Prediction of acute methanol poisoning prognosis using machine learning techniques.

Methanol poisoning is a global public health concern, especially prevalent in developing nations. This study focuses on predicting the severity of methanol intoxication using machine learning techniques, aiming to improve early identification and prognosis assessment. The study, conducted at Loghman Hakim Hospital in Tehran, Iran. The data pertaining to individuals afflicted with methanol poisoning was retrieved retrospectively and divided into training and test groups at a ratio of 70:30. The selected features were then inputted into various machine learning methods. The models were implemented using the Scikit-learn library in the Python programming language. Ultimately, the efficacy of the developed models was assessed through ten-fold cross-validation techniques and specific evaluation criteria, with a confidence level of 95%. A total number of 897 patients were included and divided in three groups including without sequel (n = 573), with sequel (n = 234), and patients who died (n = 90). The two-step feature selection was yielded 43 features in first step and 23 features in second step. In best model (Gradient Boosting Classifier) test dataset metric by 32 features younger age, higher methanol ingestion, respiratory symptoms, lower GCS scores, type of visual symptom, duration of therapeutic intervention, ICU admission, and elevated CPK levels were among the most important features predicting the prognosis of methanol poisoning. The Gradient Boosting Classifier demonstrated the highest predictive capability, achieving AUC values of 0.947 and 0.943 in the test dataset with 43 and 23 features, respectively. This research introduces a machine learning-driven prognostic model for methanol poisoning, demonstrating superior predictive capabilities compared to traditional statistical methods. The identified features provide valuable insights for early intervention and personalized treatment strategies.

Explainable artificial intelligence (XAI) for predicting the need for intubation in methanol-poisoned patients: a study comparing deep and machine learning models.

The need for intubation in methanol-poisoned patients, if not predicted in time, can lead to irreparable complications and even death. Artificial intelligence (AI) techniques like machine learning (ML) and deep learning (DL) greatly aid in accurately predicting intubation needs for methanol-poisoned patients. So, our study aims to assess Explainable Artificial Intelligence (XAI) for predicting intubation necessity in methanol-poisoned patients, comparing deep learning and machine learning models. This study analyzed a dataset of 897 patient records from Loghman Hakim Hospital in Tehran, Iran, encompassing cases of methanol poisoning, including those requiring intubation (202 cases) and those not requiring it (695 cases). Eight established ML (SVM, XGB, DT, RF) and DL (DNN, FNN, LSTM, CNN) models were used. Techniques such as tenfold cross-validation and hyperparameter tuning were applied to prevent overfitting. The study also focused on interpretability through SHAP and LIME methods. Model performance was evaluated based on accuracy, specificity, sensitivity, F1-score, and ROC curve metrics. Among DL models, LSTM showed superior performance in accuracy (94.0%), sensitivity (99.0%), specificity (94.0%), and F1-score (97.0%). CNN led in ROC with 78.0%. For ML models, RF excelled in accuracy (97.0%) and specificity (100%), followed by XGB with sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%). Overall, RF and XGB outperformed other models, with accuracy (97.0%) and specificity (100%) for RF, and sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%) for XGB. ML models surpassed DL models across all metrics, with accuracies from 93.0% to 97.0% for DL and 93.0% to 99.0% for ML. Sensitivities ranged from 98.0% to 99.37% for DL and 93.0% to 99.0% for ML. DL models achieved specificities from 78.0% to 94.0%, while ML models ranged from 93.0% to 100%. F1-scores for DL were between 93.0% and 97.0%, and for ML between 96.0% and 98.27%. DL models scored ROC between 68.0% and 78.0%, while ML models ranged from 84.0% to 96.08%. Key features for predicting intubation necessity include GCS at admission, ICU admission, age, longer folic acid therapy duration, elevated BUN and AST levels, VBG_HCO3 at initial record, and hemodialysis presence. This study as the showcases XAI's effectiveness in predicting intubation necessity in methanol-poisoned patients. ML models, particularly RF and XGB, outperform DL counterparts, underscoring their potential for clinical decision-making.

Prediction of acute organophosphate poisoning severity using machine learning techniques.

