A deep dive into burn-mediated ARDS severity assessment: a retrospective study on hematological markers.

Acute Respiratory Distress Syndrome (ARDS) is a critical form of Acute Lung Injury (ALI), challenging clinical diagnosis and severity assessment. This study evaluates the potential utility of various hematological markers in burn-mediated ARDS, including Neutrophil-to-Lymphocyte Ratio (NLR), Mean Platelet Volume (MPV), MPV-to-Lymphocyte Ratio (MPVLR), Platelet count, and Platelet Distribution Width (PDW). Employing a retrospective analysis of data collected over 12 years, this study focuses on the relationship between these hematological markers and ARDS diagnosis and severity in hospitalized patients. The study establishes NLR as a reliable systemic inflammation marker associated with ARDS severity. Elevated MPV and MPVLR also emerged as significant markers correlating with adverse outcomes. These findings suggest these economical, routinely measured markers can enhance traditional clinical criteria, offering a more objective approach to ARDS diagnosis and severity assessment. Hematological markers such as NLR, MPV, MPVLR, Platelet count, and PDW could be invaluable in clinical settings for diagnosing and assessing ARDS severity. They offer a cost-effective, accessible means to improve diagnostic accuracy and patient stratification in ARDS. However, further prospective studies are necessary to confirm these findings and investigate their integration with other diagnostic tools in diverse clinical settings.

Genetic counseling and genetic testing for pathogenic germline mutations among high-risk patients previously diagnosed with breast cancer: a traceback approach.

Genetic counseling and testing are more accessible than ever due to reduced costs, expanding indications and public awareness. Nonetheless, many patients missed the opportunity of genetic counseling and testing due to barriers that existed at that time of their cancer diagnoses. Given the identified implications of pathogenic mutations on patients' treatment and familial outcomes, an opportunity exists to utilize a 'traceback' approach to retrospectively examine their genetic makeup and provide consequent insights to their disease and treatment. In this study, we identified living patients diagnosed with breast cancer (BC) between July 2007 and January 2022 who would have been eligible for testing, but not tested. Overall, 422 patients met the eligibility criteria, 282 were reached and invited to participate, and germline testing was performed for 238, accounting for 84.4% of those invited. The median age (range) was 39.5 (24-64) years at BC diagnosis and 49 (31-75) years at the date of testing. Genetic testing revealed that 25 (10.5%) patients had pathogenic/likely pathogenic (P/LP) variants; mostly in BRCA2 and BRCA1. We concluded that long overdue genetic referral through a traceback approach is feasible and effective to diagnose P/LP variants in patients with history of BC who had missed the opportunity of genetic testing, with potential clinical implications for patients and their relatives.

Parallel detection of multiple biomarkers in a point-of-care-competent device for the prediction of exacerbations in chronic inflammatory lung disease.

Sudden aggravations of chronic inflammatory airway diseases are difficult-to-foresee life-threatening episodes for which advanced prognosis-systems are highly desirable. Here we present an experimental chip-based fluidic system designed for the rapid and sensitive measurement of biomarkers prognostic for potentially imminent asthma or COPD exacerbations. As model biomarkers we chose three cytokines (interleukin-6, interleukin-8, tumor necrosis factor alpha), the bacterial infection marker C-reactive protein and the bacterial pathogen Streptococcus pneumoniae-all relevant factors in exacerbation episodes. Assay protocols established in laboratory environments were adapted to 3D-printed fluidic devices with emphasis on short processing times, low reagent consumption and a low limit of detection in order to enable the fluidic system to be used in point-of-care settings. The final device demonstrator was validated with patient sample material for its capability to detect endogenous as well as exogenous biomarkers in parallel.

Intelligent bell facial paralysis assessment: a facial recognition model using improved SSD network.

With the continuous progress of technology, the subject of life science plays an increasingly important role, among which the application of artificial intelligence in the medical field has attracted more and more attention. Bell facial palsy, a neurological ailment characterized by facial muscle weakness or paralysis, exerts a profound impact on patients' facial expressions and masticatory abilities, thereby inflicting considerable distress upon their overall quality of life and mental well-being. In this study, we designed a facial attribute recognition model specifically for individuals with Bell's facial palsy. The model utilizes an enhanced SSD network and scientific computing to perform a graded assessment of the patients' condition. By replacing the VGG network with a more efficient backbone, we improved the model's accuracy and significantly reduced its computational burden. The results show that the improved SSD network has an average precision of 87.9% in the classification of light, middle and severe facial palsy, and effectively performs the classification of patients with facial palsy, where scientific calculations also increase the precision of the classification. This is also one of the most significant contributions of this article, which provides intelligent means and objective data for future research on intelligent diagnosis and treatment as well as progressive rehabilitation.

