Enhancing Sepsis prognosis: Integrating social determinants and demographic variables into a comprehensive model for critically ill patients.

BACKGROUND: The Sequential Organ Failure Assessment (SOFA) score monitors organ failure and defines sepsis but may not fully capture factors influencing sepsis mortality. Socioeconomic and demographic impacts on sepsis outcomes have been highlighted recently. OBJECTIVE: To evaluate the prognostic value of SOFA scores against demographic and social health determinants for predicting sepsis mortality in critically ill patients, and to assess if a combined model increases predictive accuracy. METHODS: The study utilized retrospective data from the MIMIC-IV database and prospective external validation from the Penn State Health cohort. A Random Forest model incorporating SOFA scores, demographic/social data, and the Charlson Comorbidity Index was trained and validated. FINDINGS: In the MIMIC-IV dataset of 32,970 sepsis patients, 6,824 (20.7%) died within 30 days. A model including demographic, socioeconomic, and comorbidity data with SOFA scores improved predictive accuracy beyond SOFA scores alone. Day 2 SOFA, age, weight, and comorbidities were significant predictors. External validation showed consistent performance, highlighting the importance of delta SOFA between days 1 and 3. CONCLUSION: Adding patient-specific demographic and socioeconomic information to clinical metrics significantly improves sepsis mortality prediction. This suggests a more comprehensive, multidimensional prognostic approach is needed for accurate sepsis outcome predictions.

Opportunities and challenges in using NIH's Dietary Supplement Label Database for research on non-nutrient ingredients: Use case for ingredients in honeybee products.

Background: Non-nutrient bioactive ingredients of foods such as bee products are often of interest in preclinical and clinical research to explore their possible beneficial effects. The National Institute of Health's Dietary Supplement Label Database (DSLD) contains over 165,000 labels of dietary supplements marketed in the United States of America (US), including declarations on labels for many of these ingredients, including those in honeybee products which have been used in foods and traditional medicines for centuries worldwide and are now also appearing in dietary supplements. Methods: This article presents a use case for honeybee products that describes and tests the utility of the DSLD and other databases available in the US as research tools for identifying and quantifying the prevalence of such ingredients.. It focuses on the limitations to the information on product composition in these databases and describes how to code the ingredients using the LanguaL™ or FoodEx2 description and classification systems and the strengths and limitations of information on honeybee product ingredients, including propolis, bee pollen, royal jelly, beeswax, and bee venom. Results and Conclusions: Codes for the ingredients are provided for identifying their presence in LanguaL™ or FoodEx2 ontologies used in Europe and elsewhere. The prevalence of dietary supplement products containing these ingredients in DSLD and on the US market is low compared to some other products and ingredients. Unfortunately label declarations in DSLD do not provide quantitative information and so the data can be used only to screen for their presence, but cannot be used for quantitative exposure estimates by researchers and regulators .

Highly accurate classification and discovery of microbial protein-coding gene functions using FunGeneTyper: an extensible deep learning framework.

High-throughput DNA sequencing technologies decode tremendous amounts of microbial protein-coding gene sequences. However, accurately assigning protein functions to novel gene sequences remain a challenge. To this end, we developed FunGeneTyper, an extensible framework with two new deep learning models (i.e., FunTrans and FunRep), structured databases, and supporting resources for achieving highly accurate (Accuracy > 0.99, F1-score > 0.97) and fine-grained classification of antibiotic resistance genes (ARGs) and virulence factor genes. Using an experimentally confirmed dataset of ARGs comprising remote homologous sequences as the test set, our framework achieves by-far-the-best performance in the discovery of new ARGs from human gut (F1-score: 0.6948), wastewater (0.6072), and soil (0.5445) microbiomes, beating the state-of-the-art bioinformatics tools and sequence alignment-based (F1-score: 0.0556-0.5065) and domain-based (F1-score: 0.2630-0.5224) annotation approaches. Furthermore, our framework is implemented as a lightweight, privacy-preserving, and plug-and-play neural network module, facilitating its versatility and accessibility to developers and users worldwide. We anticipate widespread utilization of FunGeneTyper (https://github.com/emblab-westlake/FunGeneTyper) for precise classification of protein-coding gene functions and the discovery of numerous valuable enzymes. This advancement will have a significant impact on various fields, including microbiome research, biotechnology, metagenomics, and bioinformatics.

