Identifying Putative Biomarkers of Foodborne Pathogens Using a Metabolomic Approach.

Metabolomics is the study of low molecular weight biochemical molecules (typically <1500 Da) in a defined biological organism or system. In case of food systems, the term "food metabolomics" is often used. Food metabolomics has been widely explored and applied in various fields including food analysis, food intake, food traceability, and food safety. Food safety applications focusing on the identification of pathogen-specific biomarkers have been promising. This chapter describes a nontargeted metabolite profiling workflow using gas chromatography coupled with mass spectrometry (GC-MS) for characterizing three globally important foodborne pathogens, Escherichia coli O157:H7, Listeria monocytogenes, and Salmonella enterica, from selective enrichment liquid culture media. The workflow involves a detailed description of food spiking experiments followed by procedures for the extraction of polar metabolites from media, the analysis of the extracts using GC-MS, and finally chemometric data analysis using univariate and multivariate statistical tools to identify potential pathogen-specific biomarkers.

A novel method for objectively classifying sequential emotion within dreams: a proof-of-concept pilot study.

Traditionally, emotions in dreams have been assessed using subjective ratings by human raters (e.g., external raters or dreamers themselves). These methods have extensive support and utility in dream science, yet they have certain innate limitations due to the subjective nature of the rating methodologies. Attempting to circumvent several of these limitations, we aimed to develop a novel method for objectively classifying and quantifying sequential (word-for-word) emotion within a dream report. We investigated whether sentiment analysis, a branch of natural language processing, could be used to generate continuous positive and negative valence ratings across a dream. In this pilot, proof-of-concept study, we used 14 dream reports collected upon awakening following overnight polysomnography. We also collected pre- and post-sleep affective data and personality metrics. Our objectives included demonstrating that (1) valence ratings derived from sentiment analysis (Valence Aware Dictionary for sEntiment Reasoning [VADER]) could be used to visualize (plot) positive and negative emotion fluctuations within a dream, (2) how the visual properties of emotion fluctuations within a dream (peaks and troughs, area under the curve) can be used to generate novel "emotion indicators" as proxies for emotion regulation throughout a dream, and (3) these emotion indicators correlate with sleep, affective, and personality variables known to be associated with dreaming and emotion regulation. We describe 6 novel, objective dream emotion indicators: Total number of Peaks, total number of Troughs, Positive, Negative, and Overall Emotion Intensity (composites from an "area under the curve" method using the trapezoid rule applied to the peaks and troughs), and the Emotion Gradient (a polynomial trendline fitted to the emotion fluctuations in the dream chart). The latter signifies the overall direction of sequential emotion changes within a dream. Results also showed that ⅚ emotion indicators correlated significantly with at least one existing sleep, affective, or personality variable known to be associated with dreaming and emotion regulation. We propose that the novel emotion indicators potentially serve as proxies for emotion regulation processes unfolding within a dream. These preliminary findings provide a methodological foundation for future studies to test and refine the method in larger and more diverse samples.

Assessing the precision of machine learning for diagnosing pulmonary arterial hypertension: a systematic review and meta-analysis of diagnostic accuracy studies.

Introduction: Pulmonary arterial hypertension (PAH) is a severe cardiovascular condition characterized by pulmonary vascular remodeling, increased resistance to blood flow, and eventual right heart failure. Right heart catheterization (RHC) is the gold standard diagnostic technique, but due to its invasiveness, it poses risks such as vessel and valve injury. In recent years, machine learning (ML) technologies have offered non-invasive alternatives combined with ML for improving the diagnosis of PAH. Objectives: The study aimed to evaluate the diagnostic performance of various methods, such as electrocardiography (ECG), echocardiography, blood biomarkers, microRNA, chest x-ray, clinical codes, computed tomography (CT) scan, and magnetic resonance imaging (MRI), combined with ML in diagnosing PAH. Methods: The outcomes of interest included sensitivity, specificity, area under the curve (AUC), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR). This study employed the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool for quality appraisal and STATA V.12.0 for the meta-analysis. Results: A comprehensive search across six databases resulted in 26 articles for examination. Twelve articles were categorized as low-risk, nine as moderate-risk, and five as high-risk. The overall diagnostic performance analysis demonstrated significant findings, with sensitivity at 81% (95% CI = 0.76-0.85, p < 0.001), specificity at 84% (95% CI = 0.77-0.88, p < 0.001), and an AUC of 89% (95% CI = 0.85-0.91). In the subgroup analysis, echocardiography displayed outstanding results, with a sensitivity value of 83% (95% CI = 0.72-0.91), specificity value of 93% (95% CI = 0.89-0.96), PLR value of 12.4 (95% CI = 6.8-22.9), and DOR value of 70 (95% CI = 23-231). ECG demonstrated excellent accuracy performance, with a sensitivity of 82% (95% CI = 0.80-0.84) and a specificity of 82% (95% CI = 0.78-0.84). Moreover, blood biomarkers exhibited the highest NLR value of 0.50 (95% CI = 0.42-0.59). Conclusion: The implementation of echocardiography and ECG with ML for diagnosing PAH presents a promising alternative to RHC. This approach shows potential, as it achieves excellent diagnostic parameters, offering hope for more accessible and less invasive diagnostic methods. Systematic Review Registration: PROSPERO (CRD42024496569).

