Interpretability-based machine learning for predicting the risk of death from pulmonary inflammation in Chinese intensive care unit patients.

Objective: The objective of this research was to create a machine learning predictive model that could be easily interpreted in order to precisely determine the risk of premature death in patients receiving intensive care after pulmonary inflammation. Methods: In this study, information from the China intensive care units (ICU) Open Source database was used to examine data from 2790 patients who had infections between January 2019 and December 2020. A 7:3 ratio was used to randomly assign the whole patient population to training and validation groups. This study used six machine learning techniques: logistic regression, random forest, gradient boosting tree, extreme gradient boosting tree (XGBoost), multilayer perceptron, and K-nearest neighbor. A cross-validation grid search method was used to search the parameters in each model. Eight metrics were used to assess the models' performance: accuracy, precision, recall, F1 score, area under the curve (AUC) value, Brier score, Jordon's index, and calibration slope. The machine methods were ranked based on how well they performed in each of these metrics. The best-performing models were selected for interpretation using both the Shapley Additive exPlanations (SHAP) and Local interpretable model-agnostic explanations (LIME) interpretable techniques. Results: A subset of the study cohort's patients (120/1668, or 7.19%) died in the hospital following screening for inclusion and exclusion criteria. Using a cross-validated grid search to evaluate the six machine learning techniques, XGBoost showed good discriminative ability, achieving an accuracy score of 0.889 (0.874-0.904), precision score of 0.871 (0.849-0.893), recall score of 0.913 (0.890-0.936), F1 score of 0.891 (0.876-0.906), and AUC of 0.956 (0.939-0.973). Additionally, XGBoost exhibited excellent performance with a Brier score of 0.050, Jordon index of 0.947, and calibration slope of 1.074. It was also possible to create an interactive internet page using the XGBoost model. Conclusion: By identifying patients at higher risk of early mortality, machine learning-based mortality risk prediction models have the potential to significantly improve patient care by directing clinical decision making and enabling early detection of survival and mortality issues in patients with pulmonary inflammation disease.

Biomass-derived porous material synthesized by one-step calcination method for the magnetic solid-phase extraction of polychlorinated biphenyls in water.

Recent findings unfold that biomass materials with the micro/mesoporous structure were often treated as adsorbents for organic substances. In this work, a one-step calcination method was adopted in the preparation of magnetic porous green bean biomass material. It has the properties of magnetism and porosity after the addition of Co(NO3 )2 and high-temperature calcination. A variety of characterizations have been operated, including energy dispersive X-ray detector, vibrating sample magnetometer, Fourier transform infrared spectroscopy, Brunauer-Emmett-Teller analysis, and so on. It has a specific surface area of 168.1611 m2 /g and a pore volume of 0.1764 cm3 /g. The material was used in magnetic solid-phase extraction of three polychlorinated biphenyls: 2-chlorobiphenyl, 4-chlorobiphenyl, and 2,2,5-trichlorobiphenyl. Several factors were investigated, such as material amount, eluents, adsorption time, solution pH, salinity, and reusability. Under optimized conditions, good recoveries (90.24-93.34%) were achieved with the relative standard deviation in a range from 2.30 to 4.83%. Three real water samples (tap, river, and lake water) were tested to verify the accuracy of the method. This method can be successfully used in the analysis of some polychlorinated biphenyls congeners in water.

Effects of moxibustion on reproduction and metabolism of polycystic ovary syndrome: a protocol for meta-analysis and systematic review.

INTRODUCTION: Polycystic ovary syndrome (PCOS) is one of the most common endocrinopathy in women of reproductive age. Recently, moxibustion, as a complementary and alternative therapy, has been commonly used in assisted reproduction and improvement of metabolic abnormalities in patients with PCOS. Currently, intervention efficacy of the use of moxibustion in PCOS treatment still remains controversial due to lack of high-quality evidence. Consequently, this study protocol was designed to objectively review and evaluate the effectiveness and safety of moxibustion treatment for PCOS. METHODS AND ANALYSIS: Electronic searches will be carried out from inception to May 2021 in the online databases of The Cochrane Library, PubMed, EMBASE, Chinese Biomedical Literature, Chongqing VIP Chinese Science and Technology Periodical Database (VIP) and China National Knowledge Infrastructure. The Chinese Clinical Trial Registry Center and Clinical Trials will be used for searching ongoing trials. Randomised controlled trials and the first period in randomised cross-over trials involving any type of moxibustion for patients with PCOS will be included. Primary outcomes will be the ovulation rate, pregnancy rate and sex hormone levels, and secondary outcomes will be changes in clinical symptoms and metabolic indicators, total effective rate and the incidences of side effects and adverse events. Briefly, two reviewers will independently conduct study selection and data extraction, and the risk of bias will be assessed. Prior to the formal meta-analysis, the heterogeneity of included studies will be assessed. Review Manager Statistical Software (RevMan) V.5.3 will be used for data processing. Finally, the Grading of Recommendations Assessment, Development and Evaluation method will be applied to evaluate the quality of evidence. ETHICS AND DISSEMINATION: Ethical approval is not necessary since this study is designed as a systematic review. This study will be disseminated by a peer-review journal or conference presentation.

