Introduction: Artificial Intelligence (AI) and machine learning (ML) are used extensively in HICs to detect and control antibiotic resistance (AMR) in laboratories and clinical institutions. ML is designed to predict outcome variables using an algorithm to enable “machines” to learn the “rules” from the data. ML is increasingly being applied in intensive care units to identify AMR and to assist empiric antibiotic therapy. This study aimed to evaluate the performance of ML models for predicting AMR bacteria and resistance to antibiotics in two Vietnamese hospitals. Patients and Methods: A cross-sectional study combined with retrospective was conducted from 1st January 2020 to 30th June 2022. Five models were developed to predict antibiotic resistance of bacterial infections of ICU patients. Two datasets were prepared to predict AMR bacteria and antibiotics with ML models. The performance of the prediction models was evaluated by various indicators (sensitivity, specificity, precision, accuracy, F1-score, PRC, AuROC, and NormMCC) to determine the optimal time point for data selection. Python version 3.8 was used for statistical analyses. Results: The accuracy, F1-score, AuROC, and normMMC of LightGBM, XGBoost, and Random Forest models were higher than those of other models in both datasets. In both datasets 1 and 2, accuracy, F1-score, AuROC and normMCC of the XGBoost model were the highest among five models (from 0.890 to 1.000). Only Random Forest models had specificity scores higher than 0.850. High scores of sensitivity, accuracy, precision, F1-score, and normMCC indicated that the models were making accurate predictions for datasets 1 and 2. Conclusion: XGBoost, LightGBM, and Random Forest were the best-performed machine learning models to predict antibiotic resistance of bacterial infections of ICUs patients using the patients’ EMRs.

Bone metabolism is regulated by endocrine systems, so people exposed to polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) may suffer adverse effects on bones and teeth. We reviewed previous publications in which effects of PCDD/Fs and dioxin-like polychlorinated biphenyls on the teeth and bones of animals and humans were found. The aim was to identify future research directions, particularly for epidemiological studies of populations exposed to PCDD/Fs in the environment. Exposure of fetuses to PCDD/Fs may affect odontogenesis, particularly enamel formation, but the effects of PCDD/Fs on bone genesis are limited to palatine bone. Exposure to PCDD/Fs in milk may affect both teeth and bones, but the effects on bones may be reversible. Exposure to high PCDD/F concentrations even during adulthood may adversely affect teeth. Exposure to PCDD/Fs may induce osteogenesis and improve bone properties because the disrupting effects of PCDD/Fs cause bone remodeling and vitamin D activation. More studies involving humans are required to investigate previously found associations between the PCDD/F concentrations humans are exposed to and biological markers for teeth and bones, including metabolites of vitamin D.

To clarify the influence of dioxin exposure on brain morphometry, the present study investigated associations between dioxin exposure at high levels and brain structural irregularities in 32 Vietnamese men. Two exposure markers were used: blood dioxin levels, as a marker of exposure in adulthood, and perinatal dioxin exposure, estimated by maternal residency in a dioxin-contaminated area during pregnancy. All subjects underwent brain magnetic resonance imaging (MRI) scans. We analyzed correlations between regional gray matter volumes and blood dioxin levels, and compared regional volumes between men with and without perinatal dioxin exposure using the voxel-based morphometry (VBM) tool from Statistical Parametric Mapping 12 (SPM12). Blood 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) was associated with low volume of the medial temporal pole and fusiform gyrus. Toxic equivalency (TEQ)-PCDDs were correlated with low medial temporal pole volume. However, 1,2,3,4,7,8-HxCDD was associated with high middle frontal gyrus and cerebellum volume. In men with perinatal dioxin exposure, the left inferior frontal gyrus pars orbitalis volume was significantly lower than in those without perinatal exposure. These results suggest that dioxin exposure during the perinatal period and in adulthood may alter regional brain volume, which might lead to cognitive deficits and unusual social emotional behavior in Vietnamese men living in dioxin-contaminated areas.

We examined children in Da Nang, a dioxin contamination hotspot in Vietnam, twice at 5 and 8 years of age, and investigated sex- and age-dependent differences in the effects of dioxin exposure on attention deficit hyperactivity disorder (ADHD) symptoms. We also studied autistic traits in children with ADHD symptoms. A total of 163 children participated in follow-up surveys at 5 and 8 years of age and were included in the present analysis. ADHD symptoms were assessed using an ADHD rating scale with inattention and hyperactivity-and-impulsivity (hyperactivity) subscales. Autistic behaviors were evaluated using the Autism Spectrum Rating Scale (ASRS). Perinatal dioxin exposure was indicated by dioxin levels in maternal breast milk. In boys, hyperactivity scores were significantly higher in the high 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) group only at 5 years of age. In girls, hyperactivity scores at 8 years of age were significantly higher in the high TCDD group, which was significantly associated with those at 5 years of age. In girls, ASRS unusual behavior scores were significantly higher with higher TCDD exposure and hyperactivity scores at 8 years of age. These results suggest that high perinatal TCDD exposure may increase ADHD likelihood and autistic traits, particularly in girls of 7–8 years of age.

Large amounts of herbicides containing polychlorinated dibenzo-p-dioxins and dibenzo-furans (PCDD/Fs) were sprayed in South Vietnam during the Vietnam War. Levels of PCDD/Fs in the environment of South Vietnam remained high even 40 years later. A total of 861 breast milk samples (597 from three areas sprayed with herbicides, Quang Tri, Da Nang, and Bien Hoa, and 264 from three unsprayed areas in North Vietnam) were collected between 2007 and 2015 and the PCDD/F concentrations in the samples were determined. Levels of TEQ-PCDD/Fs and 17 PCDD/F congeners were higher in the sprayed area samples than the unsprayed area samples. We found particular PCDD/F congener patterns for different areas. High tetrachlorodibenzo-p-dioxin (TCDD) concentrations were found in Bien Hoa, high TCDD and 1,2,3,6,7,8-hexadibenzo-p-dioxin concentrations were found in Da Nang, and high 1,2,3,4,6,7,8-heptadibenzo-p-dioxin concentrations were found in Quan Tri. High 1,2,3,4,7,8-hexadibenzofuran and 1,2,3,4,6,7,8-heptadibenzofuran concentrations were also found in Da Nang and Quang Tri. However, breast feeding may have caused associations between the TCDD and polychlorinated dibenzofuran congener concentrations. Advanced statistical analysis will need to be performed in future to assess the characteristic PCDD/F congener profiles in breast milk samples from areas of Vietnam previously sprayed with herbicides.