Unlocking Hidden Risks: Harnessing Artificial Intelligence (AI) to Detect Subclinical Conditions from an Electrocardiogram (ECG).

Recent artificial intelligence (AI) advancements in cardiovascular medicine offer potential enhancements in diagnosis, prediction, treatment, and outcomes. This article aims to provide a basic understanding of AI enabled ECG technology. Specific conditions and findings will be discussed, followed by reviewing associated terminology and methodology. In the appendix, definitions of AUC versus accuracy are explained. The application of deep learning models enables detecting diseases from normal electrocardiograms at accuracy not previously achieved by technology or human experts. Results with AI enabled ECG are encouraging as they considerably exceeded current screening models for specific conditions (i.e., atrial fibrillation, left ventricular dysfunction, aortic stenosis, and hypertrophic cardiomyopathy). This could potentially lead to a revitalization of the utilization of the ECG in the insurance domain. While we are embracing the findings with this rapidly evolving technology, but cautious optimism is still necessary at this point.

Comparative assessment of trough and peak levels and AUC24 for amikacin in nontuberculous mycobacterial infection.

BACKGROUND: Amikacin, an aminoglycoside antibiotic, is widely used for the treatment of nontuberculous mycobacterial (NTM) infections. To date, therapeutic drug monitoring (TDM) of amikacin has primarily relied on the measurement of peak and trough levels as indicators rather than the 24-hr area under the concentration-time curve (AUC24). METHODS: NTM patients referred for amikacin TDM from March 2021 to May 2023 were assessed for the AUC24 values based on administered dose. We investigated re-admission rates, all-cause mortality and AFB smear results to evaluate clinical outcome based on the actual AUC24 values. Ototoxicity and nephrotoxicity were also investigated as adverse effects in correlation with TDM parameters. RESULTS: Among 65 patients, the mean and median values of AUC24 were 234 and 249 mg·hr/L, respectively. In a group of patients with AUC24 values less than 250 mg·hr/L, 42.4 % of patients were re-admitted for pulmonary symptoms. On the contrary, another group with AUC24 values equal to or more than 250 mg·hr/L, had lower re-admission rates (25.0 %). They also showed lower all-cause mortality rates and more improvement on acid-fast bacilli smear results. Moderate to poor correlation between AUC24 values and peak/trough levels were observed. Ototoxicity and nephrotoxicity were revealed to be associated with drug exposure duration rather than AUC24 values. CONCLUSION: In this study, we performed comparative assessment of trough/peak level, traditional clinical marker for amikacin TDM, and AUC24 value. Although AUC24 values showed poor to moderate correlation to trough/peak levels, higher AUC24 correlated with favorable clinical outcomes without additional risk of toxicity.

A Comparison of Vancomycin Area Under the Curve and Trough Concentration in Specific Populations.