Poisoning with organophosphate compounds is a significant public health risk, especially in developing countries. Considering the importance of early and accurate prediction of organophosphate poisoning prognosis, the aim of this study was to develop a machine learning-based prediction model to predict the severity of organophosphate poisoning. The data of patients with organophosphate poisoning were retrospectively extracted and split into training and test sets in a ratio of 70:30. The feature selection was done by least absolute shrinkage and selection operator method. Selected features were fed into five machine learning techniques, including Histogram Boosting Gradient, eXtreme Gradient Boosting, K-Nearest Neighborhood, Support Vector Machine (SVM) (kernel = linear), and Random Forest. The Scikit-learn library in Python programming language was used to implement the models. Finally, the performance of developed models was measured using ten-fold cross-validation methods and some evaluation criteria with 95 % confidence intervals. A total of 1237 patients were used to train and test the machine learning models. According to the criteria determining severe organophosphate poisoning, 732 patients were assigned to group 1 (patients with mild to moderate poisoning) and 505 patients were assigned to group 2 (patients with severe poisoning). With an AUC value of 0.907 (95 % CI 0.89-0.92), the model developed using XGBoost outperformed other models. Feature importance evaluation found that venous blood gas-pH, white blood cells, and plasma cholinesterase activity were the top three variables that contribute the most to the prediction performance of the prognosis in patients with organophosphate poisoning. XGBoost model yield an accuracy of 90.1 % (95 % CI 0.891-0.918), specificity of 91.4 % (95 % CI 0.90-0.92), a sensitivity of 89.5 % (95 % CI 0.87-0.91), F-measure of 91.2 % (95 % CI 0.90-0.921), and Kappa statistic of 91.2 % (95 % CI 0.90-0.92). The machine learning-based prediction models can accurately predict the severity of organophosphate poisoning. Based on feature selection techniques, the most important predictors of organophosphate poisoning were VBG-pH, white blood cell count, plasma cholinesterase activity, VBG-BE, and age. The best algorithm with the highest predictive performance was the XGBoost classifier.

Demographic and Clinical Characteristics of 907 Cases with Naltrexone Intoxication; a 14-Year Cross-Sectional Study.

Introduction: Opioids have been the leading cause of death from poisoning in Iran for several years. This study aimed to evaluate the clinical and para-clinical presentations of naltrexone intoxication, its toxic dose, and its epidemiological properties. Methods: This retrospective cross-sectional study was conducted on medical records of patients presenting to Toxicology Department of Loghman Hakim Hospital, Tehran, Iran, following naltrexone intoxication, from 2002 to 2016. Patients' demographic and laboratory data, clinical signs, supposed ingested dose, and intent of naltrexone consumption were collected, analyzed, and then interpreted. Results: 907 patients with the mean age of 36.6 ±11.7 years were evaluated (94.3% male). The mean amount of naltrexone consumed by the intoxicated patients reported in the medical records was 105.8 ± 267.8 mg. One hundred thirty patients (14.3%) used naltrexone to treat substance use disorder. Two hundred eighty-seven poisoned patients (31.6%) were current opium users who intentionally or unintentionally used naltrexone concomitantly. The most common symptoms observed in these patients were agitation (41.8%), vomiting (16.4%), and nausea (14.8%). Among patients with naltrexone poisoning, 25 patients were intubated (2.8%), and three passed away. Aspartate aminotransferase (AST) levels were significantly higher in patients intoxicated with naltrexone who needed intubation (p = 0.02). Conclusion: The probability of intubation of cases with naltrexone intoxication was associated with AST elevation. It seems that, the number of intensive care unit (ICU) admissions and mortality rates are not high among these patients.

Is there any potential management against COVID-19? A systematic review and meta-analysis.