Update of the Xylella spp. host plant database - Systematic literature search up to 31 December 2023.

This scientific report provides an update of the Xylella spp. host plant database, aiming to provide information and scientific support to risk assessors, risk managers and researchers dealing with Xylella spp. Upon a mandate of the European Commission, EFSA created and regularly updates a database of host plant species of Xylella spp. The current mandate covers the period 2021-2026. This report is related to the 10th version of the database published in Zenodo in the EFSA Knowledge Junction community, covering literature published from 1 July 2023 up to 31 December 2023, and recent Europhyt outbreak notifications. Informative data have been extracted from 39 selected publications. Sixteen new host plants, five genera and one family were identified and added to the database. They were naturally infected by X. fastidiosa subsp. fastidiosa or unknown either in Portugal or the United States. No additional data were retrieved for X. taiwanensis, and no additional multilocus sequence types (STs) were identified worldwide. New information on the tolerant/resistant response of plant species to X. fastidiosa infection were added to the database. The Xylella spp. host plant species were listed in different categories based on the number and type of detection methods applied for each finding. The overall number of Xylella spp. host plants determined with at least two different detection methods or positive with one method either by sequencing or pure culture isolation (category A), reaches now 451 plant species, 204 genera and 70 families. Such numbers rise to 712 plant species, 312 genera and 89 families if considered regardless of the detection methods applied (category E).

CANTATAdb 3.0: an updated repository of plant long noncoding RNAs.

CANTATAdb 3.0 is an updated database of plant long non-coding RNAs (lncRNAs), containing 571,688 lncRNAs identified across 108 species, including 100 Magnoliopsida (flowering plants), a significant expansion from the previous version. A notable feature is the inclusion of 112,980 lncRNAs that are expressed specifically in certain plant organs or embryos, indicating their potential role in development and organ-specific processes. In addition, CANTATAdb 3.0 includes 74,886 pairs of evolutionarily conserved lncRNAs found across 47 species and inferred from genome-genome alignments as well as conserved lncRNAs obtained with a similarity-search approach in 5,479 species pairs, which would further aid in the selection of lncRNAs for functional studies. Interestingly, we find that conserved lncRNAs with tissue specific expression patterns tend to occupy the same plant organ across different species, pointing toward conserved biological roles. The database now offers extended search capabilities, and downloadable data in popular formats, further facilitating research on plant lncRNAs.

DNA profiling in India: Addressing issues of sample preservation, databasing, marker selection, & statistical approaches.

DNA technology is the gold standard with respect to the identification of individuals from biological evidence. The technology offers the convenience of a universally similar approach and methodology for analysis across the globe. However, the technology has not realised its full potential in India due to the lack of a DNA database and lacunae in sample collection and preservation from the scene of crime and victims (especially those of sexual assault). Further, statistical interpretation of DNA results is non-existent in the majority of cases. Though the latest technologies and developments in the field of DNA analysis are being adopted and implemented,very little has been enacted practically to improve optimise sample collection and preservation. This article discusses current casework scenarios that highlight the pitfalls and ambiguous areas in the field of DNA analysis, especially with respect DNA databases, sampling, andstatistical approaches to genetic data analysis. Possible solutions and mitigation measures are suggested.

Construction and validation of a nomogram for predicting the prognosis of breast cancer patients who received adjuvant therapy: an analysis based on the SEER database.

