1.Development and validation of a multi-modality fusion deep learning model for differentiating glioblastoma from solitary brain metastases
Shanshan SHEN ; Chunquan LI ; Yaohua FAN ; Shanfu LU ; Ziye YAN ; Hu LIU ; Haihang ZHOU ; Zijian ZHANG
Journal of Central South University(Medical Sciences) 2024;49(1):58-67
Objective:Glioblastoma(GBM)and brain metastases(BMs)are the two most common malignant brain tumors in adults.Magnetic resonance imaging(MRI)is a commonly used method for screening and evaluating the prognosis of brain tumors,but the specificity and sensitivity of conventional MRI sequences in differential diagnosis of GBM and BMs are limited.In recent years,deep neural network has shown great potential in the realization of diagnostic classification and the establishment of clinical decision support system.This study aims to apply the radiomics features extracted by deep learning techniques to explore the feasibility of accurate preoperative classification for newly diagnosed GBM and solitary brain metastases(SBMs),and to further explore the impact of multimodality data fusion on classification tasks. Methods:Standard protocol cranial MRI sequence data from 135 newly diagnosed GBM patients and 73 patients with SBMs confirmed by histopathologic or clinical diagnosis were retrospectively analyzed.First,structural T1-weight,T1C-weight,and T2-weight were selected as 3 inputs to the entire model,regions of interest(ROIs)were manually delineated on the registered three modal MR images,and multimodality radiomics features were obtained,dimensions were reduced using a random forest(RF)-based feature selection method,and the importance of each feature was further analyzed.Secondly,we used the method of contrast disentangled to find the shared features and complementary features between different modal features.Finally,the response of each sample to GBM and SBMs was predicted by fusing 2 features from different modalities. Results:The radiomics features using machine learning and the multi-modal fusion method had a good discriminatory ability for GBM and SBMs.Furthermore,compared with single-modal data,the multimodal fusion models using machine learning algorithms such as support vector machine(SVM),Logistic regression,RF,adaptive boosting(AdaBoost),and gradient boosting decision tree(GBDT)achieved significant improvements,with area under the curve(AUC)values of 0.974,0.978,0.943,0.938,and 0.947,respectively;our comparative disentangled multi-modal MR fusion method performs well,and the results of AUC,accuracy(ACC),sensitivity(SEN)and specificity(SPE)in the test set were 0.985,0.984,0.900,and 0.990,respectively.Compared with other multi-modal fusion methods,AUC,ACC,and SEN in this study all achieved the best performance.In the ablation experiment to verify the effects of each module component in this study,AUC,ACC,and SEN increased by 1.6%,10.9%and 15.0%,respectively after 3 loss functions were used simultaneously. Conclusion:A deep learning-based contrast disentangled multi-modal MR radiomics feature fusion technique helps to improve GBM and SBMs classification accuracy.
2.Comparison between endoscopic surgery and glucocorticoid therapy for traumatic optic neuropathy: a meta-analysis
Jinhai YU ; Yan WU ; Yaohua WANG ; Qihua XU ; Chao XIONG ; Hongfei LIAO
Chinese Journal of Experimental Ophthalmology 2023;41(1):42-46
Objective:To systematically analyze the efficacy of nasal endoscopic transsphenoidal decompression and glucocorticoid pulse therapy for traumatic optic neuropathy (TON).Methods:PubMed, Web of Science, Cochrane Library, ClinicalTrial.gov, CNKI, Wanfang, China Biomedical and other databases were searched to retrieve clinical studies on endoscopic surgery and hormone therapy for TON since their establishment to November 1, 2020.Two researchers screened the literature and evaluated the quality of the included literatures.The improvement of visual acuity before and after treatment was taken as an effective indicator.RevMan 5.3 statistical software was used for meta-analysis.The sensitivity of the results was analyzed atfer literature exclusion.The publication bias of each study was double checked by funnel plot and Begg test.Results:Eight studies were included, including 7 cohort studies and 1 randomized controlled trial.There was no significant difference between the endoscopic decompression group and glucocorticoid group in the efficiency of visual acuity improvement in the treatment of TON [odds ratio ( OR)=1.65, 95% confidence interval ( CI)∶0.75-3.66, P=0.22], neither in TON patients with residual vision before surgery ( OR=2.17, 95% CI: 0.94-4.98, P=0.07). For nasal endoscopic decompression surgery, early surgery (disease course<7 days) was more effective than late surgery (disease course>7 days) ( OR=4.73, 95% CI: 2.55-8.78, P<0.01). Sensitivity analysis suggested that the results of this literature analysis were not robust.The Begg test showed that there was no literature publication bias. Conclusions:There is no significant difference between nasal endoscopic surgery and glucocorticoid therapy in the treatment of TON.Early endoscopic surgery may help improve visual acuity in patients with residual vision.
