scholarly journals A Hybrid Method to Predict Postoperative Survival of Lung Cancer Using Improved SMOTE and Adaptive SVM

2021 ◽  
Vol 2021 ◽  
pp. 1-15
Author(s):  
Jiang Shen ◽  
Jiachao Wu ◽  
Man Xu ◽  
Dan Gan ◽  
Bang An ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Area Under The Curve ◽  
Imbalanced Data ◽  
Medical Decision ◽  
Postoperative Survival ◽  
Support Vector ◽  
Lung Cancer Patients ◽  
Second Stage ◽  
Self Tuning ◽  
Two Stages

Predicting postoperative survival of lung cancer patients (LCPs) is an important problem of medical decision-making. However, the imbalanced distribution of patient survival in the dataset increases the difficulty of prediction. Although the synthetic minority oversampling technique (SMOTE) can be used to deal with imbalanced data, it cannot identify data noise. On the other hand, many studies use a support vector machine (SVM) combined with resampling technology to deal with imbalanced data. However, most studies require manual setting of SVM parameters, which makes it difficult to obtain the best performance. In this paper, a hybrid improved SMOTE and adaptive SVM method is proposed for imbalance data to predict the postoperative survival of LCPs. The proposed method is divided into two stages: in the first stage, the cross-validated committees filter (CVCF) is used to remove noise samples to improve the performance of SMOTE. In the second stage, we propose an adaptive SVM, which uses fuzzy self-tuning particle swarm optimization (FPSO) to optimize the parameters of SVM. Compared with other advanced algorithms, our proposed method obtains the best performance with 95.11% accuracy, 95.10% G -mean, 95.02% F1, and 95.10% area under the curve (AUC) for predicting postoperative survival of LCPs.

scholarly journals An Expert System Based on Fisher Score and LS-SVM for Cardiac Arrhythmia Diagnosis

2013 ◽  
Vol 2013 ◽  
pp. 1-6 ◽  
Author(s):  
Ersen Yılmaz
Keyword(s):  
Expert System ◽  
Cardiac Arrhythmia ◽  
Feature Space ◽  
Support Vector ◽  
Feature Subset ◽  
Fisher Score ◽  
Data Set ◽  
Second Stage ◽  
Vector Machines ◽  
Two Stages

An expert system having two stages is proposed for cardiac arrhythmia diagnosis. In the first stage, Fisher score is used for feature selection to reduce the feature space dimension of a data set. The second stage is classification stage in which least squares support vector machines classifier is performed by using the feature subset selected in the first stage to diagnose cardiac arrhythmia. Performance of the proposed expert system is evaluated by using an arrhythmia data set which is taken from UCI machine learning repository.

Diagnostic and prognostic values of circular RNAs for lung cancer: a meta-analysis

2020 ◽  
pp. postgradmedj-2019-137178
Author(s):  
Qian Yang ◽  
Lizhen Chen ◽  
Li Yang ◽  
Yuanshuai Huang
Keyword(s):  
Lung Cancer ◽  
Cancer Patients ◽  
Meta Analysis ◽  
Area Under The Curve ◽  
High Sensitivity ◽  
Cochrane Library ◽  
Clinicopathological Parameters ◽  
Circular Rnas ◽  
Lung Cancer Patients ◽  
Diagnostic Values

Circular RNAs (circRNAs) may serve as potential biomarkers for patients with lung cancer. The aim of this meta-analysis was to analyse the diagnostic, prognostic and clinicopathological values of circRNAs in lung cancer patients. A systematic search of PubMed, Embase, Web of Science, Scopus and the Cochrane Library databases was performed for relevant articles from inception to 29 January 2020. Pooled parameters including sensitivity, specificity and area under the curve (AUC) were used to assess the diagnostic performance, HRs and 95% CIs were used to evaluate overall survival (OS) and ORs were used to estimate clinicopathological parameters. 52 studies from 45 articles were enrolled in this study, including 17 on diagnosis and 35 on prognosis. For diagnostic values, circRNAs could discriminate lung cancer patients from the controls, with AUC of 0.83 (95% CI: 0.79 to 0.86), a relatively high sensitivity of 0.77 (95% CI: 0.73 to 0.81) and specificity of 0.75 (95% CI: 0.71 to 0.79). For prognostic significances, overexpression of 23 upregulated circRNAs was relevant to a poor prognosis (OS: HR=2.21, 95% CI: 1.96 to 2.49, p<0.001), and overexpression of 9 downregulated circRNAs was correlated with a favourable prognosis (OS: HR=0.62, 95% CI: 0.53 to 0.73, p<0.001). As for clinicopathological parameters, high expression of 23 upregulated circRNAs was associated with unfavourable clinicopathological features while 9 downregulated circRNAs proved the contrary. In conclusion, this study confirmed that circRNAs might serve as important biomarkers for diagnostic and prognostic values of lung cancer.

