmHg) (OR=2.296, 95% CI: 1.036–5.086, P = 0.040) and early surgery (≤ 24h) (OR=2.841, 95% CI: 1.088–7.419, P = 0.033) were significant protective factors for neurological improvement. Conclusions Patients with TSCI admitted to the ICU are predominantly middle-aged men, with high falls and traffic accidents being the primary causes. Early blood pressure optimization and timely surgical intervention are significantly associated with improved neurological outcomes.

The role of HMOX1-mediated ferroptosis in blue light-induced damage to retinal pigment epithelium

Informer-based DDoS attack detection method for the power Internet of Things

With the rapid development of smart grids, power grid systems are becoming increasingly complex, posing significant challenges to their security. Traditional network intrusion detection systems often rely on manually engineered features, which are not only resource-intensive but also struggle to handle the diverse range of attack types. This paper aims to address these challenges by proposing an automated DDoS attack detection algorithm using the Informer model. We introduce a windowing technique to segment network traffic into manageable samples, which are then input into the Informer for feature extraction and classification. This model captures both the temporal dependencies and global attention information in the traffic data. Experimental results on the CICIDS-2018 dataset demonstrate the effectiveness of our approach, showing significant improvements in detection accuracy and efficiency. Our findings suggest that the proposed method offers a promising solution for real-time intrusion detection in complex power grid environments.

Compact femtosecond fiber laser tunable from 800 to 850 nm with pulse energy exceeding 5 nJ

Abstract This paper introduces a compact, tunable femtosecond laser based on an Erbium-doped fiber, utilizing the Self-Soliton Frequency Shifted technique and PPLN crystal as a Second Harmonic Generation module. Achieving an unparalleled frequency conversion efficiency up to 55% for the 800 - 850 nm wavelength range, this compact laser emits sub-100 fs pulses. The laser operates simultaneously within the first and third biological windows, delivering pulse energies of 10.4 nJ and 5.1 nJ, respectively. This performance, previously unattained in similar systems, is achieved while maintaining Second Harmonic Generation power stability below 2% RMS. The presented compact laser, developed for bladder cancer detection through multiphoton microscopy, will significantly improve the system’s compactness, precision, and cancer detection efficiency.

I told AI to make me a protein. Here’s what it came up with

Effects of dietary of Bacillus coagulans, whey powder, and their interaction on the performance of Lohmann LSL-lite laying hens in the late production phase

Due to the need to produce high-quality and healthy eggs, the current experiment was conducted to investigate the impact of dietary supplementation of whey powder (WP), Bacillus coagulans (B. coagulans), and their combination (MIX) on the production performance, egg quality, blood biochemical parameters, and histomorphological parameters of Lohmann LSL-lite laying hens. 144 Lohmann laying hen (75 weeks) were randomly assigned to 4 different dietary treatments, with 6 replications and 6 hens per cage. The hens were fed a basal diet (control, CON), the basal diet supplemented with 1 g/kg WP, 1 g/kg B. coagulans (4 × 106 CFU), and 1 g/kg WP plus 1 g/kg B. coagulans probiotic for 12 weeks. Feed intake and egg weight were not affected by the treatments at any stage of the trial (P > 0.05). No significant interaction was found between WP and B. coagulans in egg quality parameters and blood-biochemical parameters other than malondialdehyde (P < 0.05). The level of malondialdehyde in serum was reduced when WP was used along with B. coagulans compared to when WP was used alone. However, egg production in all periods and egg mass in the first period were affected by the synergistic effect of WP and B. coagulans. Furthermore, FCR was reduced in the first period (75–80 weeks) under the influence of the MIX group compared to the control group or when used alone (P < 0.05). The color of the yolk was increased in the group receiving B. coagulans compared to the control group (P < 0.05). Therefore, in birds fed with B. coagulans, a significant increase in the width of the villi in the ileum was observed (P ≤ 0.05). Interestingly, B. coagulans and WP reduced performance when used alone compared to the control, but improved performance when combined. In conclusion, the simultaneous use of WP and B. coagulans in diet can probably improve the parameters of production performance, FCR, and serum malondialdehyde level at the end of the production period of Lohmann laying hens.

Construction and validation of antibody dependent cell phagocytosis related risk model in breast cancer

Findings from transcriptomics and immunohistochemistry indicate an autoimmune disease targeting brainstem inhibitory interneurons in bovine spastic paresis

Bovine spastic paresis (BSP) is a progressive neuromuscular disease of unknown origin that causes persistent stiffness of the hind limbs. The symptoms are similar to those of human motor neuron diseases such as primary (PLS) or amyotrophic lateral sclerosis (ALS). BSP occurs worldwide in cattle production with an estimated prevalence of <1%. For Germany, this means that around 20,000 Holstein cattle are affected. BSP is generally considered a hereditary disease, but there is no prevention through breeding programs. As a result, BSP not only affects animal welfare but also leads to economic losses in milk and beef production. Here, we used transcriptomics to analyse the brainstem, spinal cord and affected gastrocnemius muscle tissue of eight animals affected by BSP and eight control animals from slaughterhouses to gain new insights into the molecular mechanisms underlying BSP. We found that the expression of several genes was significantly different in animals affected by BSP compared to control animals. Specific genes for inhibitory neurons were downregulated in the brainstems of the affected animals, namely CCK (cholecystokinin), NPY (neuropeptide Y), and SST (somatostatin). These inhibitory neurotransmitters influence cerebral movement control, among other processes. Furthermore, OOSP2 (oocyte secreted protein 2) was found to be significantly upregulated in the affected animals in all tissues. This expression could best be explained by the presence of T-follicular-helper cells which, through interleukin 21, can trigger a TH-2-dominated immune response and lead to autoimmune encephalitis. Further cases were sampled for confirmation and we detected cell infiltrates of activated microglia and T-cells in the brainstem using immunohistochemistry. Microglial foci were significantly more abundant in animals affected by BSP than control animals. We conclude that BSP is caused by an autoimmune reaction directed against inhibitory interneurons in the brainstem and is due to a combination of genetics and environmental influences. This may result in lost controlling influence on the upper motor neurons via extrapyramidal pathways and therefore triggers the specific symptoms of motor neuron disease.