PURPOSE: A recent survey has shown that the COVID-19 pandemic has culminated in dramatical and critical treatment particularly in acute infected patients. In fact, this systematic review-meta-analysis was directly pertained to estimation at the efficient value of some clinical managements to confront the COVID-19 infection. METHODS: Pubmed, Embase, Scopus, Cochrane, and Scholar databases were searched from inception to July 1, 2020, to identify studies reporting the current treatment process and medications (e.g. hydroxychloroquine, antiviral therapy, convalescent plasma, and immunomodulatory agents) for COVID-19. A random-effects model meta-analysis was performed to calculate the relative risk (RR) with 95% confidence intervals (CI). The outcomes of this study were the frequency of negative conversion cases, clinical improvements, mechanical ventilation demand, intensive care unit (ICU) entry, and mortality. The standard treatment refers to the published guidelines and specialist experience which varies in different articles, and the proposed treatment refers to the kind of interest suggested in the included studies. RESULTS: A number of 45 articles met the eligibility criteria (out of 6793 articles). Among them, 26 articles involving 3263 patients were included in quantitative analysis. Anti-COVID-19 interventions could significantly increase clinical improvement (RR 1.17, 95% CI 1.08-1.27; I2 = 49.8%) and reduce the mortality rate (RR 0.58, 95% CI 0.35-0.95; I2 = 74.8%). Although in terms of negative conversion, ICU entry, and mechanical ventilation demand, clinical intervention had no beneficial effect. The clinical effect of immunomodulatory agents (especially tocilizumab and anakinra) was noticeable compared to other medications with RR of 0.22 (95% CI 0.09-0.53; I2 = 40.9%) for mortality and 1.25 (95% CI 1.07-1.46; I2 = 45.4%) for clinical improvement. Moreover, Antivirals (RR 1.13, 95% CI 1.01-1.26; I2 = 47.0%) and convalescent plasma therapy (RR 1.41, 95% CI 1.01-1.98; I2 = 66.6%) had significant beneficial effects on clinical improvement. CONCLUSION: Based on our findings, all the included interventions significantly declined the mortality and enhanced clinical improvements with no effect on negative conversion and mechanical ventilation demand. Especially, immunomodulators and plasma therapy showed favorable outcomes. An evaluation on the efficacy of proposed treatment against COVID-19.

Fabricating an electroactive injectable hydrogel based on pluronic-chitosan/aniline-pentamer containing angiogenic factor for functional repair of the hippocampus ischemia rat model.

The hippocampus, a critical cerebral region involved in learning and memory formation, is especially vulnerable to ischemic defect. Here, we developed an injectable electroactive hydrogel based on pluronic-chitosan/aniline-pentamer with proper conductivity around 10-4 S/cm to achieve the functional repair of the hippocampus following the ischemic defect. FTIR, DSC, and TGA measurements were performed to assess the chemical structure and thermal stability of the synthesized hydrogel. Aniline pentamer decreased the swelling capacity, degradation, and drug release rate. Further, contact angle, melting point, and gelation time of hydrogels were enhanced by addition of aniline oligomer. Moreover, it endowed the on-demand electro-responsive drug release. Injectability of hydrogel was evaluated by rheometry, exhibiting proper gelling time at the body temperature. The ionic/electrical conductivity and desired in vitro biocompatibility with PC12 cells were also achieved. Injection of VEGF-loaded electroactive hydrogel in the hippocampal ischemic animal model resulted in decreased infarction volume, improved hippocampal dependent learning, and memory performance. Taken all together, the results confirmed that fabricated injectable hydrogel would be a suitable candidate for ischemic defect treatment and can lead to new horizons to treat neurological disorders.

Neuroprotective effects of lipopolysaccharide and naltrexone co­preconditioning in the photothrombotic model of unilateral selective hippocampal ischemia in rat.

Preconditioning with lipopolysaccharide (LPS) or opioid antagonists has a neuroprotective effect in ischemic insults. However, the co­preconditioning effect of toll­like receptor ligands and opioid antagonists has not been investigated. In this study we examined the neuroprotective effect of LPS and naltrexone (NTX) preconditioning and co­preconditioning in unilateral selective hippocampal ischemia in rats to assess for possible synergistic protective effects. LPS and NTX were injected unilaterally into the left cerebral ventricle of male rats. Forty­eight hours after LPS and twenty­four hours after NTX injection, ipsilateral selective hippocampal ischemia was induced using a modified version of the photothrombotic method. Protective effects for LPS and NTX were assessed by evaluating infarct volume (using 2,3,5­triphenyltetrazolium chloride staining), and cognitive function (using radial arm water maze and passive avoidance tests). Animals in the ischemic group had an infarct lesion and considerable cognitive impairment, compared with the sham group. LPS or NTX preconditioning significantly reduced the infarct size and improved cognitive function. Moreover, co­preconditioning with LPS and NTX increased the protective effect compared with preconditioning with LPS or NTX alone. Our data showed that LPS and NTX preconditioning resulted in a neuroprotective effect in hippocampal ischemia. Furthermore, co­preconditioning with LPS and NTX resulted in a synergistic protective effect.