Background: Breast cancer is the most common malignant tumor in women globally. Despite advances in primary treatment, the role of adjuvant therapy in reducing recurrence and improving survival is critical; however, there is a notable lack of tailored prognostic models for patients receiving adjuvant therapy. This study used the Surveillance, Epidemiology, and End Results (SEER) database to develop a prognostic nomogram for breast cancer patients receiving adjuvant therapy. Methods: The data of breast cancer patients who received adjuvant therapy after surgery in 2014-2015 were extracted from the SEER database. Univariate Cox regression identified significant prognostic variables that were further refined by least absolute shrinkage and selection operator (LASSO) regression and cross-validation analyses. These variables were incorporated into a multivariate Cox regression analysis to establish the predictive model. This model was visualized and validated using various statistical measures. Results: A total of 54,960 patients were included in the study, with 38,472 in the training set and 16,488 in the validation set. Age, sex, race, marital status, grade, tumor (T) stage, lymph node (N) stage, subtype, and radiotherapy were found to be significant independent risk factors of 1-, 3-, and 5-year overall survival (OS). The receiver operating characteristic curve area for 1-, 3-, and 5-year OS was >0.76 in both sets. The consistency index values were 0.768 and 0.763 for the training and validation sets, respectively. The calibration curves showed good fit, and the nomogram exhibited substantial clinical utility. Conclusions: Incorporating various significant factors, the constructed nomogram was able to effectively predict the prognosis of breast cancer patients who received adjuvant therapy. This nomogram extends understandings of complex prognosis scenarios. In addition, it could enhance personalized treatment plans and assist in patient counseling.

Establishment of a logistic regression model nomogram for clinicopathological characteristics and risk factors with axillary lymph node metastasis in T1 locally advanced breast cancer: a retrospective study.

Background: Although the research reports on locally advanced breast cancer (LABC) are increasing year by year, there are few reports on T1 LABC axillary lymph node metastasis (ALNM). By establishing a prediction model for T1 LABC ALNM, this study provides a reference value for the probability of ALNM of related patients, which helps clinicians to develop a more effective and individualized treatment plan for LABC. Methods: Cases with pathologically confirmed T1 breast cancer (BC) between 2010 and 2015 in the Surveillance, Epidemiology, and End Results (SEER) database were identified. Logistic regression was used to analyze the correlation between LABC lymph node metastasis and every factor, and the odds ratio (OR) and 95% confidence interval (CI) were used to identify any influencing factors. A nomogram was drawn after incorporating meaningful factors identified in multivariate logistic regression into the model. The receiver operating characteristic (ROC) curve of the model was drawn, and the area under the curve (AUC) and its 95% CI were calculated. Hosmer-Lemeshow goodness-of-fit test and clinical decision curve analysis (DCA) were performed. The results were validated in the validation group. Results: A total of 200,933 female T1 BC patients were included in this study. Univariate and multivariate logistic regression analysis of T1 BC showed that progesterone receptor (PR)-negative, race, age, lobular carcinoma, micropapillary ductal carcinoma, axillary tail tumor, poor differentiation, and larger tumor diameter increased the probability of ALNM in T1 LABC. A predictive nomogram was established using the above predictors, the AUC of the modeling group was 0.739 (95% CI: 0.732-0.747), and when the AUC cut-off value was 0.026, the specificity and sensitivity of the model were 65.78% and 69.99%, respectively. Validation of the model showed that the AUC of the validation group (n=60,280) was 0.741. When all the risk factors were met, the predicted probability of N2-N3 was 50.40%. Conclusions: In this study, it was found that PR-negative, Black race, age, lobular carcinoma, micropapillary ductal carcinoma, axillary tail tumor, poor differentiation, and tumor diameter increased the probability of large lymph node metastasis in T1 LABC small tumors.

A DPC Database Study on the Safety of Atezolizumab/Bevacizumab/Carboplatin/Paclitaxel in Non-small Cell Lung Cancer in Japanese Patients.