Applying an explainable machine learning model might reduce the number of negative appendectomies in pediatric patients with a high probability of acute appendicitis.

The diagnosis of acute appendicitis and concurrent surgery referral is primarily based on clinical presentation, laboratory and radiological imaging. However, utilizing such an approach results in as much as 10-15% of negative appendectomies. Hence, in the present study, we aimed to develop a machine learning (ML) model designed to reduce the number of negative appendectomies in pediatric patients with a high clinical probability of acute appendicitis. The model was developed and validated on a registry of 551 pediatric patients with suspected acute appendicitis that underwent surgical treatment. Clinical, anthropometric, and laboratory features were included for model training and analysis. Three machine learning algorithms were tested (random forest, eXtreme Gradient Boosting, logistic regression) and model explainability was obtained. Random forest model provided the best predictions achieving mean specificity and sensitivity of 0.17 ± 0.01 and 0.997 ± 0.001 for detection of acute appendicitis, respectively. Furthermore, the model outperformed the appendicitis inflammatory response (AIR) score across most sensitivity-specificity combinations. Finally, the random forest model again provided the best predictions for discrimination between complicated appendicitis, and either uncomplicated acute appendicitis or no appendicitis at all, with a joint mean sensitivity of 0.994 ± 0.002 and specificity of 0.129 ± 0.009. In conclusion, the developed ML model might save as much as 17% of patients with a high clinical probability of acute appendicitis from unnecessary surgery, while missing the needed surgery in only 0.3% of cases. Additionally, it showed better diagnostic accuracy than the AIR score, as well as good accuracy in predicting complicated acute appendicitis over uncomplicated and negative cases bundled together. This may be useful in centers that advocate for the conservative treatment of uncomplicated appendicitis. Nevertheless, external validation is needed to support these findings.

Sensorimotor tests in patients with neck pain and its associated disorders: a systematic review and meta-analysis.

This systematic review aimed to synthesize the current evidence regarding neck sensorimotor testing in individuals with neck pain, assess the differences between neck pain groups and healthy controls, and recognize factors that might affect test results. We performed the data search using PubMed, Embase, PsycINFO, CINAHL, and Scopus databases. We used a two-step screening process to identify studies. Furthermore, we screened the reference lists for additional studies. Hedges g was used to present the difference between neck pain groups and asymptomatic individuals. We assessed the quality of the studies using the QUADAS tool. The final review included 34 studies, of which 25 were related to the joint position error test, four to the smooth pursuit neck torsion test and six to the balance test. Our meta-analysis showed poorer joint-position sense, oculomotor function, and wider postural sway in individuals with neck pain than healthy controls. The size of the difference between the groups seemed to be influenced by the intensity of the pain and the presence of dizziness. Therefore, it might be helpful in future studies to differentiate patients with neck pain into subgroups based on their symptom and demographic profiles to assess other factors that significantly affect cervical sensorimotor control.

Contributing to the management of viral infections through simple immunosensing of the arachidonic acid serum level.

A trendsetting direct competitive-based biosensing tool has been developed and implemented for the determination of the polyunsaturated fatty acid arachidonic acid (ARA), a highly significant biological regulator with decisive roles in viral infections. The designed methodology involves a competitive reaction between the target endogenous ARA and a biotin-ARA competitor for the recognition sites of anti-ARA antibodies covalently attached to the surface of carboxylic acid-coated magnetic microbeads (HOOC-MµBs), followed by the enzymatic label of the biotin-ARA residues with streptavidin-horseradish peroxidase (Strep-HRP) conjugate. The resulting bioconjugates were magnetically trapped onto the sensing surface of disposable screen-printed carbon transducers (SPCEs) to monitor the extent of the biorecognition reaction through amperometry. The operational functioning of the exhaustively optimized and characterized immunosensing bioplatform was highly convenient for the quantitative determination of ARA in serum samples from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2-) and respiratory syncytial virus (RSV)-infected individuals in a rapid, affordable, trustful, and sensitive manner.

Alpha-fetoprotein and APRI as predictive markers for patients with Type C hepatitis B-related acute-on-chronic liver failure: a retrospective study.