Neuroprotective effect of monophosphoryl lipid A, a detoxified lipid A derivative, in photothrombotic model of unilateral selective hippocampal ischemia in rat.

Finding a neuroprotective strategy to rescue patients suffering from acute brain damage is of great interest. Monophosphoryl lipid A (MPL) is a derivative of lipopolysaccharide (LPS) that lacks many of the endotoxic properties of the parent molecule, and yet has similar protective effect. Here, we report the first evidence that MPL preconditioning, similar to LPS preconditioning, can induce neuroprotection against cerebral ischemia. MPL (0.5, 1, 5 µg/rat) was injected unilaterally into the left cerebral ventricle of male rats, and 48 h later, rats were subjected to ipsilateral selective hippocampal ischemia using a modified version of the photothrombotic method. The neuroprotective effects of MPL and LPS were evaluated by measuring infarct size and assessing cognitive function. The expression level of some inflammatory and anti-inflammatory cytokines involving in TLR4 signaling pathway was also measured. Cognitive impairment and infarct size were obvious in control group receiving normal saline intracerebroventricularly and then selective hippocampal ischemia, compared to the sham group. Immunologic preconditioning with MPL or LPS significantly reduced infarct size and improved cognitive function. Additionally, immunologic preconditioning resulted in inflammatory mediators, NF-κB and TNF-α, down-regulation but anti-inflammatory mediators, IRF3, IFN-ß, and TGF-ß, up-regulation. Our data showed that both MPL and LPS preconditioning may reprogram the TLR4 signaling pathway to produce a cytokine profile which eventually leads to neuroprotection against ischemia injury. MPL, unlike LPS, is safe and well tolerated in clinic, thus it could be considered as a new approach in prevention or even treatment of cerebral ischemic insult consequences.

Photothrombotically induced unilateral selective hippocampal ischemia in rat.

INTRODUCTION: The vulnerability of hippocampal formation to ischemic insult has been documented in both humans and animal models. Ischemic injury induction through photothrombosis is an invasive and reproducible model of ischemic stroke which provides the ability to induce ischemia selectively in any desired area. In this study, we describe a method to induce selective unilateral hippocampal ischemia in rat through modified photothrombotic model. METHODS: Male wistar rats (n = 66) were subjected to stereotaxic surgery to insert a guide cannula just above the ascending part of hippocampal fissure. After recovery femoral vein was cannulated and rats were mounted in stereotaxic frame to optical fiber insertion in guide cannula and illumination of hippocampal fissure for 25 min. Rose Bengal dye was slowly injected through femoral vein during the first two-minute of illumination. Twenty-four hours later, infarct volume and histological change were evaluated in rat hippocampus. Cognitive function was also evaluated 48 h after ischemia induction. RESULTS: This procedure caused significant neuronal necrosis and infarct formation in the right hemisphere hippocampus. The infarct size was consisted in different subjects and was paralleled to cognitive impairment. The mean volume of infarction was 6.5% which affected whole right hippocampus and caused significant cognitive impairment compared to sham group (P < 0.001). DISCUSSION: The method described in this study provides the ability of selective hippocampal ischemia induction and study of hippocampal injury consequences following ischemia or other neurodegenerative disease that affect hippocampus.

Triggering microglia through toll-like receptor 2 pathway induced interferon ß expression in cell and animal model of Alzheimer's disease.