3.The special consideration for using the real-world data as external control in clinical evaluation
Yuning WANG ; Minghong YAO ; Yan REN ; Jiayue XU ; Yulong JIA ; Kai LIN ; Yaohua LI ; Chunmin WEI ; Xin SUN
Chinese Journal of Epidemiology 2023;44(2):335-340
Randomized controlled trials (RCT) have long been considered the gold standard for assessing clinical efficacy. However, RCT are inappropriate for some diseases due to related ethical issues and costs, such as rare diseases that are seriously life-threatening but without adequate treatment. Using real world data (RWD) as external control for RCT could make recruitment less complicated and reduce time and cost. This paper introduces common application scenarios, data sources, study design, basic principles, and statistical methods of RWD as an external control based on the latest guidelines related to RWD and combined with our team's previous research experience. This study could provide references for scholars and sponsors who want to conduct RWD research.
4.Effects of smoking and drinking status before operation on recurrence and metastasis of esophageal squamous cell carcinoma
Jianfei ZHU ; Yawei DOU ; Wei TIAN ; Yun DAI ; Xianghui LUO ; Yaohua CHEN ; Hongtao WANG ; Zhe LI
Chinese Journal of Clinical Thoracic and Cardiovascular Surgery 2021;28(02):219-224
Objective To evaluate the effect of smoking and drinking status on the prognosis of patients with esophageal squamous cell carcinoma (ESCC). Methods The clinical data of 483 patients with ESCC who underwent surgical treatment in Shannxi Provincial People's Hospital from 2007 to 2016 were retrospectively analyzed. Among them, 352 patients were male and 131 were female, with a median age of 64 (37-80) years. There were 311 smokers and 172 drinkers. The relationship between preoperative drinking or smoking status and the clinicopathological characteristics of patients with ESCC was analyzed. Log-rank method and Cox risk regression were used to conduct univariate and multivariate survival analysis, respectively. Results The preoperative smoking status was related to the patient's tumor location (P=0.030). Drinking status was associated with tumor location (P=0.001), degree of differentiation (P=0.030), pathological T stage (P=0.024) and pathological N stage (P=0.029). Univariate survival analysis showed that smoking status did not affect the disease-free survival (DFS) (P=0.188) and overall survival (OS) (P=0.127) of patients with ESCC. However, patients who drank alcohol had worse PFS than non-drinking patients (29.37 months vs. 42.87 months, P=0.009). It was further proved that alcohol consumption was an independent risk factor affecting patients' recurrence and metastasis by using multivariate analysis (RR=1.28, P=0.040). Alcohol consumption also reduced the OS of patients by 21.47 months (P=0.014), however, multivariate analysis did not yield significant results. Conclusion Preoperative drinking status is related to the stage and differentiation of patients with ESCC. It is an independent risk factor affecting the recurrence and metastasis of ESCC.