Background: Vancomycin is an antibiotic known to cause nephrotoxicity, particularly when a vancomycin trough of 15 to 20 mg/L, a surrogate for an area under the curve (AUC) of at least 400 mgh/L, is targeted. Although monitoring vancomycin AUC is more resource intensive, it may especially benefit populations expected to be at higher risk of nephrotoxicity. Objective: To describe the proportion of discordance between vancomycin AUC and trough concentration in targeted high-risk populations. Methods: A prospective observational review was conducted on adults receiving intravenous vancomycin for more than 48 hours from May 9 to June 3, 2022. Patients included were elderly, obese, had renal dysfunction, and/or received 4 grams or more of vancomycin daily with a pending vancomycin trough concentration. A peak concentration was ordered by a project team member to calculate AUC to assess discordance. Results: A total of 47 patients were included with 87 vancomycin minimum concentration (Cmin)/AUC pairs analyzed. Discordance was observed in 52.9% of Cmin/AUC pairs in the entire cohort. The majority (79%) of the 43 Cmin levels <15 mg/L had an associated AUC >400 mgh/L and 57% of 21 Cmin levels within the 15 to 20 mg/L range had an AUC >600 mgh/L. Conclusion: A high degree of discordance between vancomycin Cmin and AUC was present in patients considered to be at high risk of nephrotoxicity. Monitoring vancomycin AUC in these patients may reduce the risk of nephrotoxicity.

Preoperative prediction of hepatocellular carcinoma microvascular invasion based on magnetic resonance imaging feature extraction artificial neural network.

BACKGROUND: Hepatocellular carcinoma (HCC) recurrence is highly correlated with increased mortality. Microvascular invasion (MVI) is indicative of aggressive tumor biology in HCC. AIM: To construct an artificial neural network (ANN) capable of accurately predicting MVI presence in HCC using magnetic resonance imaging. METHODS: This study included 255 patients with HCC with tumors < 3 cm. Radiologists annotated the tumors on the T1-weighted plain MR images. Subsequently, a three-layer ANN was constructed using image features as inputs to predict MVI status in patients with HCC. Postoperative pathological examination is considered the gold standard for determining MVI. Receiver operating characteristic analysis was used to evaluate the effectiveness of the algorithm. RESULTS: Using the bagging strategy to vote for 50 classifier classification results, a prediction model yielded an area under the curve (AUC) of 0.79. Moreover, correlation analysis revealed that alpha-fetoprotein values and tumor volume were not significantly correlated with the occurrence of MVI, whereas tumor sphericity was significantly correlated with MVI (P < 0.01). CONCLUSION: Analysis of variable correlations regarding MVI in tumors with diameters < 3 cm should prioritize tumor sphericity. The ANN model demonstrated strong predictive MVI for patients with HCC (AUC = 0.79).

Preoperative Modified Frailty Index-11 versus EuroSCORE II in Predicting Postoperative Mortality and Complications in Elderly Patients Who Underwent Elective Open Cardiac Surgery: A Retrospective Cohort Study.

OBJECTIVE: To compare sensitivity, specificity, receiver operating characteristic (ROC), and area under the curve (AUC) values using the modified Frailty Index 11 (mFI-11), EuroSCORE II, and combined mFI-11 and EuroSCORE II to predict in-hospital mortality and composite morbidities. DESIGN: Retrospective cohort study SETTING: Songklanagarind Hospital, a tertiary care center in southern Thailand. PARTICIPANTS: Elderly patients age ≥60 years who underwent elective open-heart surgical procedures on a pump between January 2017 and December 2022 were included. INTERVENTIONS: ROC curves were constructed to evaluate the discriminatory power of EuroSCORE II and mFI-11 for predicting in-hospital mortality and postoperative complications. MEASUREMENTS AND MAIN RESULTS: The actual in-hospital mortality was 2.5% for all patients. The discriminative accuracy of mFI-11, EuroSCORE II, and combined mFI-11 with EuroSCORE II for predicting in-hospital mortality was good, with respective AUC values of 0.733 (95% confidence interval [CI], 0.6157-0.8499), 0.793 (95% CI, 0.6826-0.9026), and 0.78 (95% CI, 0.6686-0.893). The AUC of mFI-11 for predicting postoperative cardiac, respiratory, neurologic, and renal complications was 0.558 (95% CI, 0.5101-0.6063), 0.606 (95% CI, 0.5542-0.6581), 0.543 (95% CI, 0.4533-0.6337), and 0.652 (95% CI, 0.5859-0.7179), respectively, and that of EuroSCORE II was 0.553 (95% CI, 0.5038-0.6013), 0.631 (95% CI, 0.578-0.6836), 0.619 (95% CI, 0.5306-0.7076), and 0.702 (95% CI, 0.6378-0.7657), respectively. CONCLUSIONS: The mFI-11 and EuroSCORE II demonstrated good discrimination in ROC analysis, with EuroSCORE II showing superior predictive accuracy for in-hospital mortality in elderly elective cardiac surgery patients. However, neither score independently predicted mortality in multiple logistic regression, nor did combining them enhance predictive power significantly. Furthermore, both scores were less effective in predicting postoperative complications.