Synaptic function and memory performance are disrupted by soluble form of ß-amyloid (Aß). In the previous study, we found that early activation of microglia by toll-like receptor 2 (TLR2) attenuated Alzheimer's disease-associated cognitive deficit. This study was designed to investigate whether pretreatment with the TLR2 receptor ligand can regulate microglia to produce interferon ß (INFß) in a rat model of Alzheimer's disease. For this purpose, the BV-2 cell line was cultured in a 24-well plate, treated with Pam3Cys (1 µg/ml), and then incubated with oligomeric Aß for 24 h. The expression of TRIF, IRF3, and INFß was measured by western blot technique. For in-vivo study, bilateral guide cannulas were streotaxically implanted in the right and left lateral ventricles. Pam3Cys/vehicle was microinjected into the right ventricle every 3 days. Two weeks later, an osmotic pump was inserted into the left ventricle to microinfuse oligomeric Aß/placebo over 14 days. In the meanwhile, treatment with Pam3Cys was continued every 3 days. Then, expression of TRIF, IRF3, and INFß was measured in the hippocampus. The results showed that although oligomeric Aß could not alter the expression of these proteins at the cell and tissue level, treatment with Pam3Cys resulted in enhanced expression of them at both cell culture and hippocampal tissue following treatment with oligomeric Aß. It is concluded that stimulation of microglia through TLR2 pathway induces INFß expression, which may in part mediate neuroprotection against oligomeric Aß.

Protective effect of thymoquinone, the active constituent of Nigella sativa fixed oil, against ethanol toxicity in rats.

OBJECTIVES: Long term consumption of ethanol may induce damage to many organs. Ethanol induces its noxious effects through reactive oxygen species production, and lipid peroxidation and apoptosis induction in different tissues and cell types. Previous experiments have indicated the antioxidant characteristics of thymoquinone, the active constituent of Nigella sativa fixed oil, against biologically dangerous reactive oxygen species. This experiment was planned to evaluate the protective effect of thymoquinone against subchronic ethanol toxicity in rats. MATERIALS AND METHODS: Experiments were performed on six groups. Each group consisted of six animals, including control group (saline, gavage), ethanol-receiving group (3 g/kg/day, gavage), thymoquinone (2.5, 5, 10 mg/Kg/day, intraperitoneally (IP)) plus ethanol and thymoquinone (10 mg/Kg/day, IP) groups. Treatments were carried out in four weeks. RESULTS: Thymoquinone reduced the ethanol-induced increase in the lipid peroxidation and severity of histopathological alteration in liver and kidney tissues. In addition it improved the levels of proinflammatory cytokines in liver tissue. Furthermore, thymoquinone corrected the liver enzymes level including alanine transaminase, aspartate transaminase and alkaline phosphatase in serum and glutathione content in liver and kidney tissues. Other experiments such as Western blot analysis and quantitative real-time RT-PCR revealed that thymoquinone suppressed the expression of Bax/Bcl-2 ratio (both protein and mRNA level), and caspases activation pursuant to ethanol toxicity. CONCLUSION: This study indicates that thymoquinone may have preventive effects against ethanol toxicity in the liver and kidney tissue through reduction in lipid peroxidation and inflammation, and also interrupting apoptosis.

Effect of Nigella sativa fixed oil on ethanol toxicity in rats.

OBJECTIVES: This study was planned to appraise the protective effect of Nigella sativa fixed oil (NSO) against subchronic ethanol induced toxicity in rats. MATERIALS AND METHODS: Studies were carried out on six groups of six animals each, including control (normal saline, gavage), ethanol (3 g/kg/day, gavage), NSO (0.125, 0.25 and 0.5 ml/Kg/day, IP) plus ethanol and NSO (0.5 ml/Kg/day, IP) groups. Treatments were continued for 4 weeks. RESULTS: According to data, treatment with NSO attenuated ethanol-induced increased levels of malondialdehyde (MDA), tumor necrosis factor α (TNF-α) and interleukin-6 (IL-6), as well as histopathological changes in liver and kidney tissues. Moreover, NSO improved the level of serum liver enzymes (including alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (ALP) and glutathione (GSH) content in liver and kidney tissues in ethanol-treated rats. Western blot analysis and quantitative real time RT-PCR showed that NSO treatment inhibited apoptosis stimulated by ethanol through decreasing the Bax/Bcl-2 ratio (both protein and mRNA levels), cleaved caspase-3, cleaved caspase-8 and cleaved caspase-9 level in liver and kidney. CONCLUSION: This study showed that NSO may have protective effects against hepatotoxicity and renal toxicity of ethanol by decreasing lipid peroxidation and inflammation and preventing apoptosis.