5.NOGEA: A Network-oriented Gene Entropy Approach for Dissecting Disease Comorbidity and Drug Repositioning
Guo ZIHU ; Fu YINGXUE ; Huang CHAO ; Zheng CHUNLI ; Wu ZIYIN ; Chen XUETONG ; Gao SHUO ; Ma YAOHUA ; Shahen MOHAMED ; Li YAN ; Tu PENGFEI ; Zhu JINGBO ; Wang ZHENZHONG ; Xiao WEI ; Wang YONGHUA
Genomics, Proteomics & Bioinformatics 2021;19(4):549-564
Rapid development of high-throughput technologies has permitted the identification of an increasing number of disease-associated genes (DAGs), which are important for understanding disease initiation and developing precision therapeutics. However, DAGs often contain large amounts of redundant or false positive information, leading to difficulties in quantifying and prioritizing potential relationships between these DAGs and human diseases. In this study, a network-oriented gene entropy approach (NOGEA) is proposed for accurately inferring master genes that contribute to specific diseases by quantitatively calculating their perturbation abilities on directed disease-specific gene networks. In addition, we confirmed that the master genes identified by NOGEA have a high reliability for predicting disease-specific initiation events and progression risk. Master genes may also be used to extract the underlying information of different diseases, thus revealing mechanisms of disease comorbidity. More importantly, approved therapeutic targets are topologically localized in a small neighborhood of master genes in the interactome network, which provides a new way for predicting drug-disease associations. Through this method, 11 old drugs were newly identified and predicted to be effective for treating pancreatic cancer and then validated by in vitro experiments. Collectively, the NOGEA was useful for identifying master genes that control disease initiation and co-occurrence, thus providing a valuable strategy for drug efficacy screening and re-positioning. NOGEA codes are publicly available at https://github.com/guozihuaa/NOGEA.
6.Analysis and management of coronary heart disease complicated by thrombocytopenia
Xingxia LI ; Yan HUO ; Xipeng SUN ; Yan CHEN ; Yaohua LU ; Cheng GUO
Journal of Pharmaceutical Practice 2020;38(1):77-80
Objective To summarize the factors and the corresponding treatments for patients with coronary heart disease (CHD) complicated by thrombocytopenia, and provide medical advises for clinical treatment. Methods Literatures and case reports were analyzed and summarized. Results Thrombocytopenia in patient with CHD could be mainly divided into two types, one group was induced by the antithrombotic therapies for CHD, and the other group was caused by some concurrent diseases or combined non-antithrombotic medications. There were different medical decisions and prognoses according to the causes in different groups. Conclusion The treatment strategies of CHD will be considered in the CHD patients with thrombocytopenia. Identifying thrombocytopenia by monitoring the platelet counts in early stage, finding out the causes quickly and providing proper treatments are the key for the prognosis of the patients.
7.Changes of pupil diameter and pupillary centroid shift of myopia from scotopic to photopic condition
Qing ZHU ; Yan WANG ; Wenxing NING ; Yaohua ZHANG ; Wenbo CHENG ; Tong CUI ; Weiting HAO
Chinese Journal of Experimental Ophthalmology 2020;38(6):510-514
Objective:To study the pupillary centroid shift of myopia and characteristics of pupil diameter change from scotopic to photopic condition.Methods:A case series study was carried out, 140 eyes of 70 myopia patients from September to November 2016 in Tianjin Eye Hospital were enrolled.The pupillary centroid shift and pupil diameter parameters were measured by visual quality analyzer from scotopic (0.017 lx) to photopic (10.400 lx) condition.This study followed the Declaration of Helsinki.Results:Under the scotopic and photopic conditions, the pupil diameter was positively correlated between the bilateral eyes (scotopic: rs=0.85, P<0.001; photopic: r=0.85, P<0.001), and the pupil diameter variation from scotopic to photopic condition was positively correlated between the bilateral eyes ( r=0.75, P<0.001). The pupil diameter in scotopic and photopic conditions, and the change of pupil diameter in the right eyes were significantly higher than those in the left eyes (all at P<0.05). The pupillary centroid shift was within 0.2 mm in the left eyes of 94.2% (66/70) subjects and in the right eyes of 97.1%(68/70) subjects.The pupillary centroid shift of all subjects was within 0.3 mm.From scotopic to photopic condition, the pupil centroid was mainly shift to the nasal superior direction.There was no significant correlation between pupil diameter and age or gender.There was no significant correlation between pupillary centroid shift and age, diopter or pupil diameter. Conclusions:The binocular pupillary centroid shift are symmetrical from scotopic to photopic condition in myopic eyes, and the pupil centroid mainly shift to nasal superior direction.