Prediction of neonatal outcomes using gestational age vs ACOG definitions of maternal disease severity in hypertensive disorders of pregnancy.

PURPOSE: Hypertensive disorders of pregnancy cause significant neonatal complications. Disease severity is often used to predict neonatal outcomes, however gestational age (GA) at delivery may be a better predictor. We aimed to assess whether disease severity or GA was more predictive of adverse neonatal outcomes. METHODS: We included 165 participants with confirmed HELLP syndrome or severe preeclampsia (sPE). Two predictive models were constructed to assess the ability of disease severity compared to GA to predict a composite adverse neonatal outcome. The composite outcome included low birth weight, SGA, IUGR, Apgar score, and neonatal death. RESULTS: Using severity as a predictor of binary neonatal outcome had an AUC of 0.73 (0.65-0.81), with a sensitivity (SE) of 70.3% and a specificity (SP) of 64.4%. For GA, we observed an AUC of 0.82 (0.75-0.89), with a SE of 75.7% and a SP of 76.7%. CONCLUSION: For the composite neonatal outcome, GA was a better predictor than ACOG diagnosis (severity). This observation underscores the need for further research to validate these findings in larger cohorts and to determine their applicability to maternal outcomes.

The Role of Preoperative Abdominal Ultrasound in the Preparation of Patients Undergoing Primary Metabolic and Bariatric Surgery: A Machine Learning Algorithm on 4418 Patients' Records.

BACKGROUND: The utility of preoperative abdominal ultrasonography (US) in evaluating patients with obesity before metabolic bariatric surgery (MBS) remains ambiguously defined. METHOD: Retrospective analysis whereby patients were classified into four groups based on ultrasound results. Group 1 had normal findings. Group 2 had non-significant findings that did not affect the planned procedure. Group 3 required additional or follow-up surgeries without changing the surgical plan. Group 4, impacting the procedure, needed further investigations and was subdivided into 4A, delaying surgery for more assessments, and 4B, altering or canceling the procedure due to critical findings. Machine learning techniques were utilized to identify variables. RESULTS: Four thousand four hundred eighteen patients' records were analyzed. Group 1 was 45.7%. Group 2, 35.7%; Group 3, 17.0%; Group 4, 1.5%, Group 4A, 0.8%; and Group 4B, 0.7%, where surgeries were either canceled (0.3%) or postponed (0.4%). The hyperparameter tuning process identified a Decision Tree classifier with a maximum tree depth of 7 as the most effective model. The model demonstrated high effectiveness in identifying patients who would benefit from preoperative ultrasound before MBS, with training and testing accuracies of 0.983 and 0.985. It also showed high precision (0.954), recall (0.962), F1 score (0.958), and an AUC of 0.976. CONCLUSION: Our study found that preoperative ultrasound demonstrated clinical utility for a subset of patients undergoing metabolic bariatric surgery. Specifically, 15.9% of the cohort benefited from the identification of chronic calculous cholecystitis, leading to concomitant cholecystectomy. Additionally, surgery was postponed in 1.4% of the cases due to other findings. While these findings indicate a potential benefit in certain cases, further research, including a cost-benefit analysis, is necessary to fully evaluate routine preoperative ultrasound's overall utility and economic impact in this patient population.

The crucial relationship between vancomycin minimum inhibitory concentration and therapeutic efficacy against methicillin-resistant coagulase-negative staphylococci.

