Objective To explore the value of CT-based radiomics and machine learning in predicting the severity of community acquired pneumonia(CAP)in children.Methods The clinical and imaging data of 158 patients diagnosed with CAP in children were analyzed retrospectively.All patients were randomly divided into training set(n=110)and validation set(n=48)in a ratio of 7︰3.Radiomics features were outlined and extracted using 3D Slicer software,and feature selection was achieved using maximum relevance and minimum redundancy(MRMR)and least absolute shrinkage and selection operator(LASSO)algorithms.The construction of the nomogram model and the machine learning combined model was performed by combining clinical features and Radiomics score(Radscore),and its performance was evaluated and validated.Results The area under the curve(AUC)of the clinical model,the radiomics model and the nomogram model in the validation set were classified as 0.82,0.86 and 0.91,respectively.The AUC of the combined multi-layer perceptron(MLP),random forest(RF),and adaptive boosting(ADB)models were 0.926,0.934 and 0.917,respectively in the validation set.Conclusion Radiomics combined with clinical data is expected to be a novel predictor of the severity of CAP in children.MLP,RF and ABD machine learning algorithms can further enable model performance.

Objective To explore the value of a CT radiomics model in differentiating Xpert MTB/RIF-negative pulmonary tuber-culosis(PTB)from non-tuberculous mycobacteria pulmonary disease(NTM-PD).Methods A retrospective analysis was performed on 90 patients with Xpert MTB/RIF-negative PTB and 127 patients with NTM-PD.All patients were randomly divided into training set and testing set at the ratio of 7∶3.Radiomics features were extracted from chest CT images.Feature dimensionality reduction and selection were sequentially performed using the maximum relevance and minimum redundancy(mRMR)algorithm and the least absolute shrinkage and selection operator(LASSO)algorithm.Clinical,radiomics,and combined models were constructed by multi-variable logistic regression.The area under the curve(AUC)of receiver operating characteristic(ROC)curve was utilized to assess the model diagnostic performance.Calibration curves were used to evaluate model stability,and the decision curve analysis(DCA)was used to evaluate the clinical utility.Results The combined model had the highest diagnostic performance in both training and testing sets,with AUC of 0.90 and 0.86,respectively,which were higher than clinical and radiomics models.The calibration curve showed that the combined model had a good consistency between the predicted and the actual observations,and DCA revealed the highest clinical benefit.Conclusion The clinical-radiomics combined model has excellent predictive ability in differentiating Xpert MTB/RIF-negative PTB from NTM-PD,which can provide robust support for clinical diagnosis.

Objective To explore the value of CT-based radiomics and machine learning in predicting the severity of community acquired pneumonia(CAP)in children.Methods The clinical and imaging data of 158 patients diagnosed with CAP in children were analyzed retrospectively.All patients were randomly divided into training set(n=110)and validation set(n=48)in a ratio of 7︰3.Radiomics features were outlined and extracted using 3D Slicer software,and feature selection was achieved using maximum relevance and minimum redundancy(MRMR)and least absolute shrinkage and selection operator(LASSO)algorithms.The construction of the nomogram model and the machine learning combined model was performed by combining clinical features and Radiomics score(Radscore),and its performance was evaluated and validated.Results The area under the curve(AUC)of the clinical model,the radiomics model and the nomogram model in the validation set were classified as 0.82,0.86 and 0.91,respectively.The AUC of the combined multi-layer perceptron(MLP),random forest(RF),and adaptive boosting(ADB)models were 0.926,0.934 and 0.917,respectively in the validation set.Conclusion Radiomics combined with clinical data is expected to be a novel predictor of the severity of CAP in children.MLP,RF and ABD machine learning algorithms can further enable model performance.

Objective To explore the value of a CT radiomics model in differentiating Xpert MTB/RIF-negative pulmonary tuber-culosis(PTB)from non-tuberculous mycobacteria pulmonary disease(NTM-PD).Methods A retrospective analysis was performed on 90 patients with Xpert MTB/RIF-negative PTB and 127 patients with NTM-PD.All patients were randomly divided into training set and testing set at the ratio of 7∶3.Radiomics features were extracted from chest CT images.Feature dimensionality reduction and selection were sequentially performed using the maximum relevance and minimum redundancy(mRMR)algorithm and the least absolute shrinkage and selection operator(LASSO)algorithm.Clinical,radiomics,and combined models were constructed by multi-variable logistic regression.The area under the curve(AUC)of receiver operating characteristic(ROC)curve was utilized to assess the model diagnostic performance.Calibration curves were used to evaluate model stability,and the decision curve analysis(DCA)was used to evaluate the clinical utility.Results The combined model had the highest diagnostic performance in both training and testing sets,with AUC of 0.90 and 0.86,respectively,which were higher than clinical and radiomics models.The calibration curve showed that the combined model had a good consistency between the predicted and the actual observations,and DCA revealed the highest clinical benefit.Conclusion The clinical-radiomics combined model has excellent predictive ability in differentiating Xpert MTB/RIF-negative PTB from NTM-PD,which can provide robust support for clinical diagnosis.