INTRODUCTION: Atezolizumab, bevacizumab, carboplatin, and paclitaxel (ABCP) combination therapy is a standard of care for advanced non-squamous non-small cell lung cancer (NSQ-NSCLC); however, the lack of safety data limits its clinical application in Japan. METHODS: This study compared the safety of ABCP with that of bevacizumab, carboplatin, and paclitaxel (BCP) combination for the treatment of advanced NSQ-NSCLC in Japanese patients by evaluating the clinical background and incidence of adverse events (AEs) based on data extracted from the Diagnosis Procedure Combination (DPC) database. Incidence rates and restricted mean survival times (RMSTs) for up to 1 year were analyzed for 19 clinically important AEs. Covariates were adjusted using the inverse probability weighting method. RESULTS: A search conducted using the International Statistical Classification of Diseases and Related Health Problems 10th Revision codes identified 350,987 patients, of whom 202 were included in the ABCP cohort and 232 in the BCP cohort. Among the 19 AEs, the incidence of skin disorder and febrile neutropenia (FN) was significantly higher in the ABCP cohort versus the BCP cohort. The adjusted incidence rate ratios were 2.65 [95% confidence interval (CI) 1.43-4.91] for skin disorder and 1.70 (95% CI 1.01-2.85) for FN. The adjusted RMST differences were - 64.2 days (95% CI - 93.0 to - 35.4 days) and - 46.0 days (95% CI - 73.5 to - 18.5 days) for skin disorder and FN, respectively. These results were comparable to those of other pivotal clinical trials. CONCLUSIONS: The findings of this DPC database study highlight the safety of ABCP in Japanese clinical practice, and this methodology may facilitate more efficient research in real-world settings. TRIAL REGISTRATION: UMIN Clinical Trials Registry ID UMIN000041507.

Prevalence of Tendon Rupture and Tendinopathies Among Patients with Atherosclerotic Cardiovascular Disease Derived From United States Administrative Claims Data.

INTRODUCTION: The prevalence of tendon rupture and tendinopathies (TRT) has not been determined in a large population of patients with atherosclerotic cardiovascular disease (ASCVD). We investigated TRT prevalence among patients with ASCVD and in the general population, using data from the Symphony Health Integrated Dataverse, a large US medical and pharmacy claims database. METHODS: This retrospective, observational study included patients aged ≥ 19 years from the claims database during the identification period (January 2019 to December 2020) and 12 months of continuous enrollment. The primary outcome was evidence of TRT in the 12 months following the index date (first ASCVD diagnosis in the ASCVD cohort; first claim in the claims database in the overall population). Diagnostic codes (ICD-10 and/or CPT) were used to define ASCVD and TRT diagnosis. RESULTS: The ASCVD cohort and overall population included 5,589,273 and 61,715,843 patients, respectively. In the ASCVD cohort, use of medications with a potential or known association with TRT was identified in 67.9% (statins), 17.7% (corticosteroids), and 16.7% (fluoroquinolones) of patients. Bempedoic acid use was reported in 1556 (< 0.1%) patients. TRT prevalence during 12-month follow-up was 3.4% (ASCVD cohort) and 1.9% (overall population). Among patients with ASCVD, 83.5% experienced TRT in only one region of the body. Factors most associated with TRT in the ASCVD cohort were increasing age, most notably in those aged 45-|64 years (odds ratio [OR] 2.19; 95% confidence interval [CI] 2.07-2.32), obesity (OR 1.51; 95% CI 1.50-1.53), and rheumatoid arthritis (OR 1.47; 95% CI 1.45-1.79). Use of statins or bempedoic acid was not associated with increased TRT risk. CONCLUSION: Patients with ASCVD may have greater risk of TRT than the general population, which may be driven by an increased prevalence of comorbidities and use of medications with a potential or known association with TRT.