The area under the curve (AUC)/minimum inhibitory concentration (MIC) ratio was used as an indicator of the clinical efficacy of vancomycin. However, the target AUC/MIC has not been set for methicillin-resistant coagulase-negative staphylococci (MR-CNS), and the effectiveness of vancomycin in strains with high MIC is unknown. Therefore, we aimed to investigate the relationship between the vancomycin MIC and therapeutic efficacy in patients with MR-CNS bacteremia. The primary outcome was the difference in treatment failure rate when the MR-CNS vancomycin MIC was 1 or 2 µg/mL. The treatment failure rate did not significantly differ between the two groups (MIC 1 vs. MIC 2: 27.0% vs. 31.0%; p = 0.779). As a result of multivariate analysis, AUC/MIC0-24 h ≤230 was extracted as risk factor for treatment failure, suggesting the importance of a sufficient initial loading dose and early blood concentration monitoring to increase AUC/MIC0-24 h for successful treatment.

Population pharmacokinetics of cyclosporine A in hematopoietic stem cell transplant recipients: A systematic review.

Cyclosporine A (CsA) is the prevalent immunosuppressive drug for preventing and treating graft-versus-host disease after hematopoietic stem cell transplantation (HSCT) in both children and adults. Population pharmacokinetic studies have identified covariates, owing to their large between-subject variability, facilitating individualized therapy. However, no review has summarized CsA's population pharmacokinetics post-HSCT. This systematic review aims to synthesize population pharmacokinetic studies of CsA therapy in HSCT recipients and explore influencing covariates. Thirteen studies, comprising five involving children, one involving both children and adults and seven involving adults, were included. The median apparent clearance in children surpassed that in adults, influenced notably by hematocrit level and body. While liver function impacted clearance, the effect was insignificant. Co-administration with cytochrome P450 enzyme inhibitors (e.g., fluconazole or itraconazole) decreased drug clearance, whereas inducers (e.g., rifampicin or rifapentine) increased it. Area under the curve analysis is recommended over trough concentration-based monitoring for HSCT recipients on CsA. In cases of insufficient trough concentration, additional sampling points are recommended for improved area under the curve estimation. Further studies are needed to evaluate the optimal sampling points required for the area under the curve estimation in CsA therapy post-HSCT.

Pharmacokinetic Simulation and Area under the Curve Estimation of Drugs Subject to Enterohepatic Circulation.

Enterohepatic circulation (EHC) is a complex process where drugs undergo secretion and reabsorption from the intestinal lumen multiple times, resulting in pharmacokinetic profiles with multiple peaks. The impact of EHC on area under the curve (AUC) has been a topic of extensive debate, questioning the suitability of conventional AUC estimation methods. Moreover, a universal model for accurately estimating AUC in EHC scenarios is lacking. To address this gap, we conducted a simulation study evaluating five empirical models under various sampling strategies to assess their performance in AUC estimation. Our results identify the most suitable model for EHC scenarios and underscore the critical role of meal-based sampling strategies in accurate AUC estimation. Additionally, we demonstrate that while the trapezoidal method performs comparably to other models with a large number of samples, alternative models are essential when sample numbers are limited. These findings not only illuminate how EHC influences AUC but also pave the way for the application of empirical models in real-world drug studies.

Blunted anticipatory stress responses on competition day in team sports athletes compared to individual sports athletes.

Knowledge of anticipatory stress responses before sports competitions is limited, thus this study examined the relationship between anticipatory stress in terms of salivary cortisol secretion in athletes on the morning of a competition and a comparison baseline day. Thirty-seven athletes collected three saliva samples over a 45-min period post-awakening (0, 30 and 45 min). Anticipatory stress was expressed as Area Under the Curve compared to ground (AUCg; total cortisol secretion). There was no significant difference in AUCg between baseline and competition days. However, a mixed two-factor ANOVA with day and sport type (individual vs. team) revealed a significant main effect of sport type (p < 0.01) and a significant interaction (p = 0.001). Individual athletes demonstrated increased AUCg on competition day compared to baseline, while team athletes demonstrated decreased AUCg on competition day compared to baseline. This blunting response was also observed when analysing the raw cortisol secretion levels upon awakening. These findings suggest there may be substantive differences in anticipatory stress between individual and team sport athletes.

"What's yours is mine": Partners' everyday emotional experiences and cortisol in older adult couples.