BACKGROUND: Type C hepatitis B-related acute-on-chronic liver failure (HBV-ACLF), which is based on decompensated cirrhosis, has different laboratory tests, precipitating events, organ failure and clinical outcomes. The predictors of prognosis for type C HBV-ACLF patients are different from those for other subgroups. This study aimed to construct a novel, short-term prognostic score that applied serological indicators of hepatic regeneration and noninvasive assessment of liver fibrosis to predict outcomes in patients with type C HBV-ACLF. METHOD: Patients with type C HBV-ACLF were observed for 90 days. Demographic information, clinical examination, and laboratory test results of the enrolled patients were collected. Univariate and multivariate logistic regression were performed to identify independent prognostic factors and develop a novel prognostic scoring system. A receiver operating characteristic (ROC) curve was used to analyse the performance of the model. RESULTS: A total of 224 patients with type C HBV-ACLF were finally included. The overall survival rate within 90 days was 47.77%. Age, total bilirubin (TBil), international normalized ratio (INR), alpha-fetoprotein (AFP), white blood cell (WBC), serum sodium (Na), and aspartate aminotransferase/platelet ratio index (APRI) were found to be independent prognostic factors. According to the results of the logistic regression analysis, a new prognostic model (named the A3Twin score) was established. The area under the curve (AUC) of the receiver operating characteristic curve (ROC) was 0.851 [95% CI (0.801-0.901)], the sensitivity was 78.8%, and the specificity was 71.8%, which were significantly higher than those of the MELD, IMELD, MELD-Na, TACIA and COSSH-ACLF II scores (all P < 0.001). Patients with lower A3Twin scores (<-9.07) survived longer. CONCLUSIONS: A new prognostic scoring system for patients with type C HBV-ACLF based on seven routine indices was established in our study and can accurately predict short-term mortality and might be used to guide clinical management.

Methodological identification of anomalies episodes in ECG streams: a systematic mapping study.

An electrocardiogram is a medical examination tool for measuring different patterns of heart blood flow circle either in the form of usual or non-invasive patterns. These patterns are useful for the identification of morbidity condition of the heart especially in certain conditions of heart abnormality and arrhythmia. Myocardial infarction (MI) is one of them that happened due to sudden blockage of blood by the cause of malfunction of heart. In electrocardiography (ECG) intensity of MI is highlighted on the basis of unusual patterns of T wave changes. Various studies have contributed for MI through T wave's classification, but more to the point of T wave has always attracted the ECG researchers. Methodology. This Study is primarily designed for proposing the combination of latest methods that are worked for the solutions of pre-defined research questions. Such solutions are designed in the form of the systematic review process (SLR) by following the Kitchen ham guidance. The literature survey is a two phase's process, at first phase collect the articles that were published in IEEE Xplore, Scopus, science direct and Springer from 2008 to 2023. It consist of steps; the first level is executed by filtrating the articles on the basis of keyword phase of title and abstract filter. Similarly, at two level the manuscripts are scanned through filter of eligibility criteria of articles selection. The last level belongs to the quality assessment of articles, in such level articles are rectified through evaluation of domain experts. Results. Finally, the selected articles are addressed with research questions and briefly discuss these selected state-of-the-art methods that are worked for the T wave classification. These address units behave as solutions to research problems that are highlighted in the form of research questions. Conclusion and future directions. During the survey process for these solutions, we got some critical observations in the form of gaps that reflected the other directions for researchers. In which feature engineering, different dependencies of ECG features and dimensional reduction of ECG for the better ECG analysis are reflection of future directions.

ECG autoencoder based on low-rank attention.

The prevalence of cardiovascular disease (CVD) has surged in recent years, making it the foremost cause of mortality among humans. The Electrocardiogram (ECG), being one of the pivotal diagnostic tools for cardiovascular diseases, is increasingly gaining prominence in the field of machine learning. However, prevailing neural network models frequently disregard the spatial dimension features inherent in ECG signals. In this paper, we propose an ECG autoencoder network architecture incorporating low-rank attention (LRA-autoencoder). It is designed to capture potential spatial features of ECG signals by interpreting the signals from a spatial perspective and extracting correlations between different signal points. Additionally, the low-rank attention block (LRA-block) obtains spatial features of electrocardiogram signals through singular value decomposition, and then assigns these spatial features as weights to the electrocardiogram signals, thereby enhancing the differentiation of features among different categories. Finally, we utilize the ResNet-18 network classifier to assess the performance of the LRA-autoencoder on both the MIT-BIH Arrhythmia and PhysioNet Challenge 2017 datasets. The experimental results reveal that the proposed method demonstrates superior classification performance. The mean accuracy on the MIT-BIH Arrhythmia dataset is as high as 0.997, and the mean accuracy and F 1 -score on the PhysioNet Challenge 2017 dataset are 0.850 and 0.843.