EFFECTS OF SVM PARAMETER OPTIMIZATION BASED ON THE PARAMETER DESIGN OF TAGUCHI METHOD

2011 ◽  
Vol 20 (03) ◽  
pp. 563-575 ◽  
Author(s):  
MEI LING HUANG ◽  
YUNG HSIANG HUNG ◽  
EN JU LIN
Keyword(s):  
Taguchi Method ◽  
Parameter Design ◽  
Support Vector ◽  
Svm Classifier ◽  
Liver Disorders ◽  
Second Stage ◽  
Vector Machines ◽  
Class Labels ◽  
Two Stages ◽  
Decision Boundaries

Support Vector Machines (SVMs) are based on the concept of decision planes that define decision boundaries, and Least Squares Support Vector (LS-SVM) Machine is the reformulation of the principles of SVM. In this study a diagnosis on a BUPA liver disorders dataset, is conducted LS-SVM with the Taguchi method. The BUPA Liver Disorders dataset includes 345 samples with 6 features and 2 class labels. The system approach has two stages. In the first stage, in order to effectively determine the parameters of the kernel function, the Taguchi method is used to obtain better parameter settings. In the second stage, diagnosis of the BUPA liver disorders dataset is conducted using the LS-SVM classifier; the classification accuracy is 95.07%; the AROC is 99.12%. Compared with the results of related research, our proposed system is both effective and reliable.

scholarly journals Improving activated sludge classification based on imbalanced data

2014 ◽  
Vol 16 (6) ◽  
pp. 1331-1342 ◽  
Author(s):  
Y. Qian ◽  
Y. C. Liang ◽  
R. C. Guan
Keyword(s):  
Neural Network ◽  
Wastewater Treatment ◽  
Activated Sludge ◽  
Recognition Rate ◽  
Back Propagation ◽  
Area Under The Curve ◽  
Imbalanced Data ◽  
Real Data ◽  
Support Vector ◽  
Classification Problems

A fast and accurate classification method for sewage sludge biological activity classification is of great significance for wastewater treatment. However, the data are often imbalanced and the accuracy of traditional classification algorithms applied to imbalanced small classes of data is very low. Such small classes are crucial application data. Therefore, based on the analysis of eight microorganisms, a novel method is proposed in this paper for the classification of activated sludge known as balanced support-vector-based back-propagation (SV-BP) neural network. It first splits the multiclass classification problem into a plurality of pairwise classification problems and uses a support vector machine (SVM) to achieve equalization. Second, the new dataset is produced, following which back-propagation neural network (BPNN) is used for training and classification. To examine the efficiency of the model, 1731 real data points are collected from a wastewater treatment factory and divide the data into four classes with the help of wastewater experts. Based on the new model, data redundancy and noise are greatly reduced. With area under the curve (AUC) measurements, we find that the AUC of SV-BP is 6.9% higher than classical BPNN. In addition, the small-class recognition rate of SV-BP is far better than that by classical BPNN and SVM algorithms.

Discovery of a novel linc01125 isoform in serum exosomes as a promising biomarker for NSCLC diagnosis and survival assessment

2021 ◽  
Author(s):  
Jianfeng Xian ◽  
Yuyuan Zeng ◽  
Shizhen Chen ◽  
Liming Lu ◽  
Li Liu ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Area Under The Curve ◽  
Lung Cancer Patients ◽  
Non Invasive ◽  
Invasive Method ◽  
Highly Correlated ◽  
Nsclc Patients ◽  
Serum Exosomes

Abstract A non-invasive method to distinguish potential lung cancer patients would improve lung cancer prevention. We employed the RNA-Seq analysis to profile serum exosomal long non-coding RNAs (lncRNAs) from non-small cell lung cancer (NSCLC) patients and pneumonia controls, and then determined the diagnostic and prognostic value of a promising lncRNA in four datasets. We identified 90 dysregulated lncRNAs for NSCLC and found the most significant lncRNA was a novel isoform of linc01125. Serum exosomal linc01125 could distinguish NSCLC cases from disease-free and tuberculosis controls, with the area under the curve (AUC) values as 0.662 (95% confidence interval [CI]= 0.614-0.711) and 0.624 (95%CI= 0.522–0.725), respectively. High expression of exosomal linc01125 was also correlated with an unfavorable overall survival of NSCLC (hazard ratio [HR] = 1.58, 95%CI = 1.01–2.49). Clinic treatment decreased serum exosomal linc01125 in NSCLC patients (P = 0.036). Linc01125 functions to inhibit cancer growth and metastasis via acting as a competing endogenous RNA to up-regulate TNFAIP3 expression by sponging miR-19b-3p. Notably, the oncogenic transformation of 16HBE leads to decreased linc01125 in cells but increased linc01125 in cell-derived exosomes. The expression of linc01125 in total exosomes was highly correlated with that in tumor-associated exosomes in serum. Moreover, lung cancer cells were capable of releasing linc01125 into exosomes in vitro and in vivo. Our analyses suggest serum exosomal linc01125 as a promising biomarker for non-invasively diagnosing NSCLC and predicting the prognosis of NSCLC.