Patients with atherosclerosis, the main cause of heart attacks, strokes, and peripheral vascular disease, typically require several drugs to control the disease. Some of the drugs used to treat atherosclerosis have been linked to a higher occurrence of tendon tears (or ruptures) or swelling/inflammation of the tendons (tendinopathies). However, there may be other factors present in these patients that increase the risk of tendon injuries that are not related to these drugs. This study used the medical records of over 5.5 million patients with atherosclerosis and over 63 million patients reflecting the general population in the United States to determine the prevalence of tendon injury. Additionally, the researchers looked at other factors that might be related to a higher risk of tendon injury in each group. Over a 12-month period, tendon injuries occurred in 3.4% of patients with atherosclerosis and 1.8% of patients in the general population. In patients with atherosclerosis, factors such as being obese, older (45­64 years), or having rheumatoid arthritis were also linked to an increased risk of tendon injuries. There was no association seen between statin or bempedoic acid use and tendon injuries. These results may help healthcare providers to determine the underlying risk of tendon injuries and guide treatment of this patient population.

Burden of illness in patients with pulmonary hypertension due to interstitial lung disease: a real-world analysis.

BACKGROUND: Pulmonary hypertension due to interstitial lung disease (PH-ILD) is associated with high rates of respiratory failure and death. Healthcare resource utilization (HCRU) and cost data are needed to characterize PH-ILD disease burden. METHODS: A retrospective cohort analysis of the Truven Health MarketScan® Commercial Claims and Encounters Database and Medicare Supplemental Database between June 2015 to June 2019 was conducted. Patients with ILD were identified and indexed based on their first claim with a PH diagnosis. Patients were required to be 18 years of age on the index date and continuously enrolled for 12-months pre- and post-index. Patients were excluded for having a PH diagnosis prior to ILD diagnosis or the presence of other non-ILD, PH-associated conditions. Treatment patterns, HCRU, and healthcare costs were compared between the 12 months pre- versus 12 months post-index date. RESULTS: In total, 122 patients with PH-ILD were included (mean [SD] age, 63.7 [16.6] years; female, 64.8%). The same medication classes were most frequently used both pre- and post-index (corticosteroids: pre-index 43.4%, post-index 53.5%; calcium channel blockers: 25.4%, 36.9%; oxygen: 12.3%, 25.4%). All-cause hospitalizations increased 2-fold, with 29.5% of patients hospitalized pre-index vs. 59.0% post-index (P < 0.0001). Intensive care unit (ICU) utilization increased from 6.6 to 17.2% (P = 0.0433). Mean inpatient visits increased from 0.5 (SD, 0.9) to 1.1 (1.3) (P < 0.0001); length of stay (days) increased from 5.4 (5.9) to 7.5 (11.6) (P < 0.0001); bed days from 2.5 (6.6) to 8.0 (16.3) (P < 0.0001); ICU days from 3.8 (2.3) to 7.0 (13.2) (P = 0.0362); and outpatient visits from 24.5 (16.8) to 32.9 (21.8) (P < 0.0001). Mean (SD) total all-cause healthcare costs increased from $43,201 ($98,604) pre-index to $108,387 ($190,673) post-index (P < 0.0001); this was largely driven by hospitalizations (which increased from a mean [SD] of $13,133 [$28,752] to $63,218 [$75,639] [P < 0.0001]) and outpatient costs ($16,150 [$75,639] to $25,604 [$93,964] [P < 0.0001]). CONCLUSION: PH-ILD contributes to a high HCRU and cost burden. Timely identification, management, and treatment are needed to mitigate the clinical and economic consequences of PH-ILD development and progression.

Census-Dependent Mortality of Ventilated Patients With COVID-19 in Israel: Noninterventional Observational Cohort Study.

