Employing a random forest model to forecast the likelihood of coronary artery lesions in Kawasaki disease: a study centered on four biomarkers

Background:Kawasaki disease is an acute immune vasculitis,which is more common in children under 5 years old.Kawasaki disease mainly affects the cardiovascular system,especially the coronary arteries.Once coronary artery damage occurs,it can significantly impact the patient’s prognosis.Therefore,in some countries and regions,Kawasaki disease has become a common acquired heart disease.Methods:First,univariate analysis was conducted on each predictive factor.Then,Least Absolute Shrinkage and Selection Operator and random forest algorithms were used to screen all predictive factors,and the prediction model was evaluated using receiver operating characteristic curve,calibration curve,and Decision Curve Analysis.Results:This study,based on data from 228 Kawasaki disease patients,utilized a random forest model to identify four predictive factors:white blood cell count,creatine kinase isoenzyme MB,albumin,and neutrophil count.These factors were used to construct a prediction model,which achieved an area under the curve of 0.743.Conclusions:We developed a forest plot based on white blood cell count,creatine kinase isoenzyme MB,albumin,and neutrophil count to effectively predict the occurrence of coronary artery lesions in Kawasaki disease.

Successful cardiopulmonary resuscitation combined with thrombolysis for massive pulmonary embolism during peri-cardiac arrest

Dear editor,Pulmonary thromboembolism(PTE)has become the third most lethal type of thrombotic disease in the world.[1]Thrombolysis has been advised for high-risk PTE patients,however due to its unique pathophysiology,which differs from other atherosclerotic diseases,this can still be a challenge for PTE patients.[2]We reported two cases with a high-risk of PTE that had the onset of or progressed to cardiac arrest(CA)during hospitalization and were successfully managed with cardiopulmonary resuscitation(CPR)combined with thrombolytic therapy.

Recombinant tissue plasminogen activator plus heparin compared with heparin alone for patients with acute submassive pulmonary embolism: one-year outcome

OBJECTIVE: To evaluate the long-term effects of thrombolysis on patients with submassive pulmonary embolism (PE)。 METHODS: Data of 136 patients with acute submassive PE and low risk of bleeding were prospectively collected from January 2005 to October 2011 in a single medical center。 Patients received recombinant tissue plasminogen activator (r-tPA) plus low molecular weight heparin (LMWH, TT group, n = 79) or LMWH alone (AT group, n = 57), depending on treating physician's recommendation and patient's preference。 Echocardiography was performed at admission, 24 h, 6 and 12 months to evaluate right ventricular function。 Computed tomography pulmonary angiography (CTPA) and lung perfusion scan were performed on admission, at 7 days, 6 and 12 months to evaluate clot burden。 RESULTS: Seventy-nine patients received r-tPA plus LMWH (TT group) while 57 received LMWH alone (AT group)。 The baseline characteristics and risk factors did not differ between the two groups。 Respiratory rate, heart rate, and systolic blood pressure improved within two hours in both groups。 Systolic pulmonary arterial pressure and tricuspid regurgitation improved to a greater extent in the TT group at 24 h, and at 12 months (P < 0。001), as compared to those in the AT group。 At one week, and 12 months, clot burden decreased more in AT group, as compared to that in AT group (P < 0.001)。 There was no death due to bleeding in both groups。 Recurrent PE were similar in both groups (2.5% in TT vs。 1.8% in AT)。 The rates of minor hemorrhages were 6.3% in TT group and 1.8% in AT group (P < 0.05)。 In submassive PE patient who has low risk of bleeding, thrombolysis plus anticoagulation can lead to greater improvement of right ventricular dysfunction and clot burden reduction as compared to anticoagulation therapy alone。