PPIH acts as a potential predictive biomarker for patients with common solid tumors.

BACKGROUND: Our previous studies have indicated that mRNA and protein levels of PPIH are significantly upregulated in Hepatocellular Carcinoma (LIHC) and could act as predictive biomarkers for patients with LIHC. Nonetheless, the expression and implications of PPIH in the etiology and progression of common solid tumors have yet to be explored, including its potential as a serum tumor marker. METHODS: We employed bioinformatics analyses, augmented with clinical sample evaluations, to investigate the mRNA and protein expression and gene regulation networks of PPIH in various solid tumors. We also assessed the association between PPIH expression and overall survival (OS) in cancer patients using Kaplan-Meier analysis with TCGA database information. Furthermore, we evaluated the feasibility and diagnostic efficacy of PPIH as a serum marker by integrating serological studies with established clinical tumor markers. RESULTS: Through pan-cancer analysis, we found that the expression levels of PPIH mRNA in multiple tumors were significantly different from those in normal tissues. This study is the first to report that PPIH mRNA and protein levels are markedly elevated in LIHC, Colon adenocarcinoma (COAD), and Breast cancer (BC), and are associated with a worse prognosis in these cancer patients. Conversely, serum PPIH levels are decreased in patients with these tumors (LIHC, COAD, BC, gastric cancer), and when combined with traditional tumor markers, offer enhanced sensitivity and specificity for diagnosis. CONCLUSION: Our findings propose that PPIH may serve as a valuable predictive biomarker in tumor patients, and its secreted protein could be a potential serum marker, providing insights into the role of PPIH in cancer development and progression.

Prediction of subsequent fragility fractures: application of machine learning.

BACKGROUND: Machine learning (ML) has shown exceptional promise in various domains of medical research. However, its application in predicting subsequent fragility fractures is still largely unknown. In this study, we aim to evaluate the predictive power of different ML algorithms in this area and identify key features associated with the risk of subsequent fragility fractures in osteoporotic patients. METHODS: We retrospectively analyzed data from patients presented with fragility fractures at our Fracture Liaison Service, categorizing them into index fragility fracture (n = 905) and subsequent fragility fracture groups (n = 195). We independently trained ML models using 27 features for both male and female cohorts. The algorithms tested include Random Forest, XGBoost, CatBoost, Logistic Regression, LightGBM, AdaBoost, Multi-Layer Perceptron, and Support Vector Machine. Model performance was evaluated through 10-fold cross-validation. RESULTS: The CatBoost model outperformed other models, achieving 87% accuracy and an AUC of 0.951 for females, and 93.4% accuracy with an AUC of 0.990 for males. The most significant predictors for females included age, serum C-reactive protein (CRP), 25(OH)D, creatinine, blood urea nitrogen (BUN), parathyroid hormone (PTH), femoral neck Z-score, menopause age, number of pregnancies, phosphorus, calcium, and body mass index (BMI); for males, the predictors were serum CRP, femoral neck T-score, PTH, hip T-score, BMI, BUN, creatinine, alkaline phosphatase, and spinal Z-score. CONCLUSION: ML models, especially CatBoost, offer a valuable approach for predicting subsequent fragility fractures in osteoporotic patients. These models hold the potential to enhance clinical decision-making by supporting the development of personalized preventative strategies.

A systematic review of machine learning models for management, prediction and classification of ARDS.