BACKGROUND: The COVID-19 pandemic led to several surges in the mass hospitalization rate. Extreme increases in hospital admissions without adequate medical resources may increase mortality. No study has addressed the impact of daily census of ventilated patients on mortality in the context of the pandemic in a nationwide setting. OBJECTIVE: This study aimed to determine whether daily census of ventilated patients affected COVID-19 mortality rates nationwide in Israel. METHODS: We conducted a cohort study using nationwide, public-domain, population-based COVID-19 data of hospitalized patients from an Israeli database from March 11, 2020, until February 11, 2021. We included all COVID-19 hospital admissions, classified as mild to severe per the Centers for Diseases Control and Prevention classification irrespective of whether they were mechanically ventilated. Outcome measures were daily death rates and death rates expressed as a percentage of ventilated patients. RESULTS: During the study period (338 days from March 11, 2020, to February 11, 2021), 715,743 patients contracted and were clinically confirmed as having COVID-19. Among them, 5577 (0.78%) patients died. In total, 3398 patients were ventilated because of severe COVID-19. Daily mortality correlated with daily census of ventilated patients (R2=0.828, P<.001). The daily percent mortality of ventilated patients also correlated with the daily census of ventilated patients (R2=0.365, P<.001)-backward multiple regression analysis demonstrated that this positive correlation was still highly significant even when correcting for the average age or gender of ventilated patients (R2=0.4328, P<.001) or for the surge in their number. Overall, 40% of the variation in mortality was explained by variations in the daily census of ventilated patients. ANOVA revealed that at less than 50 ventilated patients per day, the daily mortality of ventilated patients was slightly above 5%, and it nearly doubled (10%) with 50-149 patients; moreover, in all categories of ≥200 patients ventilated per day, it more than tripled at ≥15% (P<.001). CONCLUSIONS: Daily mortality rates per ventilated patient increased with an increase in the number of ventilated patients, suggesting the saturation of medical resources. Policy makers should be aware that expanding medical services without adequate resources may increase mortality. Governments should perform similar analyses to provide indicators of system saturation, although further validation of these results might be needed to use this indicator to drive public policy.

Disproportionality Analysis of Nusinersen in the Food and Drug Administration Adverse Event Reporting System: A Real-World Postmarketing Pharmacovigilance Assessment.

BACKGROUND: Nusinersen is the first drug for precise targeted therapy of spinal muscular atrophy, a rare disease that occurs in one of 10,000 to 20,000 live births. Therefore, thorough and comprehensive reports on the safety of nusinersen in large, real-world populations are necessary. This study aimed to mine the adverse event (AE) signals related to nusinersen through the Food and Drug Administration Adverse Event Reporting System (FAERS) database. METHODS: We extracted reports of AEs with nusinersen as the primary suspect from FAERS between December 2016 and March 2023. Reporting odds ratio (ROR) and Bayesian confidence propagation neural network (BCPNN) were used for AE signal detection. RESULTS: We extracted a total of 4807 suspected AE cases with nusinersen as the primary suspect from the FAERS database. Among them, 106 positive signals were obtained using the ROR and BCPNN. The highest frequency reported systemic organ class was general disorders and administration site conditions. Common clinical AEs of nusinersen were detected in the FAERS database, such as pneumonia, vomiting, back pain, headache, pyrexia, and post-lumbar puncture syndrome. In addition, we identified potential unexpected serious AEs through disproportionality analysis, including sepsis, seizure, epilepsy, brain injury, cardiorespiratory arrest, and cardiac arrest. CONCLUSIONS: Analyzing large amounts of real-world data from the FAERS database, we identified potential new AEs of nusinersen by disproportionate analysis. It is advantageous for health care professionals and pharmacists to concentrate on effectively managing high-risk AEs of nusinersen, improve medication levels in clinical settings, and uphold patient medication safety.

Association of Definitive Radiotherapy for Esophageal Cancer and the Incidence of Secondary Head and Neck Cancers: A SEER Population-Based Study.