A phenomenographic study on Chinese EFL teachers’ cognitions of positive and negative educational, social, and psychological consequences of high-stake tests

The removal of black ink via Emericella quadrilineata as a green alternative technique to recycling ink waste papers

In order to fight deforestation, biological methods of recycling printed waste papers must be used. In addition to identifying and isolating A. quadrilineatus, the current study attempts to ascertain the best physiological conditions and mechanisms underlying this species’ ability to deink. Five isolates such as Cladosporium sp., Aspergillus sp., Fusarium sp., Penicillium sp., and Rhizopus sp. isolated from soil containing ink remains using Bushnell and Hass media carried out the deinking tests. SEM, FT-IR, the molecular method, and factors affecting ink eradication were all carried out. The deinking of ink-loaded filter paper, Langmuir and Freundlich adsorption isotherms, and A. quadrilineatus enzyme activities were also investigated in this work. Ninety per percentage of the black ink was removed by A. quadrilineatus. Six days later, under ideal conditions (pH 6, temperature 30°C, initial ink concentrations of 20,000 mg L ⁻ ¹, and inoculum dose of three fungal discs), the optimal deinking percentage from solution through a culture of A. quadrilineatus reached up to 97%. The deinking mechanism of A. quadrilineatus was shown by SEM and FT-IR studies. A good level of agreement between the Langmuir adsorption isotherm model and the adsorption process was shown by the Freundlich and Langmuir adsorption isotherms. Otherwise, on agar plates, A. quadrilineatus demonstrated its capacity to manufacture the enzymes lipase and xylanase. Overall, the results indicated that A. quadrilineatus may open up new possibilities for recycling printed waste papers.

Research on multi-algorithm and explainable AI techniques for predictive modeling of acute spinal cord injury using multimodal data

NSF plan to slash ‘indirect’ science funding: will it stick?

The impact of baseline laboratory tests on the management of new-onset hypertension in primary care: A pilot study

Background Hypertension is a key contributor to the global cardiovascular disease burden. In 2021, the World Health Organization (WHO) hypertension management guideline suggested baseline laboratory tests for patients with newly diagnosed hypertension but noted limited evidence in primary care contexts. This study evaluates the impact of baseline laboratory assessments on blood pressure control and comorbidities in newly diagnosed hypertensive patients. Methods This is a multicenter retrospective study that included all patients with new-onset hypertension between January 2015 and January 2020, followed in three primary health care centers until 2022. Data collection included 8 items of paraclinical tests performed at the diagnosis of hypertension: serum sodium, potassium and creatinine, lipid panel, electrocardiogram, glucose, HbA1c and urine dipstick. Complete workup was defined as having the 8 items checked and partial workup included 1–7 items. Blood pressure was assessed at one year and the final visit, which was beyond one year, in the two workup groups. Results Of 621 hypertensive patients, 107 with incident hypertension were analyzed (mean age: 54.8 ± 12.7 years; 58.9% women). A complete workup was done for 48 patients, partial for 52 and none for 7. Abnormalities detected included: 8.4% of patients with fasting blood glucose > 125 mg/dL, 7.5% with HbA1c > 6.5%, 1.9% with serum potassium < 3.5 mmoL/L, 54.2% with LDL Cholesterol > 100 mg/dL, 35.5% with serum creatinine > 0.8 mg/dL, and 7.5% with an estimated GFR < 60 mL/min/1.73 m2. Significant systolic blood pressure improvement was seen at 12 months in the complete workup group (129.9 ± 13.6 mmHg) vs. partial workup group (142.8 ± 18.9 mm Hg) (P = 0.003). Men and smokers were tested more often than women and non-smokers. Conclusion Baseline laboratory tests in patients with newly diagnosed hypertension help unmask comorbidities such as chronic kidney disease, diabetes, and dyslipidemia. Our findings support the use of baseline laboratory testing in order to optimize blood pressure control and individual patient management.

Trainable embedding quantum physics informed neural networks for solving nonlinear PDEs

Abstract This paper proposes a novel approach for solving nonlinear partial differential equations (PDEs) with a quantum computer, the trainable embedding quantum physics informed neural network (TE-QPINN). We combine quantum machine learning (QML) with physics informed neural networks (PINNs) in a hybrid approach. By leveraging the advantages of classical and quantum computers, we can create algorithms that have a potential to be run on noisy intermediate-scale quantum devices (NISQ). We use feedforward neural networks (FNN) as problem-agnostic embedding functions, giving the used quantum circuit greater expressibility than previously introduced embedding. This expressibility allows us to solve a wide range of problems without using a problem specific ansatz. Additionally, we introduce a hybrid backpropagation algorithm that allows efficient updates of the used weights and biases in the FNN embedding functions. In this paper we showcase the capabilities of TE-QPINNs of a wide range of problems, including the two-dimensional Poisson, Burgers and Navier-Stokes equations. In direct comparison with classical PINNs, this approach showed an ability to achieve superior results while using the same number of parameters, highlighting their potential for more efficient optimization in high-dimensional parameter spaces, which could be transformative for future applications.