The existing literature consistently finds that emotional experiences and cortisol secretion are linked at the within-person level. Further, relationship partners tend to covary in emotional experience, and in cortisol secretion. However, we are only beginning to understand whether and how an individuals' emotions are linked to their relationship partners' cortisol secretion. In this project, we harmonized data from three intensive measurement studies originating from Canada and Germany to investigate the daily dynamics of emotions and cortisol within 321 older adult couples (age range=56-87 years). Three-level multilevel models accounted for the nested structure of the data (repeated assessments within individuals within couples). Actor-Partner Interdependence Models were used to examine the effect of own emotional experiences (actor effects) and partner emotional experiences (partner effects) on momentary and daily cortisol secretion. Adjusting for age, sex, education, comorbidities, assay version, diurnal cortisol rhythm, time spent together, medication, and time-varying behaviors that may increase cortisol secretion, results suggest that higher relationship partner's positive emotions are linked with lower momentary cortisol and total daily cortisol. Further, this association was stronger for older participants and those who reported higher relationship satisfaction. We did not find within-couple links between negative emotions and cortisol. Overall, our results suggest that one's relationship partner's positive emotional experience may be a protective factor for their physiological responding, and that these more fleeting and day-to-day fluctuations may accumulate over time, contributing to overall relationship satisfaction.

Lifelog-based daily step counts, walking speed, and metabolically healthy status.

Objective: Optimal metabolically healthy status is important to prevent various chronic diseases. This study investigated the association between lifelog-derived physical activity and metabolically healthy status. Methods: This cross-sectional study included 51 Korean adults aged 30-40 years with no history of chronic diseases. Physical activity data were obtained by the International Physical Activity Questionnaire-Short Form (IPAQ-SF). Lifelog-derived physical activity was defined by step counts and walking speed for 1 week, as recorded by the Samsung Health application on both the Samsung Galaxy Fit2 and mobile phones. Participants without metabolic syndrome components were categorized as the metabolically healthy group (n = 31) and the remaining participants as the metabolically unhealthy group (n = 20). Prevalence ratios and 95% confidence intervals were estimated using Poisson regression models. The predictive ability of each physical activity measure was evaluated according to the area under the curve (AUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) values. Results: Among the physical activity measures, lifelog-derived walking speed was significantly inversely associated with prevalent metabolically unhealthy status. The lifelog component model including walking speed, age, and sex had the highest AUC value for metabolically unhealthy status. Adding lifelog-derived step counts to the IPAQ-SF-derived metabolic equivalent (MET) model (including age, sex, and IPAQ-SF-METs) yielded 37% and 13% increases in the NRI and IDI values, respectively. Incorporating walking speed into the IPAQ-SF-derived MET model improved metabolically unhealthy status prediction by 42% and 21% in the NRI and IDI analyses, respectively. Conclusions: Slow walking speed derived from the lifelog was associated with a higher prevalence of metabolically unhealthy status. Lifelog-derived physical activity information may aid in identifying individuals with metabolic abnormalities.

Accuracy of salivary biomarkers in the diagnosis of periodontal status and coronary heart disease.

Inflammatory illnesses, such as periodontitis and atherosclerotic coronary heart disease (ASCHD), trigger the production of pro-inflammatory mediators. The aim of this study was to assess the accuracy of using salivary interleukin-1ß (IL-1ß), interleukin-18 (IL-18), and gasdermin D (GSDMD) in discerning patients with periodontitis with and without ASCHD from healthy individuals, and to assess their correlation with clinical periodontal parameters and low-density lipoprotein (LDL) levels. The study involved 120 participants: 30 were healthy subjects (control group, C), 30 had generalized periodontitis (group P), 30 had ASCHD and clinically healthy periodontium (group AS-C), and 30 had ASCHD and generalized periodontitis (group AS-P). Saliva and blood samples were collected, and periodontal characteristics such as plaque index, bleeding on probing, probing pocket depth, and clinical attachment loss were examined. IL-1ß, IL-18, and GSDMD levels from saliva were determined using ELISA. LDL levels were determined from the blood samples. Groups P, AS-C, and AS-P had higher levels of salivary IL-1ß, IL-18, and GSDMD than group C. The receiver operating characteristic (ROC) curves of all biomarkers showed high diagnostic accuracy, with a significant positive correlation with the clinical parameters and LDL levels. The observed correlations between the studied pro-inflammatory mediators and disease severity suggest that these biomarkers could serve as indicators of disease progression in conditions such as periodontitis and ASCHD.

Application of peripheral blood routine parameters in the diagnosis of influenza and Mycoplasma pneumoniae.