Background: Impact of radiotherapy (RT) for esophageal cancer (EC) patients on the development of secondary head and neck cancer (SHNC) remains equivocal. The objective of this study was to investigate the link between definitive RT used for EC treatment and subsequent SHNC. Methods: This study was conducted using the Surveillance, Epidemiology, and End Results (SEER) database to collect the data of primary EC patients. Fine-Gray competing risk regression and standardized incidence ratio (SIR) and propensity score matching (PSM) method were used to match SHNC patients with only primary head and neck cancer (HNC) patients. Overall survival (OS) rates were applied by Kaplan-Meier analysis. Results: In total, 14,158 EC patients from the SEER database were included, of which 9,239 patients (65.3%) received RT and 4,919 patients (34.7%) received no radiation therapy (NRT). After a 12-month latency period, 110 patients (1.2%) in the RT group and 36 patients (0.7%) in the NRT group experienced the development of SHNC. In individuals with primary EC, there was an increased incidence of SHNC compared to the general US population (SIR = 5.95, 95% confidence interval (CI): 5.15 - 6.84). Specifically, the SIR for SHNC was 8.04 (95% CI: 6.78 - 9.47) in the RT group and 3.51 (95% CI: 2.64 - 4.58) in the NRT group. Patients who developed SHNC after RT exhibited significantly lower OS compared to those after NRT. Following PSM, the OS of patients who developed SHNC after RT remained significantly lower than that of matched patients with only primary HNC. Conclusion: An association was discovered between RT for EC and increased long-term risk of SHNC. This work enables radiation oncologists to implement mitigation strategies to reduce the long-term risk of SHNC in patients who have received RT following primary EC.

A Comparison of Machine Learning Approaches for predicting Hepatotoxicity potential using Chemical Structure and Targeted Transcriptomic Data.

Animal toxicity testing is time and resource intensive, making it difficult to keep pace with the number of substances requiring assessment. Machine learning (ML) models that use chemical structure information and high-throughput experimental data can be helpful in predicting potential toxicity . However, much of the toxicity data used to train ML models is biased with an unequal balance of positives and negatives primarily since substances selected for in vivo testing are expected to elicit some toxicity effect. To investigate the impact this bias had on predictive performance, various sampling approaches were used to balance in vivo toxicity data as part of a supervised ML workflow to predict hepatotoxicity outcomes from chemical structure and/or targeted transcriptomic data. From the chronic, subchronic, developmental, multigenerational reproductive, and subacute repeat-dose testing toxicity outcomes with a minimum of 50 positive and 50 negative substances, 18 different study-toxicity outcome combinations were evaluated in up to 7 ML models. These included Artificial Neural Networks, Random Forests, Bernouilli Naïve Bayes, Gradient Boosting, and Support Vector classification algorithms which were compared with a local approach, Generalised Read-Across (GenRA), a similarity-weighted k-Nearest Neighbour (k-NN) method. The mean CV F1 performance for unbalanced data across all classifiers and descriptors for chronic liver effects was 0.735 (0.0395 SD). Mean CV F1 performance dropped to 0.639 (0.073 SD) with over-sampling approaches though the poorer performance of KNN approaches in some cases contributed to the observed decrease (mean CV F1 performance excluding KNN was 0.697 (0.072 SD)). With under-sampling approaches, the mean CV F1 was 0.523 (0.083 SD). For developmental liver effects, the mean CV F1 performance was much lower with 0.089 (0.111 SD) for unbalanced approaches and 0.149 (0.084 SD) for under-sampling. Over-sampling approaches led to an increase in mean CV F1 performance (0.234, (0.107 SD)) for developmental liver toxicity. Model performance was found to be dependent on dataset, model type, balancing approach and feature selection. Accordingly tailoring ML workflows for predicting toxicity should consider class imbalance and rely on simpler classifiers first.

The Risk of Infection-Caused Mortality in Gastric Adenocarcinoma: A Population-Based Study.

