scholarly journals Predicting Lung Cancer Survivability using SVM and Logistic Regression Algorithms

2017 ◽  
Vol 174 (2) ◽  
pp. 19-24 ◽  
Author(s):  
Animesh Hazra ◽  
Nanigopal Bera ◽  
Avijit Mandal
Keyword(s):  
Lung Cancer ◽  
Logistic Regression ◽  
Regression Algorithms

scholarly journals Risk factors for radiation-induced lung injury in patients with advanced non-small cell lung cancer: implication for treatment strategies

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Sha Sha ◽  
Jigang Dong ◽  
Maoyu Wang ◽  
Ziyu Chen ◽  
Peng Gao
Keyword(s):  
Risk Factors ◽  
Lung Cancer ◽  
Logistic Regression ◽  
Concurrent Chemotherapy ◽  
Small Cell ◽  
Regression Analyses ◽  
Small Cell Lung ◽  
Radiation Induced ◽  
Radiotherapy Dose ◽  
Nsclc Patients

Abstract Background The radiation-induced lung injury (RILI) in patients with advanced non-small cell lung cancer (NSCLS) is very common in clinical settings; we aimed to evaluate the risk factors of RILI in NSCLS patients, to provide insights into the treatment of NSCLS. Methods NSCLC patients undergoing three-dimensional conformal radiotherapy (3D-CRT) in our hospital from June 1, 2018, to June 30, 2020, were included. The characteristics and treatments of RILI and non-RILI patients were analyzed. Logistic regression analyses were conducted to assess the risk factors of RILI in patients with NSCLC. Results A total of 126 NSCLC patients were included; the incidence of RILI in NSCLC patients was 35.71%. There were significant differences in diabetes, smoke, chronic obstructive pulmonary disease (COPD), concurrent chemotherapy, radiotherapy dose, and planning target volume (PTV) between the RILI group and the non-RILI group (all P < 0.05). Logistic regression analyses indicated that diabetes (OR 3.076, 95%CI 1.442~5.304), smoke (OR 2.745, 95%CI 1.288~4.613), COPD (OR 3.949, 95%CI 1.067~5.733), concurrent chemotherapy (OR 2.072, 95%CI 1.121~3.498), radiotherapy dose ≥ 60 Gy (OR 3.841, 95%CI 1.932~5.362), and PTV ≥ 396 (OR 1.247, 95%CI 1.107~1.746) were the independent risk factors of RILI in patients with NSCLC (all P < 0.05). Conclusions RILI is commonly seen in NSCLS patients; early targeted measures are warranted for patients with those risk factors; future studies with larger sample sizes and different areas are needed to further elucidate the influencing factors of RILI in the treatment of NSCLS.

scholarly journals Downregulation of MicroRNA-330 Correlates with the Radiation Sensitivity and Prognosis of Patients with Brain Metastasis from Lung Cancer

2017 ◽  
Vol 42 (6) ◽  
pp. 2220-2229 ◽  
Author(s):  
Li-Peng Jiang ◽  
Zhi-Tu Zhu ◽  
Yue Zhang ◽  
Chun-Yan He
Keyword(s):  
Lung Cancer ◽  
Logistic Regression ◽  
Radiation Therapy ◽  
Survival Rate ◽  
Brain Metastasis ◽  
Cox Regression ◽  
Radiation Sensitivity ◽  
Extracranial Metastasis ◽  
Radiation Resistant ◽  
N Stage