Objective To investigate the value of CT radiomics models based on intranodal and perinodal in distinguishing pulmonary cryptococcosis(PC)from lung adenocarcinoma.Methods A total of 194 patients,including PC(n=94)and lung adenocarcinoma(n=100),confirmed by surgical or puncture pathology were analyzed retrospectively and randomly divided into training set and test set in a ratio of 7∶3.3D Slicer was used to delineate and extract the intranodal and perinodal volume of interest(VOI)radiomics features within a 5 mm range.The minimum redundancy maximum relevance(mRMR)and least absolute shrinkage and selection operator(LASSO)methods were used to dimensionality reduction.Statistically significant indicators were screened by one-way logistic regression and further incorporated into the multifactor logistic regression model.Support vector machine(SVM)was used to construct the intranodal image-based radiomics model,the perinodal image-based radiomics model,the intranodal-and-perinodal image-based radiomics model,and the combined model.The diagnostic efficacy of each model was evaluated by receiver operating characteristic(ROC)curve.Results In the test set,the area under the curve(AUC)of the clinical imaging model,the intranodal image-based radiomics model,the perinodal image-based radiomics model,the intranodal-and-perinodal image-based radiomics model,and the combined model were 0.84,0.88,0.85,0.90,and 0.94,respectively.Conclusion The combined model based on clinical imaging features,intranodal and perinodal radiomics features can improve the ability of differentiating PC from lung adenocarcinoma.

Objective To compare if the Montreal cognitive assessment (MoCA) performed in the morning or afternoon would affect abnormal rate of cognitive function in the elderly with stroke.Methods A total of 378 senile patients (≥ 65 years) with acute ischemic stroke and low NIHSS score (≤ 3) were enrolled in the prospective study,which was held in the Department of Neurology at Cangzhou Hospital of Integrated Traditional Chinese Medicine.MoCA was assessed after one month of hospitalization.Based on the time of MoCA assessment,all patients were randomly divided into the group A (assessed in the morning,9 am-12 am) and the group B (assessed in the afternoon,12 am to 5 pm).Clinical data were collected,and RANKIN scale (mRS) examination was performed.Moreover,patients were further divided into severe cognitive impairment (SCI) subgroup (score ＜ 20),mild cognitive impairment (MCI) subgroup (score 20-25) and no cognitive impairment (NCI) subgroup (score ＞ 26) according to the MoCA score.Results There were 189 patients in the group A (50％),and 189 cases in the group B (50％).There was no significant difference in age,gender,education level,disability (mRS score ＜ 1),history of hypertension,diabetes,hyperlipidemia,smoking and atrial fibrillation between the two groups.Based on the MoCA score,211 cases had NCI,142 had MCI,and 25 had SCI.Compared with patients in group B,patients in group A was associated with significantly higher positive rate of SCI[12.2％ (23/189)vs.1.1％ (2/189),P=0.000],MCI[40.2％ (76/189)vs.34.9％ (66/189),P=0.013]and slightly higher positive rate of NCI[56.6％ (107/189)vs.55.0％ (104/189),P=0.214].Conclusions The test time of MoCA may have an effect on the cognitive function detection rate in elderly patients with stroke,and the time of MoCA examination should be considered in clinical examination.

Objective To explore the diagnostic value of quantitative radiomics features from dual-modal ultrasound composed of elastography and B-mode for axillary lymph node metastasis in breast cancer patients. Methods We retrospectively analyzed 161 axillary lymph nodes (69 benign and 92 metastatic) undergoing real-time elastography and B-mode ultrasound from 158 patients with breast cancer. We extracted a total of 428 features, consisting of morphologic features from B-mode, and intensity features and gray-level co-occurrence matrix features from the dual modalities,and the optimal subsut of features was selected through least absolute shrinkage and selection operator (Lasso) under the condition of leave-one-out cross validation. We used SVM for the classification of benign and metastatic nodes. Results The sensitivity, specificity, accuracy and Youden's index of the 35 radiomics features selected with Lasso were 86.96%, 85.51%, 86.34% and 72.46%, respectively. Conclusion The radiomics features from dual-modal ultrasound (elastography and B-mode) have demonstrated good performance for classification and have potential to be applied to clinical diagnosis of axillary lymph node metastasis.

Aim To investigate the immunomodulatory effects of sea cucumber fucoidan ( SC-FUC) on macro-phage and the signaling pathways. Methods Cell via-bilities in response to different concentrations of SC-FUC were analyzed by MTT, phagocytosis ability was detected by neutral red,and nitric oxide ( NO) produc-tion was examined by Griess reaction kit. The mRNA expression levels of IL-6 , IL-10 , Toll-like receptors (TLRs) and related signal molecules MyD88, TRIF, NF-κB were assayed by real-time PCR. All the experi-ments were based on murine RAW264. 7 cell line. Re-sults SC-FUC could promote RAW264 . 7 cell prolif-eration, phagocytosis as evidenced by uptake of neutral red and release of NO. The effects were significant at the early stage (6 h and 12 h) . SC-FUC could up-reg-ulate the expression of IL-6 , IL-10 , TLR4 , TLR5 , TLR9. Moreover, mRNA expressions of TLRs signaling molecules were increased, as well as MyD88, TRIF, NF-κB. Conclusions SC-FUC could activate macro-phage, and then promote the immune function by pro-moting production or expression of NO, IL-6, IL-10. It is speculated to be relevant to activated cell surface re-ceptors in macrophage, including TLR4, TLR5, TLR9, and NF-κB signaling pathways.