AIM: Acute respiratory distress syndrome or ARDS is an acute, severe form of respiratory failure characterised by poor oxygenation and bilateral pulmonary infiltrates. Advancements in signal processing and machine learning have led to promising solutions for classification, event detection and predictive models in the management of ARDS. METHOD: In this review, we provide systematic description of different studies in the application of Machine Learning (ML) and artificial intelligence for management, prediction, and classification of ARDS. We searched the following databases: Google Scholar, PubMed, and EBSCO from 2009 to 2023. A total of 243 studies was screened, in which, 52 studies were included for review and analysis. We integrated knowledge of previous work providing the state of art and overview of explainable decision models in machine learning and have identified areas for future research. RESULTS: Gradient boosting is the most common and successful method utilised in 12 (23.1%) of the studies. Due to limitation of data size available, neural network and its variation is used by only 8 (15.4%) studies. Whilst all studies used cross validating technique or separated database for validation, only 1 study validated the model with clinician input. Explainability methods were presented in 15 (28.8%) of studies with the most common method is feature importance which used 14 times. CONCLUSION: For databases of 5000 or fewer samples, extreme gradient boosting has the highest probability of success. A large, multi-region, multi centre database is required to reduce bias and take advantage of neural network method. A framework for validating with and explaining ML model to clinicians involved in the management of ARDS would be very helpful for development and deployment of the ML model.

Risk for cardiovascular disease development in rheumatoid arthritis.

BACKGROUND: Patients with rheumatoid arthritis have significant cardiovascular mortality and morbidity. OBJECTIVE: To investigate the effects of chronic inflammation in rheumatoid arthritis on cardiovascular morbidity association with cardiovascular risk factors risk factors. Mortality report is secondary just to show trends without sufficient statistical power as it is accidental endpoint. METHODS: A total of 201 individuals without previous cardiovascular disease, 124 with rheumatoid arthritis (investigation group) and 77 with osteoarthritis (control group), were included in the study and followed up for an average of 8 years to assess the development of fatal or non-fatal cardiovascular diseases. The incidence and prevalence of cardiovascular risk factors were also investigated. RESULTS: The total incidence of one or more fatal or nonfatal cardiovascular events was 43.9% in the investigation group and 37.5% in the control group. Of these patients, 31.7% and 30.9% survived cardiovascular events in the investigation and control groups, respectively. The most common cardiovascular disease among participants who completed the study and those who died during the study was chronic heart failure. The results of the subgroup analysis showed that strict inflammation control plays a central role in lowering cardiovascular risk. CONCLUSION: A multidisciplinary approach to these patients is of paramount importance, especially with the cooperation of immunologists and cardiologists for early detection, prevention, and management of cardiovascular risks and diseases.

miR-486-5p predicted adverse outcomes of SCAP and regulated K. pneumonia infection via FOXO1.

PURPOSE: Severe community-acquired pneumonia (SCAP) is a common respiratory system disease with rapid development and high mortality. Exploring effective biomarkers for early detection and development prediction of SCAP is of urgent need. The function of miR-486-5p in SCAP diagnosis and prognosis was evaluated to identify a promising biomarker for SCAP. PATIENTS AND METHODS: The serum miR-486-5p in 83 patients with SCAP, 52 healthy individuals, and 68 patients with mild CAP (MCAP) patients were analyzed by PCR. ROC analysis estimated miR-486-5p in screening SCAP, and the Kaplan-Meier and Cox regression analyses evaluated the predictive value of miR-486-5p. The risk factors for MCAP patients developing SCAP were assessed by logistic analysis. The alveolar epithelial cell was treated with Klebsiella pneumonia to mimic the occurrence of SCAP. The targeting mechanism underlying miR-486-5p was evaluated by luciferase reporter assay. RESULTS: Upregulated serum miR-486-5p screened SCAP from healthy individuals and MCAP patients with high sensitivity and specificity. Increasing serum miR-486-5p predicted the poor outcomes of SCAP and served as a risk factor for MCAP developing into SCAP. K. pneumonia induced suppressed proliferation, significant inflammation and oxidative stress in alveolar epithelial cells, and silencing miR-486-5p attenuated it. miR-486-5p negatively regulated FOXO1, and the knockdown of FOXO1 reversed the effect of miR-486-5p in K. pneumonia-treated alveolar epithelial cells. CONCLUSION: miR-486-5p acted as a biomarker for the screening and monitoring of SCAP and predicting the malignancy of MCAP. Silencing miR-486-5p alleviated inflammation and oxidative stress induced by K. pneumonia via negatively modulating FOXO1.

Role of secreted frizzled-related protein 4 in prediabetes and type 2 diabetes: a cross sectional study.