Background: Gastric adenocarcinoma (GAC) is a deadly tumor. Postoperative complications, including infections, worsen its prognosis and may affect overall survival. Little is known about perioperative complications as well as modifiable and non-modifiable risk factors. Early detection and treatment of these risk factors may affect overall survival and mortality. Methods: We extracted GAC patient's data from the Surveillance, Epidemiology, and End Results (SEER) database and analyzed using Pearson's Chi-square, Cox regression, Kaplan-Meier, and binary regression methods in SPSS. Results: At the time of analysis, 59,580 GAC patients were identified, of which 854 died of infection. Overall, mean survival in months was better for younger patients, age < 50 years vs. ≥ 50 years (60.45 vs. 56.75), and in females vs. males (65.23 vs. 53.24). The multivariate analysis showed that the risk of infectious mortality was higher in patients with age ≥ 50 years (hazard ratio (HR): 3.137; 95% confidence interval (CI): 2.178 - 4.517), not treated with chemotherapy (HR: 1.669; 95% CI: 1.356 - 2.056), or surgery (HR: 1.412; 95% CI:1.132 - 1.761) and unstaged patients (HR: 1.699; 95% CI: 1.278 - 2.258). In contrast, the mortality risk was lower in females (HR: 0.658; 95% CI: 0.561 - 0.773) and married patients (HR: 0.627; 95% CI: 0.506 - 0.778). The probability of infection was higher in older patients (odds ratio (OR) of 2.094 in ≥ 50 years), other races in comparison to Whites and Blacks (OR: 1.226), lesser curvature, not other specified (NOS) as a primary site (OR: 1.325), and patients not receiving chemotherapy (OR: 1.258). Conclusion: Older, unmarried males with GAC who are not treated with chemotherapy or surgery are at a higher risk for infection-caused mortality and should be given special attention while receiving treatment.

Association between blood urea nitrogen to serum albumin ratio and in-hospital mortality in critical patients with diabetic ketoacidosis: a retrospective analysis of the eICU database.

Background: This study aimed to investigate the association between blood urea nitrogen to serum albumin ratio (BAR) and the risk of in-hospital mortality in patients with diabetic ketoacidosis. Methods: A total of 3,962 diabetic ketoacidosis patients from the eICU Collaborative Research Database were included in this analysis. The primary outcome was in-hospital death. Results: Over a median length of hospital stay of 3.1 days, 86 in-hospital deaths were identified. One unit increase in LnBAR was positively associated with the risk of in-hospital death (hazard ratio [HR], 1.82 [95% CI, 1.42-2.34]). Furthermore, a nonlinear, consistently increasing correlation between elevated BAR and in-hospital mortality was observed (P for trend =0.005 after multiple-adjusted). When BAR was categorized into quartiles, the higher risk of in-hospital death (multiple-adjusted HR, 1.99 [95% CI, (1.1-3.6)]) was found in participants in quartiles 3 to 4 (BAR≥6.28) compared with those in quartiles 1 to 2 (BAR<6.28). In the subgroup analysis, the LnBAR-hospital death association was significantly stronger in participants without kidney insufficiency (yes versus no, P-interaction=0.023). Conclusion: There was a significant and positive association between BAR and the risk of in-hospital death in patients with diabetic ketoacidosis. Notably, the strength of this association was intensified among those without kidney insufficiency.

Higher-Order Spectral Analysis Combined with a Convolution Neural Network for Atrial Fibrillation Detection-Preliminary Study.

The global burden of atrial fibrillation (AFIB) is constantly increasing, and its early detection is still a challenge for public health and motivates researchers to improve methods for automatic AFIB prediction and management. This work proposes higher-order spectra analysis, especially the bispectrum of electrocardiogram (ECG) signals combined with the convolution neural network (CNN) for AFIB detection. Like other biomedical signals, ECG is non-stationary, non-linear, and non-Gaussian in nature, so the spectra of higher-order cumulants, in this case, bispectra, preserve valuable features. The two-dimensional (2D) bispectrum images were applied as input for the two CNN architectures with the output AFIB vs. no-AFIB: the pre-trained modified GoogLeNet and the proposed CNN called AFIB-NET. The MIT-BIH Atrial Fibrillation Database (AFDB) was used to evaluate the performance of the proposed methodology. AFIB-NET detected atrial fibrillation with a sensitivity of 95.3%, a specificity of 93.7%, and an area under the receiver operating characteristic (ROC) of 98.3%, while for GoogLeNet results for sensitivity and specificity were equal to 96.7%, 82%, respectively, and the area under ROC was equal to 96.7%. According to preliminary studies, bispectrum images as input to 2D CNN can be successfully used for AFIB rhythm detection.