8.Advances in biomechanical effects of femtosecond laser corneal refractive surgery
Chinese Journal of Experimental Ophthalmology 2020;38(6):534-538
Femtosecond laser has been widely used in corneal refractive surgery.The refractive state, corneal tissue structure and biomechanical properties are changed through cutting the corneal tissue.Femtosecond laser-assisted in situ keratomileusis(FS-LASIK)and small incision lenticule extraction(SMILE)are the most commonly used femtosecond laser corneal refractive surgery.In this study, the structural basis of corneal biomechanical properties, the commonly used methods for corneal biomechanics measuring, biomechanical effect of corneal flap thickness, side-cut angles, residual stromal bed thickness, flap hinge position for F-LASIK and the size of corneal incision, optical zone diameter, cap thickness, tissue cutting thickness for SMILE on corneal biomechanics are reviewed.
9. Analysis of anterior chamber morphological changes after small incision lenticule extraction
Chinese Journal of Experimental Ophthalmology 2019;37(11):901-906
Objective:
To investigate the changes of anterior chamber profiles after small incision lenticule extraction(SMILE) and analyze the possible influencing factors.
Methods:
A case observation study was designed.Thirty-one patients(59 eyes) with myopia and myopic astigmatism who underwent SMILE surgery at the Refractive Surgery Center of Tianjin Eye Hospital from April to December 2015 were enrolled and examined preoperatively and at 1 month, 3 and 6 months postoperatively.The anterior chamber profiles, including anterior chamber depth (ACD), anterior chamber volume (ACV) and anterior chamber angle (ACA) were measured using the Pentacam.The correlations between parameters were analyzed and the differences of the parameters at different time points were compared.This study followed the Declaration of Helsinki.This study protocol was approved by Ethic Committee of Tianjin Eye Hospital (No.TJYYLL-2015-20). Written informed consent was obtained from each subject before the operation.
Results:
Compared with pre-operation, the central corneal thickness (CCT) was significantly decreased at 1 month, 3 and 6 months postoperatively (all at
10.Correlation analysis between corneal biomechanical properties and high-order aberration in myopia
Yaohua ZHANG ; Yan WANG ; Rui DOU ; Pinghui WEI
Chinese Journal of Experimental Ophthalmology 2018;36(5):368-372
Objective To investigate the correlations between corneal biomechanical properties and corneal high-order aberrations (HOAs) in myopic eyes.Methods In this retrospective study,263 right eyes with myopia or myopic astigmatism were enrolled from January to October 2015 in Tianjin Eye Hospital.The values of first applanation time (A1T),first applanation length (A1L),first applanation velocity (A1V),second applanation time (A2T),second applanation length (A2L),second applanation velocity (A2V),highest concavity time (HCT),highest concavity deformation amplitude (HCDA),highest concavity peak distance (HCPD),highest concavity radius (HCR) and non-contact intraocular pressure (IOPnct) were measured using the corneal visualization Scheimpflug technology (Corvis ST).The central corneal thickness (CCT) and wavefront aberrations of the anterior,posterior surface,and the total cornea were measured with Pentacam.Pearson correlation and stepwise multivariate linear regression analysis were applied to analyze the correlations between biomechanical properties and corneal HOAs.This study follows the Helsinki declaration.The experiment was approved by the Ethic Committee of Tianjin Eye Hospital.Informed consent was signed from each patient.Results There was a correlation between biomechanical properties and corneal HOAs.Negative correlations were found between A1L and the spherical aberration,HOAs RMS of anterior surface,total RMS and HOAs RMS of the posterior surface and HOAs RMS of the total cornea (P<0.05).HCRs were negatively correlated with total RMS of the total cornea and the anterior surface (P<0.05).The A2V was correlated with various aberrations.Multiple stepwise linear regression model analysis showed that,the correlation between the aberrations of anterior surface and biomechanical parameters,and the correlations between the total corneal aberration and biomechanical parameters were weak,the correlations between the aberrations of posterior surface and biomechanical parameters were strong.Conclusions There is a correlation between the biomechanical properties and corneal HOAs.The changes of the corneal biomechanical properties are mainly associated with the posterior corneal aberrations,which may contribute to the early diagnosis of subclinical keratoconus or dilated corneal disease.

Result Analysis
Print
Save
E-mail