Background: The present study sought to explore the role of microRNA-330 (miR-330) in predicting the radiation response and prognosis of patients with brain metastasis (BM) from lung cancer (LC). Methods: Patients with BM from LC were identified and classified into radiation-sensitive and radiation-resistant groups according to the overall survival rate, local and distant recurrence rate after conventional whole-brain radiation therapy. Quantitative realtime polymerase chain reaction (qRT-PCR) was used to detect miR-330 expression in serum. Receiver operating characteristic (ROC) curves were used to evaluate the prognostic value of miR-330 for the radiation sensitivity of brain metastasis from LC. Related clinical factors for radiation sensitivity were assessed by logistic regression analysis, and a survival analysis was conducted using COX regression and the Kaplan-Meier method. Results: MiR-330 exhibited lower expression in the radiation-sensitive group than in the radiation-resistant group. The area under the ROC curve of miR-330 for predicting radiation sensitivity was 0.898 (optimal cut-off value, 0.815), with a sensitivity of 71.7% and a specificity of 90.1%. After radiation therapy, patients with low miR-330 expression, compared to patients with high miR-330 expression, displayed a lower survival rate and a median survival time. MiR-330 expression was correlated with extracranial metastasis, maximum BM diameter, tumor-node-metastasis (TNM) stage and node (N) stage. Logistic regression and COX regression analyses revealed that extracranial metastasis, TNM stage, N stage and miR-330 expression were factors that influenced both radiation sensitivity and individual prognostic factors in patients with BM from LC. Conclusions: These findings indicate that the downregulation of miR-330 correlates with radiation sensitivity and poor prognosis in patients with BM from LC.

scholarly journals Clinical Significance of CYFRA21-1, AFP, CA-153, CEA, and CA-199 in the Diagnosis of Lung Cancer Ocular Metastasis in Hypertension Population

2021 ◽  
Vol 8 ◽  
Author(s):  
Jing Tang ◽  
Qian-Min Ge ◽  
Rong Huang ◽  
Hui-Ye Shu ◽  
Ting Su ◽  
...  
Keyword(s):  
Risk Factors ◽  
Lung Cancer ◽  
Logistic Regression ◽  
Sensitivity And Specificity ◽  
Cancer Metastasis ◽  
Binary Logistic Regression ◽  
Diagnostic Value ◽  
Ca 125 ◽  
Pathological Type ◽  
Independent Risk Factors

Purpose: To detect lung metastases, we conducted a retrospective study to improve patient prognosis.Methods: Hypertension patients with ocular metastases (OM group; n = 58) and without metastases (NM group; n = 1,217) were selected from individuals with lung cancer admitted to our hospital from April 2005 to October 2019. The clinical characteristics were compared by Student's t-test and chi-square test. Independent risk factors were identified by binary logistic regression, and their diagnostic value evaluated by receiver operating characteristic curve analysis.Results: Age and sex did not differ significantly between OM and NM groups; There were significant differences in pathological type and treatment. Adenocarcinoma was the main pathological type in the OM group (67.24%), while squamous cell carcinoma was the largest proportion (46.43%) in the NM group, followed by adenocarcinoma (34.10%). The OM group were treated with chemotherapy (55.17%), while the NM group received both chemotherapy (39.93%) and surgical treatment (37.06%). Significant differences were detected in the concentrations of cancer antigen (CA)−125, CA-199, CA-153, alpha fetoprotein (AFP), carcinoembryonic antigen (CEA), cytokeratin fraction 21-1 (CYFRA21-1), total prostate-specific antigen, alkaline phosphatase, and hemoglobin (Student's t-test). Binary logistic regression analysis indicated that CA-199, CA-153, AFP, CEA, and CYRFA21-1 were independent risk factors for lung cancer metastasis. AFP (98.3%) and CEA (89.3%) exhibited the highest sensitivity and specificity, respectively, while CYRFA21-1 had the highest area under the ROC curve value (0.875), with sensitivity and specificity values of 77.6 and 87.0%, respectively. Hence, CYFRA21-1 had the best diagnostic value.

scholarly journals Study of a Privacy Preserving Logistic Regression Algorithm (PPLRA) For Data Privacy in the Context of Big Data

2021 ◽  
Vol 2083 (3) ◽  
pp. 032059
Author(s):  
Qiang Chen ◽  
Meiling Deng
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Privacy Protection ◽  
Data Privacy ◽  
Absolute Error ◽  
Average Absolute Error ◽  
Regression Algorithms ◽  
Hadoop Platform ◽  
Logistic Regression Algorithm ◽  
Computing Speed