BACKGROUND: Type 2 diabetes (T2D) has become an epidemic. Delays in diagnosis and as a consequent late treatment has resulted in high prevalence of complications and mortality. Secreted frizzled-related protein 4 (SFRP4), has been recently identified as a potential early biomarker of T2D related to obesity, due to its association with low grade inflammation in adipose tissue and impaired glucose metabolism. We aimed to evaluate the role of SFRP4 in prediabetes and T2D in a Mexican population. METHODS: This was a cross-sectional study that included 80 subjects with T2D, 50 subjects with prediabetes and 50 healthy individuals. Fasting SFRP4 and insulin concentrations were measured by ELISA. Human serum IL-10, IL-6, IL-1ß and IL-8 levels were quantified by flow cytometry. Genotyping was performed by TaqMan® probes. RESULTS: Prediabetes and T2D patients had significantly higher SFRP4 levels than controls (P < 0.05). In turn, prediabetes subjects had higher SFRP4 concentrations than control subjects (P < 0.05). Additionally, the prediabetes and T2D groups had higher concentrations of proinflammatory molecules such as IL-6, IL-1ß and IL-8, and lower concentrations of IL-10, an anti-inflammatory cytokine, than controls (P < 0.001). The serum SFRP4 concentrations were positively correlated with parameters that are elevated in prediabetes and T2D states, such as, HbA1c and homeostasis model assessment insulin resistance (HOMA-IR), (r = 0.168 and 0.248, respectively, P < 0.05). Also, serum SFRP4 concentrations were positively correlated with concentrations of pro-inflammatory molecules (CRP, IL-6, IL-1ß and IL-8) and negatively correlated with the anti-inflammatory molecule IL-10, even after adjusting for body mass index and age (P < 0.001). The genetic variant rs4720265 was correlated with low HDL concentrations in T2D (P < 0.05). CONCLUSIONS: SFRP4 correlates positively with the stage of prediabetes, suggesting that it may be an early biomarker to predict the risk of developing diabetes in people with high serum concentrations of SFRP4, although further longitudinal studies are required.

Assessment of the appropriate cutoff points for anthropometric indices and their relationship with cardio-metabolic indices to predict the risk of metabolic associated fatty liver disease.

BACKGROUND: Research on Metabolic Associated Fatty Liver Disease (MAFLD) is still in its early stages, with few studies available to identify and predict effective indicators of this disease. On the other hand, early diagnosis and intervention are crucial to reduce the burden of MAFLD. Therefore, the aim of this research was to investigate the effectiveness of eleven anthropometric indices and their appropriate cut-off values as a non-invasive method to predict and diagnose MAFLD in the Iranian population. METHODS: In this cross-sectional study, we analyzed baseline data from the Hoveyzeh Cohort Study, a prospective population-based study conducted in Iran that enrolled a total of 7836 subjects aged 35 to 70 years from May 2016 through August 2018. RESULTS: The optimal cut-off values of anthropometric indices for predicting MAFLD risk were determined for waist circumference(WC) (102.25 cm for males and 101.45 cm for females), body mass index (BMI) (27.80 kg/m2 for males and 28.75 kg/m2 for females), waist-to-hip ratio (WHR) (0.96 for both males and females), waist-to-height ratio (WHtR) (0.56 for males and 0.63 for females), body adiposity index (BAI) (23.24 for males and 32.97 for females), visceral adiposity index (VAI) (1.64 for males and 1.88 for females), weight-adjusted waist index (WWI) (10.63 for males and 11.71 for females), conicity index (CI) (1.29 for males and 1.36 for females), body roundness index (BRI) (4.52 for males and 6.45 for females), relative fat mass (RFM) (28.18 for males and 44.91 for females) and abdominal volume index (AVI) (18.85 for males and for 21.37 females). VAI in males (sensitivity: 77%, specificity: 60%, Youden's Index: 0.37) and RFM in females (sensitivity: 76%, specificity: 59%, Youden's Index: 0.35) were found to have higher sensitivity and specificity compared to other anthropometric indices. Furthermore, anthropometric indices demonstrated statistically significant correlations with various hepatic and cardiometabolic indices. Among these, the strongest positive correlations were observed between WC, BMI, BAI, BRI, and AVI with the Hepatic Steatosis Index (HSI), TyG-BMI, and TyG-WC, as well as between VAI and the Atherogenic Index of Plasma (AIP), Lipid Accumulation Product (LAP), Cardiometabolic Index (CMI), and the Triglyceride and Glucose (TyG) Index. CONCLUSION: Anthropometric indices are effective in predicting MAFLD risk among Iranian adults, with WWI, VAI, and RFM identified as the strongest predictors. The proposed cutoff values could serve as a straightforward and non-invasive methods for the early diagnosis of MAFLD.