scholarly journals The investigation of the volatile metabolites of lung cancer from the microenvironment of malignant pleural effusion

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ke-Cheng Chen ◽  
Shih-Wei Tsai ◽  
Xiang Zhang ◽  
Chian Zeng ◽  
Hsiao-Yu Yang
Keyword(s):  
Lung Cancer ◽  
Pleural Effusion ◽  
Pleural Fluid ◽  
Malignant Pleural Effusion ◽  
Characteristic Curve ◽  
Final Analysis ◽  
Gas Chromatography Mass Spectrometry ◽  
Support Vector ◽  
Lung Cancer Patients ◽  
Volatile Metabolites

AbstractFor malignant pleural effusions, pleural fluid cytology is a diagnostic method, but sensitivity is low. The pleural fluid contains metabolites directly released from cancer cells. The objective of this study was to diagnose lung cancer with malignant pleural effusion using the volatilomic profiling method. We recruited lung cancer patients with malignant pleural effusion and patients with nonmalignant diseases with pleural effusion as controls. We analyzed the headspace air of the pleural effusion by gas chromatography-mass spectrometry. We used partial least squares discriminant analysis (PLS-DA) to identify metabolites and the support vector machine (SVM) to establish the prediction model. We split data into a training set (80%) and a testing set (20%) to validate the accuracy. A total of 68 subjects were included in the final analysis. The PLS-DA showed high discrimination with an R2 of 0.95 and Q2 of 0.58. The accuracy of the SVM in the test set was 0.93 (95% CI 0.66, 0.998), the sensitivity was 83%, the specificity was 100%, and kappa was 0.85, and the area under the receiver operating characteristic curve was 0.96 (95% CI 0.86, 1.00). Volatile metabolites of pleural effusion might be used in patients with cytology-negative pleural effusion to rule out malignancy.

scholarly journals The Investigation of the Volatile Metabolites of Lung Cancer From the Microenvironment of Malignant Pleural Effusion

2021 ◽  
Author(s):  
Ke-Cheng Chen ◽  
Shih-Wei Tsai ◽  
Xiang Zhang ◽  
Chian Zeng ◽  
Hsiao-Yu Yang
Keyword(s):  
Lung Cancer ◽  
Pleural Effusion ◽  
Pleural Fluid ◽  
Malignant Pleural Effusion ◽  
Characteristic Curve ◽  
Final Analysis ◽  
Gas Chromatography Mass Spectrometry ◽  
Support Vector ◽  
Lung Cancer Patients ◽  
Volatile Metabolites

Abstract Lung cancer is the leading cause of cancer death. For malignant pleural effusions, pleural fluid cytology is a diagnostic method, but sensitivity is low. Many patients need to undergo invasive diagnostic tests such as thoracoscopic pleural biopsy. Pleural space is an enclosed microenvironment, and the pleural fluid contains metabolites directly released from cancer cells. The objective of this study was to diagnose lung cancer with malignant pleural effusion using the volatilomic profiling method. We recruited lung cancer patients with malignant pleural effusion and patients with nonmalignant diseases with pleural effusion as controls. We analyzed the headspace air of the pleural effusion by gas chromatography-mass spectrometry. We used partial least squares discriminant analysis (PLS-DA) to identify metabolites and the support vector machine (SVM) to establish the prediction model. We split data into a training set (80%) and a testing set (20%) to validate the accuracy. A total of 68 subjects were included in the final analysis. The PLS-DA showed high discrimination with an R2 of 0.95 and Q2 of 0.58. The accuracy of the SVM in the test set was 0.93 (95% CI: 0.66, 0.998), and kappa was 0.85, and the area under the receiver operating characteristic curve was 0.96 (95% CI: 0.86, 1.00). Pathway analysis revealed disturbances in pyruvate metabolism, the tricarboxylic acid, glycolysis, and lysine degradation. The volatile metabolites identified from malignant pleural effusion of lung cancer were primarily methylated alkanes. The pleural effusion contained volatile metabolites that have high accuracy in diagnosing lung cancer with malignant pleural effusion.