Abstract Regression algorithms are commonly used in machine learning. Based on encryption and privacy protection methods, the current key hot technology regression algorithm and the same encryption technology are studied. This paper proposes a PPLAR based algorithm. The correlation between data items is obtained by logistic regression formula. The algorithm is distributed and parallelized on Hadoop platform to improve the computing speed of the cluster while ensuring the average absolute error of the algorithm.

scholarly journals Association Between Triglyceride Glucose Index and Non-Small Cell Lung Cancer Risk in Chinese Population

2021 ◽  
Vol 11 ◽  
Author(s):  
Xin Yan ◽  
Yujuan Gao ◽  
Jingzhi Tong ◽  
Mi Tian ◽  
Jinghong Dai ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Cancer Risk ◽  
Small Cell Lung Cancer ◽  
Roc Curve ◽  
Small Cell ◽  
Small Cell Lung ◽  
Tyg Index ◽  
The Relationship

BackgroundNumerous studies showed that insulin resistance (IR) was associated with cancer risk. However, few studies investigated the relationship between IR and non-small cell lung cancer (NSCLC). The aim of this study is to explore the association of triglyceride glucose (TyG) index, a simple surrogate marker of IR, with NSCLC risk.Methods791 histologically confirmed NSCLC cases and 787 controls were enrolled in the present study. Fasting blood glucose and triglyceride were measured. The TyG index was calculated as ln [fasting triglycerides (mg/dl) ×fasting glucose (mg/dl)/2]. Logistic regression analysis was performed to estimate the relationship between NSCLC risk and the TyG index.ResultsThe TyG index was significantly higher in patients with NSCLC than that in controls (8.42 ± 0.55 vs 8.00 ± 0.45, P &lt; 0.01). Logistic regression analysis showed that the TyG index (OR = 3.651, 95%CI 2.461–5.417, P &lt; 0.001) was independently associated with NSCLC risk after adjusting for conventional risk factors. In addition, a continuous rise in the incidence of NSCLC was observed along the tertiles of the TyG index (29.4 vs 53.8 vs 67.2%, P &lt; 0.001). However, there were no differences of the TyG index in different pathological or TNM stages. In receiver operating characteristic (ROC) curve analysis, the optimal cut-off level for the TyG index to predict incident NSCLC was 8.18, and the area under the ROC curve (AUROC) was 0.713(95% CI 0.688–0.738).ConclusionsThe TyG index is significantly correlated with NSCLC risk, and it may be suitable as a predictor for NSCLC.

Predictors of early hospice or death in patients with inoperable lung cancer treated with curative intent.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e20525-e20525
Author(s):  
Anna Mary Brown Laucis ◽  
Kimberly A. Hochstedler ◽  
Thomas Pence Boike ◽  
Benjamin Movsas ◽  
Craig William Stevens ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Logistic Regression ◽  
Predictive Model ◽  
Performance Status ◽  
Ecog Performance Status ◽  
Curative Intent ◽  
Heart Dose ◽  
Patient Reported ◽  
Mean Heart Dose ◽  
Nsclc Patients

