Prediction models for postoperative delirium in elderly patients with machine-learning algorithms and SHapley Additive exPlanations.

Postoperative delirium (POD) is a common and severe complication in elderly patients with hip fractures. Identifying high-risk patients with POD can help improve the outcome of patients with hip fractures. We conducted a retrospective study on elderly patients (≥65 years of age) who underwent orthopedic surgery with hip fracture between January 2014 and August 2019. Conventional logistic regression and five machine-learning algorithms were used to construct prediction models of POD. A nomogram for POD prediction was built with the logistic regression method. The area under the receiver operating characteristic curve (AUC-ROC), accuracy, sensitivity, and precision were calculated to evaluate different models. Feature importance of individuals was interpreted using Shapley Additive Explanations (SHAP). About 797 patients were enrolled in the study, with the incidence of POD at 9.28% (74/797). The age, renal insufficiency, chronic obstructive pulmonary disease (COPD), use of antipsychotics, lactate dehydrogenase (LDH), and C-reactive protein are used to build a nomogram for POD with an AUC of 0.71. The AUCs of five machine-learning models are 0.81 (Random Forest), 0.80 (GBM), 0.68 (AdaBoost), 0.77 (XGBoost), and 0.70 (SVM). The sensitivities of the six models range from 68.8% (logistic regression and SVM) to 91.9% (Random Forest). The precisions of the six machine-learning models range from 18.3% (logistic regression) to 67.8% (SVM). Six prediction models of POD in patients with hip fractures were constructed using logistic regression and five machine-learning algorithms. The application of machine-learning algorithms could provide convenient POD risk stratification to benefit elderly hip fracture patients.

Syntheses, crystal structures and magnetic properties of three novel cobalt(II) complexes containing imidazole derivative groups.

Three Co(II) complexes with the formulas: {[Co2(Bib)3Cl2]Cl(CH3COO)}.CH3OH.H2O (1), [Co2(Bib)3Cl2]Cl2.(CH3OH)2.H2O, (2) and [Co3K1(Tib)2(CH3COO)6]PF6 (3), were obtained by self-assembly of a cobalt(II) salt with Bib and Tib (Bib = 1,3-bis(4,5-dihydro-1H-imidazol-2-yl)benzene; Tib = 1,3,5-tris(4,5-dihydro-1H-imidazol-2-yl)benzene) and were structurally and magnetically characterized. X-Ray single-crystal diffraction showed that each Co(II) ion was in a highly distorted tetrahedral coordination geometry with a cis-trans ratio of 1 : 2 from the Bib ligand, which functioned in a bidentate fashion in the binuclear triple-helical [Co2(Bib)3Cl2]2+ cations in 1 and 2. In the [Co3K1(Tib)2(CH3COO)6]- anions in 3, each Co(II) ion was also in a highly distorted tetrahedral coordination geometry and the Tib ligands acted in an offset fashion in C, C, C and A, A, A coordination to the Co(II) ions with pi-pi stacking interactions between two benzene rings from the Tib ligand in the cluster cation. Each Tib ligand in a cluster unit acted as a tridentate entity to coordinate three Co(II) ions resulting in a cylinder-like cluster structure. The intermolecular hydrogen bonds in the solid-state resulted in the well-shaped 2D layer network which formed a honeycomb in 1, the 3D supramolecular architecture which was connected to the 2D sheet into 3D in 2 and the 3D supramolecular architecture, which was extended into a well-shaped 2D honeycomb layer network in 3. The results from magnetic data, in the high-temperature region, showed that 1 and 2 obeyed the Curie-Weiss law with a Weiss constants theta = -12.3, and -9.8 K and a Curie constants C = 5.31 and 5.32 cm3 K mo1(-1), respectively, indicating antiferromagnetic interactions between adjacent cobalt(II) ions. Both complex 1 and 2 showed magnetic ordering at low temperature due to the canting effect. The zero-field AC magnetic susceptibility measurements for 1 and 2 displayed a maximum which was frequency dependent owing to a slow relaxation process, which could be caused by either domain wall movements or spin-glass behaviours.