scholarly journals Liquid Biopsy in Lung Cancer Screening: The Contribution of Metabolomics. Results of A Pilot Study

Cancers ◽  
2019 ◽  
Vol 11 (8) ◽  
pp. 1069 ◽  
Author(s):  
Sandeep Singhal ◽  
Christian Rolfo ◽  
Andrew W. Maksymiuk ◽  
Paramjit S. Tappia ◽  
Daniel S. Sitar ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Cancer Patients ◽  
Cancer Progression ◽  
Area Under The Curve ◽  
Lung Cancer Screening ◽  
Lung Cancer Mortality ◽  
Treatment Monitoring ◽  
Monitoring Tool ◽  
Lung Cancer Patients ◽  
Metabolite Concentrations

Background: Lung cancer is the most common cause of cancer-related deaths worldwide. Early diagnosis is crucial to increase the curability chance of the patients. Low dose CT screening can reduce lung cancer mortality, but it is associated with several limitations. Metabolomics is a promising technique for cancer diagnosis due to its ability to provide chemical phenotyping data. The intent of our study was to explore metabolomic effects and profiles of lung cancer patients to determine if metabolic perturbations in the SSAT-1/polyamine pathway can distinguish between healthy participants and lung cancer patients as a diagnostic and treatment monitoring tool. Patients and Methods: Plasma samples were collected as part of the SSAT1 Amantadine Cancer Study. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used to identify and quantify metabolite concentrations in lung cancer patient and control samples. Standard statistical analyses were performed to determine whether metabolite concentrations could differentiate between healthy subjects and lung cancer patients, as well as risk prediction modeling applied to determine whether metabolic profiles could provide an indication of cancer progression in later stage patients. Results: A panel consisting of 14 metabolites, which included 6 metabolites in the polyamine pathway, was identified that correctly discriminated lung cancer patients from controls with an area under the curve of 0.97 (95% CI: 0.875-1.0). Conclusion: When used in conjunction with the SSAT-1/polyamine pathway, these metabolites may provide the specificity required for diagnosing lung cancer from other cancer types and could be used as a diagnostic and treatment monitoring tool.

scholarly journals Utility of Thromboelastography to Identify Hypercoagulability in Lung Cancer Related Ischemic Stroke Patients

2020 ◽  
Vol 26 ◽  
pp. 107602962097550
Author(s):  
Xuemei Quan ◽  
Qixiong Qin ◽  
Xianting Que ◽  
Ya Chen ◽  
Yunfei Wei ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Ischemic Stroke ◽  
Cancer Patients ◽  
Area Under The Curve ◽  
Maximum Amplitude ◽  
Healthy Controls ◽  
Lung Cancer Patients ◽  
Coagulation Tests ◽  
D Dimer ◽  
Routine Coagulation

Lung cancer related hypercoagulability could increase the risk of ischemic stroke. Routine coagulation tests may have limited capacity in evaluating hypercoagulability. The aim of this study was to investigate the ability of thromboelastography (TEG) in the identification of hypercoagulability in patients with lung cancer and cryptogenic ischemic stroke (LCIS). Between January 2016 and December 2018, whole citrated blood from LCIS patients (n = 35) and age- and gender-matched lung cancer patients and healthy volunteers were used for TEG and routine coagulation tests. The coagulation indicator and clinical data were compared among the 3 groups. There were 27/35 (77.14%) on TEG and 18/35 (51.43%) on routine coagulation tests of LCIS patients who had evidence of hypercoagulability. The detection rate of hypercoagulability by TEG in LCIS patients was higher than routine coagulation tests ( P = 0.018). Comparing with lung cancer patients and healthy controls, LCIS patients have a significantly higher maximum amplitude (MA), fibrinogen, and D-dimer. Multivariate analysis showed that D-dimer and MA were significantly associated with ischemic stroke in lung cancer patients. ROC curve showed that the area under the curve of TEG (0.790 ± 0.048, 95% CI: 0.697-0.864) was significantly higher than routine coagulation tests (0.673 ± 0.059, 95% CI: 0.572-0.763) ( P = 0.04) in identifying hypercoagulability in LCIS patients. Therefore, TEG could identify hypercoagulability in LCIS patients and healthy controls. Identification of hypercoagulability in lung cancer patients by TEG may be helpful to prevent the occurrence of LCIS.

Sign in / Sign up

Export Citation Format

Share Document