e20525 Background: Treatment for inoperable stage II-III non-small cell lung cancer (NSCLC) involves aggressive chemo-radiotherapy (CRT). While outcomes have improved with immunotherapy, some patients transition to hospice or die early in their treatment course. To help identify these patients, we developed a predictive model for early poor outcomes in NSCLC patients treated with curative intent. Methods: In a statewide consortium involving 27 sites, information was collected prospectively on stage II-III NSCLC patients who received curative CRT from April 2012 to November 2019. We defined an early poor outcome as termination of treatment due to hospice enrollment or death within 5 months of initiating radiation therapy. Potential predictors included clinical characteristics and patient reported outcomes (PROs) from validated questionnaires. Logistic regression models were used to assess potential predictors and build predictive models. Multiple imputation was used to handle missing data. We used Lasso regularized logistic regression to build a predictive model with multiple predictor variables. Results: Of the total of 2267 included patients, 128 patients discontinued treatment early due to hospice enrollment or death. The mean age of the 128 patients was 71 years old (range 48-91) and 59% received concurrent chemotherapy. Significant uni-variable predictors of early hospice or death were advanced age, worse ECOG performance status, high PTV volume, short distance to normal tissue critical structures, high mean heart dose, uninsured status, lower scores on the Functional and Physical Well-Being scale and the Lung Cancer Symptoms sub-scale of the FACT-L quality of life instrument, as well as higher levels of patient-reported lack of energy, cough, and shortness of breath. The best predictive model included age, ECOG performance status, PTV volume, mean heart dose, patient insurance status, and patient-reported lack of energy and cough. The pooled estimate of area under the curve (AUC) for this multivariable model was 0.71, with a negative predictive value of 95%, specificity of 97%, positive predictive value of 23%, and sensitivity of 16% at a predicted risk threshold of 20%. Conclusions: Our models identified a combination of clinical variables and PROs that may help identify individuals with inoperable NSCLC undergoing curative intent chemo-radiotherapy who are at a high risk of early hospice enrollment or death. These preliminary results are encouraging and warrant further evaluation in a larger cohort of patients.

scholarly journals Preoperative Risk Assessment of Lymph Node Metastasis in cT1 Lung Cancer: A Retrospective Study from Eastern China

2019 ◽  
Vol 2019 ◽  
pp. 1-9
Author(s):  
Chengyan Zhang ◽  
Guanchao Pang ◽  
Chengxi Ma ◽  
Jingni Wu ◽  
Pingli Wang ◽  
...  
Keyword(s):  
Risk Factors ◽  
Lung Cancer ◽  
Logistic Regression ◽  
Lymph Node ◽  
Lymph Node Metastasis ◽  
Predictive Model ◽  
Smoking History ◽  
Lymph Node Status ◽  
Node Metastasis ◽  
Node Status

Background. Lymph node status of clinical T1 (diameter≤3 cm) lung cancer largely affects the treatment strategies in the clinic. In order to assess lymph node status before operation, we aim to develop a noninvasive predictive model using preoperative clinical information. Methods. We retrospectively reviewed 924 patients (development group) and 380 patients (validation group) of clinical T1 lung cancer. Univariate analysis followed by polytomous logistic regression was performed to estimate different risk factors of lymph node metastasis between N1 and N2 diseases. A predictive model of N2 metastasis was established with dichotomous logistic regression, externally validated and compared with previous models. Results. Consolidation size and clinical N stage based on CT were two common independent risk factors for both N1 and N2 metastases, with different odds ratios. For N2 metastasis, we identified five independent predictors by dichotomous logistic regression: peripheral location, larger consolidation size, lymph node enlargement on CT, no smoking history, and higher levels of serum CEA. The model showed good calibration and discrimination ability in the development data, with the reasonable Hosmer-Lemeshow test (p=0.839) and the area under the ROC being 0.931 (95% CI: 0.906-0.955). When externally validated, the model showed a great negative predictive value of 97.6% and the AUC of our model was better than other models. Conclusion. In this study, we analyzed risk factors for both N1 and N2 metastases and built a predictive model to evaluate possibilities of N2 metastasis of clinical T1 lung cancers before the surgery. Our model will help to select patients with low probability of N2 metastasis and assist in clinical decision to further management.

Logistic Regression Analysis on the Survival Time for Patients with Lung Cancer

2015 ◽  
Vol 713-715 ◽  
pp. 1757-1760
Author(s):  
Sheng Ma
Keyword(s):  
Lung Cancer ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Data Analysis ◽  
Survival Time ◽  
Linear Model ◽  
Logistic Regression Analysis ◽  
Binary Classification ◽  
R Software

In the paper, the survival time of patients with lung cancer is inferred based on binary classification variables Logistic regression linear model and the data analysis is implemented by R software.

Sign in / Sign up

Export Citation Format

Share Document