OBJECTIVES: Influenza and Mycoplasma pneumoniae infections often present concurrent and overlapping symptoms in clinical manifestations, making it crucial to accurately differentiate between the two in clinical practice. Therefore, this study aims to explore the potential of using peripheral blood routine parameters to effectively distinguish between influenza and Mycoplasma pneumoniae infections. METHODS: This study selected 209 influenza patients (IV group) and 214 Mycoplasma pneumoniae patients (MP group) from September 2023 to January 2024 at Nansha Division, the First Affiliated Hospital of Sun Yat-sen University. We conducted a routine blood-related index test on all research subjects to develop a diagnostic model. For normally distributed parameters, we used the T-test, and for non-normally distributed parameters, we used the Wilcoxon test. RESULTS: Based on an area under the curve (AUC) threshold of ≥ 0.7, we selected indices such as Lym# (lymphocyte count), Eos# (eosinophil percentage), Mon% (monocyte percentage), PLT (platelet count), HFC# (high fluorescent cell count), and PLR (platelet to lymphocyte ratio) to construct the model. Based on these indicators, we constructed a diagnostic algorithm named IV@MP using the random forest method. CONCLUSIONS: The diagnostic algorithm demonstrated excellent diagnostic performance and was validated in a new population, with an AUC of 0.845. In addition, we developed a web tool to facilitate the diagnosis of influenza and Mycoplasma pneumoniae infections. The results of this study provide an effective tool for clinical practice, enabling physicians to accurately diagnose and differentiate between influenza and Mycoplasma pneumoniae infection, thereby offering patients more precise treatment plans.

Automatic text classification of drug-induced liver injury using document-term matrix and XGBoost.

Introduction: Regulatory agencies generate a vast amount of textual data in the review process. For example, drug labeling serves as a valuable resource for regulatory agencies, such as U.S. Food and Drug Administration (FDA) and Europe Medical Agency (EMA), to communicate drug safety and effectiveness information to healthcare professionals and patients. Drug labeling also serves as a resource for pharmacovigilance and drug safety research. Automated text classification would significantly improve the analysis of drug labeling documents and conserve reviewer resources. Methods: We utilized artificial intelligence in this study to classify drug-induced liver injury (DILI)-related content from drug labeling documents based on FDA's DILIrank dataset. We employed text mining and XGBoost models and utilized the Preferred Terms of Medical queries for adverse event standards to simplify the elimination of common words and phrases while retaining medical standard terms for FDA and EMA drug label datasets. Then, we constructed a document term matrix using weights computed by Term Frequency-Inverse Document Frequency (TF-IDF) for each included word/term/token. Results: The automatic text classification model exhibited robust performance in predicting DILI, achieving cross-validation AUC scores exceeding 0.90 for both drug labels from FDA and EMA and literature abstracts from the Critical Assessment of Massive Data Analysis (CAMDA). Discussion: Moreover, the text mining and XGBoost functions demonstrated in this study can be applied to other text processing and classification tasks.

Automatic detection of facial expressions during the Cyberball paradigm in Borderline Personality Disorder: a pilot study.

Borderline Personality Disorder (BPD) symptoms include inappropriate control of anger and severe emotional dysregulation after rejection in daily life. Nevertheless, when using the Cyberball paradigm, a tossing game to simulate social exclusion, the seven basic emotions (happiness, sadness, anger, surprise, fear, disgust, and contempt) have not been exhaustively tracked out. It was hypothesized that these patients would show anger, contempt, and disgust during the condition of exclusion versus the condition of inclusion. When facial emotions are automatically detected by Artificial Intelligence, "blending", -or a mixture of at least two emotions- and "masking", -or showing happiness while expressing negative emotions- may be most easily traced expecting higher percentages during exclusion rather than inclusion. Therefore, face videos of fourteen patients diagnosed with BPD (26 ± 6 years old), recorded while playing the tossing game, were analyzed by the FaceReader software. The comparison of conditions highlighted an interaction for anger: it increased during inclusion and decreased during exclusion. During exclusion, the masking of surprise; i.e., displaying happiness while feeling surprised, was significantly more expressed. Furthermore, disgust and contempt were inversely correlated with greater difficulties in emotion regulation and symptomatology, respectively. Therefore, the automatic detection of emotional expressions during both conditions could be useful in rendering diagnostic guidelines in clinical scenarios.