Comparative effectiveness of herb-partitioned moxibustion plus lifestyle modification treatment for patients with simple obesity: A study protocol for a randomized controlled trial.

INTRODUCTION: Obesity is a global public health issue, which results in many health complications. Moxibustion may serve as an alternative management for simple obesity, where pharmacological therapy is always difficult to be accepted by the majority of obese patients based on its safety. However, the effects of herb-partitioned moxibustion as obesity intervention have not been confirmed. This study is designed as a single-blinded, 3-dummy randomized controlled trial to evaluate the efficacy and safety of herb-partitioned moxibustion plus lifestyle modification treatment in patients with simple obesity. METHODS AND ANALYSIS: This study will be a randomized, controlled trial conducted from April, 2019 to April, 2021 that includes 108 participants who have simple obesity and meet the eligibility criteria. The participants will be randomly divided into 3 treatment groups: heat application group, medicated plaster group, or herb-partitioned moxibustion group. Each treatment will last 4 weeks. The primary outcomes will be the clinical effectiveness. The secondary outcome measures include participants' obesity-related indicators, the IWQOL-Lite scale, and the syndrome score of Traditional Chinese Medicine. Adverse events will be recorded during the intervention period. ETHICS AND DISSEMINATION: Ethical approval of this study was granted by the Ethics Committee of Hubei Provincial Hospital of Traditional Chinese Medicine on 15 November 2018 (Ethics Reference No: HBZY2018-C24-01). Written informed consents will be provided by all participants before they were enrolled in this study. TRIAL REGISTRATION NUMBER: NCT04606680.

[The correlation between obstructive sleep apnea and olfaction function].

Objective:To investigate the olfactory function of patients with obstructive sleep apnea（OSA） without any treatment, and to providereference for clinical olfactory research. Methods:One hundred and thirty-one participants underwent polysomnography（PSG） overnight, and were divided into OSA group （observation group） and non-OSA group （control group） according to the results. The two groups were examined by the Korean version of the olfactory stick test Ⅱ（Korean Bersion） of Sniffin Sticks Test （KVSS Test Ⅱ）. SPSS 26.0 statistical software were used to analyze the data. Results:There was a significant difference in the incidence of olfactory disorders between the observation group and the control group（χ²=12.000, P=0.001）. The rate of olfactory disorders in patients with severe OSA was significantly higher than that in patients with mild OSA（P<0.05）, and the proportion of patients with olfactory disorders increased with the increase of OSA severity（χ²=10.672, P=0.001）. There was no statistically significant difference between the observation group and the control group in the severity of olfactory disorders（hyposmia and anosmia）（P=1.000）. KVSS between two groups of Test Ⅱ total score has no statistical difference（t=1.166, P=0.249）, the sense of smell recognition scores also has no statistical difference（t=1.598, P=0.116）, but the olfactory threshold score and olfactory cognition scores were statistically significant（t=5.346, t=6.405, P<0.001）. Conclusion:OSA has a negative effect on the sense of smell, and the severity of OSA is positively correlated with the incidence of olfactory disorder. OSA olfactory disorder is mainly anosmia, and the main manifestation is the decrease of olfactory cognitive scores.

Carboxyl-functionalized biochar derived from walnut shells with enhanced aqueous adsorption of sulfonamide antibiotics.

The novel HNO3-modifitied biochar (NBC) was synthesized from walnut shell. The NBC was characterized from scanning electron microscopy, Fourier transform infrared spectroscopy, X-ray diffraction, X-ray photoelectron spectroscopy and Raman spectrum. The NBC was then used in the adsorption of sulfadiazine, sulfamethazine and sulfachloropyridazine from aqueous solution. The material surface has carbon/oxygen-contained groups, which is benefit for the adsorption. The results showed the adsorption ability of NBC on three sulfonamides were 32, 46, and 40 mg g-1, respectively. The kinetic was found to follow the Elovich model and the isotherm conformed Freundlich. Adsorption was more favorable at weak acidic solution. The interactions mainly include π-π EDA, electrostatic interaction, Lewis acid-base interaction, hydrophobic interaction and H-bond.