BACKGROUND:Many studies concern the comparison of micro-titanium plate fixation and suture suspension fixation during cervical posterior expansive open-door laminoplasty, but the sample size of many studies has limitations. There is lack of objective evaluation on advantages and disadvantages of micro-titanium plate.
OBJECTIVE:To systemicaly evaluate the efficacy and safety of micro-titanium plate fixationversus suture suspension fixation in cervical posterior expansive open-door laminoplasty.
METHODS: English and Chinese randomized controled trials were searched by two reviewers. They retrieved the Cochrane Central Register of Controled Trials (CENTRAL), PubMed, EMbase, the ISI Web of Knowledge Database, CNKI, CMB, VIP and Wanfang database for randomized controled trials addressing micro-titanium plate fixationversus suture suspension fixation in cervical posterior expansive open-door laminoplasty published from database foundation to March 1, 2015. The references were also searched by hand. Meta-analyses were performed by using the Rev-Man 5.3 software, provided by the Cochrane Colaboration.
RESULTS AND CONCLUSION: A total of 9 studies involving 642 patients were included. The results of meta-analyses showed that: (1) safety: There were no significant differences between the two groups in operation time [SMD=-0.02, 95%CI (-0.57, 0.54),P=0.95 > 0.05], and intraoperative blood loss [SMD=0.07, 95%CI (-0.26, 0.40),P=0.69 > 0.05]. (2) Efficacy: compared with suture suspension fixation, Japanese Orthopaedic Association Scores were higher [SMD=0.26, 95%CI (0.10, 0.42),P=0.001 < 0.05], the angle of the opened laminae was bigger [SMD=0.25, 95%CI (0.02, 0.48),P=0.04 < 0.05], cervical curvature was better [SMD=0.46, 95%CI (0.27, 0.65),P < 0.000 01], and incidence of axial symptoms was lower [RR=0.40, 95%CI(0.29, 0.56),P< 0.000 01] after micro-titanium plate fixation. These findings suggest that during expansive open-door laminoplasty for treatment of cervical spondylosis, micro-titanium plate fixation and suture suspension fixation can obtain good clinical outcomes. However, Japanese Orthopaedic Association Scores were higher and the angle of the opened laminae was better in micro-titanium plate fixation than in suture suspension fixation. Micro-titanium plate fixation could effectively prevent loss of cervical curvature and reduce the incidence of axial symptoms. For the poor quality of the original studies and smal sample size, a prudent choice is suggested. More high-quality large-sample studies are needed for further verification.

Objective To determine whether the aberrant white matter network is shared by patients with schizophrenia and their healthy siblings.Methods Fourteen-three first-episode,treatment-naive patients with schizophrenia (patients),40 healthy siblings (siblings),and 55 healthy controls (controls)were scanned with 3.0T MRI scanner and diffusion tensor imaging tractography was used to construct the whiter matter brain network.The differences of white matter network were compared with analysis of variance among the 3 groups.Results The white matter network connectivity strength and the global efficiency significantly reduced in both patients (5.14 ±0.36,0.25 ±0.02) and their siblings (5.25 ±0.27,0.25 ±0.01) comparing to controls (5.41 ± 0.24,0.26 ± 0.01 ; F =16.55 P ＜ 0.01),without significant difference between patients and siblings (P ＞ 0.05).The degree in left precuneus,left anterior cingulate and right orbitofrontal gyrus was significantly lower in patients (7.42 ± 1.04,7.58 ± 1.25 and 3.72 ± 1.46)and siblings (7.51 ± 1.18,7.87 ± 1.10 and 4.42 ± 1.09) than controls (8.22 ± 1.07,8.31 ±0.99 and 4.80 ±0.92,P ＜0.05,FDR corrected); patients was also lower than siblings in right orhitofrontal gyrus (P ＜ 0.05,FDR corrected).Additionally significantly reduced betweenness centrality in left precuneus,left anterior cingulate and right orbitofrontal gyrus in patients (0.31 ± 0.02,0.32 ± 0.03 and 0.25 ± 0.03) and siblings (0.31 ±0.02,0.33 ±0.02 and 0.27 ±0.03) compared to controls (0.32 ±0.02,0.34 ± 0.02and 0.28 ± 0.02,P ＜ 0.05,FDR corrected) ; patients was also lower than siblings in right orbitofrontal gyrus (P ＜ 0.05,FDR corrected).Conclusions These findings suggest that schizophrenia patients and their healthy siblings share the aberrant white matter network,which may be the susceptibility biological marker for schizophrenia.