基于Logistic回归的外阴上皮内瘤变危险度评价模型的构建

目的应用临床资料和流行病学资料建立外阴上皮内瘤变危险度评价模型。方法模型样本为101例外阴上皮内瘤变患者、134例外阴良性病变者及150例外阴完全正常者。对其临床资料和流行病学资料进行单因素分析,通过受试者工作特征(ROC)曲线获取影响因素的截断值,构建Logistic回归模型,采用混淆矩阵回代分析验证模型的正确率。结果经Logistic回归分析,年龄、吸烟、人类乳头瘤病毒(HPV)感染、外阴瘙痒、外阴触血、外阴肿物、外阴疼痛、绝经与否、中性粒细胞绝对值(GRA)、中性粒细胞与淋巴细胞的比值(NLR)、血清糖类抗原125(CA125)指标进入回归方程。回归模型最大似然比(LR)检验具有差异(P<0.05)。混淆矩阵回代分析结果发现该模型总体预测正确率为78.44%(302/385)。结论应用临床资料和流行病学资料构建外阴上皮内瘤变危险度评价模型,能有效提高外阴上皮内瘤变正确筛查率。

Genetic, environmental and other risk factors for progression of retinitis pigmentosa

Retinitis pigmentosa(RP) is a commonly inherited disease of the retina, which is characterized by progressive loss of visual function due to specific genetic mutations. There are many risk factors that may have effect on the progression of RP, such as inheritance patterns, genotype, gender, age, smoking, physical activity, and other demographic and environmental factors. Baseline visual field conditions, changes of ellipsoid zone, photoreceptor layer thickness, and choroidal structure are reported to be the phenotype risk factors for RP progression. Moreover, aqueous flare and high-sensitivity C-reactive protein are probable inflammation biomarkers for assessing the progression of RP. Increased oxidative stress is considered to be one of the potential factors for the existence of RP. The risk factors can be combined to form a corresponding prediction model to predict disease progression. This review is to summarize the current literature that studies the genetic, environmental, phenotypic, demographic, inflammatory and other risk factors of RP progression and discuss the most reliable risk factors that could provide predictive models.

An unprecedented dumbbell-shaped pentadeca-nuclear W-Er heterometal cluster stabilizing nanoscale hexameric arsenotungstate aggregate and electrochemical sensing properties of its conductive hybrid film-modified electrode

Cluster-based functional materials have made remarkable progress owing to their wonderful structures and distinctive physicochemical performances,one of on-going advancements of which is basically driven by synthetic chemistry of exploring and constructing novel nanosized gigantic polyoxometalate(POM)aggregates.In this article,an unprecedented nanoscale hexameric arsenotungstate aggregate Na_(9)K_(16)H_(4)[Er_(0.5)K_(0.5)(H_(2)O)_(7)][Er_(5)W10O_(2)6(H_(2)O)_(14)][B-α-AsW_(9)O_(33)]_(6)·102H_(2)O(1)has been synthesized by the combined synthetic strategy of simultaneously using the arsenotungstate precursor and simple tungstate material in a highly acidic aqueous solution.The{[Er_(5)W_(10)O_(26)(H_(2)O)_(14)][B-α-AsW_(9)O_(33)]6}31-polyanion in 1 consists of an intriguing dumbbell-shaped pentadeca-nuclear W-Er heterometal{Er_(5)W10O_(2)6(H_(2)O)_(14)}^(23+)cluster connecting six trilacunary[B-α-AsW_(9)O_(33)]^(9-)moieties,which has never been seen previously.Furthermore,through electropolymerization of 1 and pyrrole on the conductive substrate,a thickness-controllable and robust 1-PPY(PPY=polypyrrole)hybrid film was successfully prepared,which represents the first POM-PPY film assembled from high-nuclear lanthanide(Ln)encapsulated POM and PPY hitherto.The 1-PPY film-based electrochemical biosensor exhibits a favorable recognition performance for ochratoxin A in multiple media.This work not only provides a feasible combined synthetic strategy of the POM precursor and simple tungstate material for constructing complicated multi-Ln-inserted POM aggregates,but also offers a promising electrochemical platform constructed from POM-based conductive films for identifying trace biomolecules in complex environments.