Deep learning of lumbar spine X-ray for osteopenia and osteoporosis screening: A multicenter retrospective cohort study.

Osteoporosis is a prevalent but underdiagnosed condition. As compared to dual-energy X-ray absorptiometry (DXA) measures, we aimed to develop a deep convolutional neural network (DCNN) model to classify osteopenia and osteoporosis with the use of lumbar spine X-ray images. Herein, we developed the DCNN models based on the training dataset, which comprising 1616 lumbar spine X-ray images from 808 postmenopausal women (aged 50 to 92 years). DXA-derived bone mineral density (BMD) measures were used as the reference standard. We categorized patients into three groups according to DXA BMD T-score: normal (T ≥ -1.0), osteopenia (-2.5 < T < -1.0), and osteoporosis (T ≤ -2.5). T-scores were calculated by using the BMD dataset of young Chinese female aged 20-40 years as a reference. A 3-class DCNN model was trained to classify normal BMD, osteoporosis, and osteopenia. Model performance was tested in a validation dataset (204 images from 102 patients) and two test datasets (396 images from 198 patients and 348 images from 147 patients respectively). Model performance was assessed by the receiver operating characteristic (ROC) curve analysis. The results showed that in the test dataset 1, the model diagnosing osteoporosis achieved an AUC of 0.767 (95% confidence interval [CI]: 0.701-0.824) with sensitivity of 73.7% (95% CI: 62.3-83.1), the model diagnosing osteopenia achieved an AUC of 0.787 (95% CI: 0.723-0.842) with sensitivity of 81.8% (95% CI: 67.3-91.8); In the test dataset 2, the model diagnosing osteoporosis yielded an AUC of 0.726 (95% CI: 0.646-0.796) with sensitivity of 68.4% (95% CI: 54.8-80.1), the model diagnosing osteopenia yielded an AUC of 0.810 (95% CI, 0.737-0.870) with sensitivity of 85.3% (95% CI, 68.9-95.0). Accordingly, a deep learning diagnostic network may have the potential in screening osteoporosis and osteopenia based on lumbar spine radiographs. However, further studies are necessary to verify and improve the diagnostic performance of DCNN models.

Deactivating Influence of 3-O-Glycosyl Substituent on Anomeric Reactivity of Thiomannoside Observed in Oligomannoside Synthesis.

It has been long recognized that, in chemical glycosylation, the anomeric reactivity of glycosyl donor can be influenced greatly by protecting groups. As opposed to the effects of protecting groups, we report herein a finding on how O-glycosyl substituent can affect the reactivity of oligosaccharyl donor, which in turn should have impact on convergent assembly of oligosaccharide. During our synthetic efforts toward Pichia holstii oligomannoside, a type of α-1,3-linked dimannosyl thioglycosides was found to exhibit unexpected low reactivity toward the activation of NIS/TMSOTf. This observation prompted us to perform a series of comparative reactivity studies, which attributed the donor deactivation to the presence of 3-O-glycosyl substituent, by comparison with O-acetyl group and O-glycosidic linkages at C-4/C-6 positions. To rationalize the unusual phenomenon, we hypothesize that O-glycosyl moiety at C-3 could destabilize the oxocarbenium ion intermediate by additionally increasing the O2-C2-C3-O3 torsional strain, which was further supported by DFT calculation of the hypothetical 4H3-like oxocarbeniums. The observed deactivating influence provides a basis for estimation of donor reactivity and logical selection of synthetic strategy in oligosaccharide synthesis. Following this finding, we opted to use an iterative strategy for the synthesis of targeted pentamannoside 1 by using monomeric thiomannosides that ensured sufficient reactivity.

An ylide transformation of rhodium(I) carbene: enantioselective three-component reaction through trapping of rhodium(I)-associated ammonium ylides by ß-nitroacrylates.

The chiral Rh(I)-diene-catalyzed asymmetric three-component reaction of aryldiazoacetates, aromatic amines, and ß-nitroacrylates was achieved to obtain γ-nitro-α-amino-succinates in good yields and with high diastereo- and enantioselectivity. This reaction is proposed to proceed through the enantioselective trapping of Rh(I)-associated ammonium ylides by nitroacrylates. This new transformation represents the first example of Rh(I)-carbene-induced ylide transformation.