Development and external validation of a machine learning model for the prediction of persistent acute kidney injury stage 3 in multi-centric, multi-national intensive care cohorts.

BACKGROUND: The aim of this retrospective cohort study was to develop and validate on multiple international datasets a real-time machine learning model able to accurately predict persistent acute kidney injury (AKI) in the intensive care unit (ICU). METHODS: We selected adult patients admitted to ICU classified as AKI stage 2 or 3 as defined by the "Kidney Disease: Improving Global Outcomes" criteria. The primary endpoint was the ability to predict AKI stage 3 lasting for at least 72 h while in the ICU. An explainable tree regressor was trained and calibrated on two tertiary, urban, academic, single-center databases and externally validated on two multi-centers databases. RESULTS: A total of 7759 ICU patients were enrolled for analysis. The incidence of persistent stage 3 AKI varied from 11 to 6% in the development and internal validation cohorts, respectively and 19% in external validation cohorts. The model achieved area under the receiver operating characteristic curve of 0.94 (95% CI 0.92-0.95) in the US external validation cohort and 0.85 (95% CI 0.83-0.88) in the Italian external validation cohort. CONCLUSIONS: A machine learning approach fed with the proper data pipeline can accurately predict onset of Persistent AKI Stage 3 during ICU patient stay in retrospective, multi-centric and international datasets. This model has the potential to improve management of AKI episodes in ICU if implemented in clinical practice.

Actigraph-based quantification of sleep in children with dystonia undergoing deep brain stimulation.

OBJECTIVE: Dystonia is among the most common pediatric movement disorders and can manifest with a range of debilitating symptoms, including sleep disruptions. The duration and quality of sleep are strongly associated with quality of life in these individuals and could serve as biomarkers of dystonia severity and the efficacy of interventions such as deep brain stimulation (DBS). Thus, this study investigated sleep duration and its relationship to disease severity and DBS response in pediatric dystonia. METHODS: Actigraphs (wearable three-axis accelerometers) were used to record multiday sleep data in 22 children with dystonia, including 6 patients before and after DBS implantation, and age- and sex- matched healthy controls. Data were preprocessed, and metrics of sleep duration and quality were extracted. Repeated-measures statistical analyses were used. RESULTS: Children with dystonia slept less than typically developing children (p = 0.009), and shorter sleep duration showed trending correlation with worse dystonia severity (r = -0.421, p = 0.073). Of 4 patients who underwent DBS and had good-quality data, 1 demonstrated significantly improved sleep (p < 0.001) postoperatively. Reduction in dystonia severity strongly correlated with increased sleep duration after DBS implantation (r = -0.965, p = 0.035). CONCLUSIONS: Sleep disturbances are an underrecognized marker of pediatric dystonia severity, as well as the effectiveness of interventions such as DBS. They can serve as objective biomarkers of disease burden and symptom progression after treatment.

Can the DEX/CRH test or markers of oxidative stress distinguish work-related stress from major depressive disorder and normal controls?

Hypothalamic-pituitary-adrenal (HPA)-axis hyperactivity measured by the combined dexamethasone-CRH test (DEX-CRH test) has been found in patients with major depressive disorder (MDD), whereas hypoactivity has been found in patients with work-related stress. We aimed to investigate the DEX-CRH test as a biomarker to distinguish between MDD and work-related stress (exhaustion disorder - ED). We hypothesized that there would be lower cortisol and ACTH response in participants with ED compared to MDD and healthy controls (HC). Also, we explored if the cortisol response of those patients interacted with robust markers of oxidative stress. Thirty inpatients with MDD and 23 outpatients with ED were recruited. Plasma cortisol and ACTH were sampled during a DEX-CRH test. The main outcome measure, area under the curve (AUC) for cortisol and ACTH, was compa-red between MDD vs. ED participants and a historical HC group. Secondary markers of oxidative stress urinary 8-oxodG and 8-oxoGuo; quality of sleep and psychometrics were obtained. Cortisol concentrations were higher in MDD and ED participants compared to HC, and no differences in AUC cortisol and ACTH were found between ED vs. MDD. Compared to ED, MDD participants had higher stress symptom severity and a lower sense of well-being. No differences in oxidative stress markers or quality of sleep between the groups were found. The result indicates that the patients with ED, like patients with MDD, are non-suppressors in DEX-CRH test and not hypocortisolemic as suggested.