Machine learning to identify lung cancer with tuberculosis from isolated tuberculosis (Preprint)

2021 ◽  
Author(s):  
Zhenhao Li
Keyword(s):  
Machine Learning ◽  
Lung Cancer ◽  
Cancer Patients ◽  
Tumor Markers ◽  
Learning Algorithm ◽  
Laboratory Data ◽  
Pathological Examination ◽  
Training Set ◽  
Lung Cancer Patients ◽  
Tb Pcr

UNSTRUCTURED Tuberculosis (TB) is a precipitating cause of lung cancer. Lung cancer patients coexisting with TB is difficult to differentiate from isolated TB patients. The aim of this study is to develop a prediction model in identifying those two diseases between the comorbidities and TB. In this work, based on the laboratory data from 389 patients, 81 features, including main laboratory examination of blood test, biochemical test, coagulation assay, tumor markers and baseline information, were initially used as integrated markers and then reduced to form a discrimination system consisting of 31 top-ranked indices. Patients diagnosed with TB PCR ＞1mtb/ml as negative samples, lung cancer patients with TB were confirmed by pathological examination and TB PCR ＞1mtb/ml as positive samples. We used Spatially Uniform ReliefF (SURF) algorithm to determine feature importance, and the predictive model was built using machine learning algorithm Random Forest. For cross-validation, the samples were randomly split into four training set and one test set. The selected features are composed of four tumor markers (Scc, Cyfra21-1, CEA, ProGRP and NSE), fifteen blood biochemical indices (GLU, IBIL, K, CL, Ur, NA, TBA, CHOL, SA, TG, A/G, AST, CA, CREA and CRP), six routine blood indices (EO#, EO%, MCV, RDW-S, LY# and MPV) and four coagulation indices (APTT ratio, APTT, PTA, TT ratio). This model presented a robust and stable classification performance, which can easily differentiate the comorbidity group from the isolated TB group with AUC, ACC, sensitivity and specificity of 0.8817, 0.8654, 0.8594 and 0.8656 for the training set, respectively. Overall, this work may provide a novel strategy for identifying the TB patients with lung cancer from routine admission lab examination with advantages of being timely and economical. It also indicated that our model with enough indices may further increase the effectiveness and efficiency of diagnosis.

scholarly journals ECMarker: interpretable machine learning model identifies gene expression biomarkers predicting clinical outcomes and reveals molecular mechanisms of human disease in early stages

2020 ◽  
Author(s):  
Ting Jin ◽  
Nam D Nguyen ◽  
Flaminia Talos ◽  
Daifeng Wang
Keyword(s):  
Gene Expression ◽  
Machine Learning ◽  
Lung Cancer ◽  
Cancer Patients ◽  
Human Disease ◽  
Gene Networks ◽  
Regulatory Mechanisms ◽  
Disease Development ◽  
Lung Cancer Patients ◽  
Machine Learning Model

Abstract Motivation Gene expression and regulation, a key molecular mechanism driving human disease development, remains elusive, especially at early stages. Integrating the increasing amount of population-level genomic data and understanding gene regulatory mechanisms in disease development are still challenging. Machine learning has emerged to solve this, but many machine learning methods were typically limited to building an accurate prediction model as a ‘black box’, barely providing biological and clinical interpretability from the box. Results To address these challenges, we developed an interpretable and scalable machine learning model, ECMarker, to predict gene expression biomarkers for disease phenotypes and simultaneously reveal underlying regulatory mechanisms. Particularly, ECMarker is built on the integration of semi- and discriminative-restricted Boltzmann machines, a neural network model for classification allowing lateral connections at the input gene layer. This interpretable model is scalable without needing any prior feature selection and enables directly modeling and prioritizing genes and revealing potential gene networks (from lateral connections) for the phenotypes. With application to the gene expression data of non-small-cell lung cancer patients, we found that ECMarker not only achieved a relatively high accuracy for predicting cancer stages but also identified the biomarker genes and gene networks implying the regulatory mechanisms in the lung cancer development. In addition, ECMarker demonstrates clinical interpretability as its prioritized biomarker genes can predict survival rates of early lung cancer patients (P-value < 0.005). Finally, we identified a number of drugs currently in clinical use for late stages or other cancers with effects on these early lung cancer biomarkers, suggesting potential novel candidates on early cancer medicine. Availabilityand implementation ECMarker is open source as a general-purpose tool at https://github.com/daifengwanglab/ECMarker. Contact [email protected] Supplementary information Supplementary data are available at Bioinformatics online.

Training and validation study for sequential monitoring of CAMLs in circulation to predict ongoing progression in lung cancer patients undergoing definitive radiotherapy.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. 3053-3053
Author(s):  
Daniel Adams ◽  
Jianzhong He ◽  
Yawei Qiao ◽  
Ting Xu ◽  
Hui Gao ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Radiation Therapy ◽  
Cancer Patients ◽  
Stage Iv ◽  
Progression Free Survival ◽  
Training Set ◽  
Lung Cancer Patients ◽  
Therapeutic Decisions ◽  
Validation Set ◽  
Nsclc Patients

3053 Background: Cancer Associated Macrophage-Like cells (CAMLs) are a recently described circulating stromal cell common in the peripheral blood of cancer patients that are prognostic for progressive disease. Further, it has been shown that changes in CAML size (i.e. enlargement above 50µm) can predict progression free survival (PFS) in thoracic cancers (e.g. lung). We enrolled 104 unresectable non-small cell lung cancer (NSCLC) patients, with an initial training set review of 54 patients, to determine if change in CAML size after radiation therapy was predictive PFS. Methods: A 2 year single blind prospective study was undertaken to test the relationship of ≥50µm CAMLs to PFS based on imaging in lung patients before and after induction of chemo radiation, or radiation therapy. To achieve a 2-tailed 90% power (α = 0.05) we recruited a training set of 54 patients and validation set of 50 patients all with pathologically confirmed unresectable NSCLC: Stage I (n = 14), Stage II (n = 16), Stage III (n = 61) & Stage IV (n = 13). Baseline (BL) blood samples were taken prior to start of therapy & a 2nd blood sample (T1) was taken after completion of radiotherapy (~30 days). Blood was filtered by CellSieve filtration and CAMLs quantified. Analysis by CAML size of < 49 µm or ≥50 µm was used to evaluate PFS hazard ratios (HRs) by censored univariate & multivariate analysis. Results: CAMLs were found in 95% of samples averaging 2.7 CAMLs/7.5mL sample at BL, with CAMLs ≥50 µm having reduced PFS (HR = 2.2, 95%CI1.3-3.8, p = 0.003). At T1, 18 patients had increased CAML size ≥50 µm with PFS (HR = 4.6, 95%CI2.5-8.3, p < 0.001). In total, ≥50 µm CAMLs at BL was 76% accurate at predicting progression within 24 months while ≥50 µm CAMLs at T1 was 83% accurate at predicting progression. Conclusions: In unresectable NSCLC patients, enlargement of CAMLs during treatment is an indicator active progression. We identify that a single ≥50 µm CAML after induction of radiotherapy, in our training set and confirmed in our validation set, is an indicator of poor prognosis. We suggest that changes in CAML size during therapy may indicate the efficacy of therapy and could potentially help shape subsequent therapeutic decisions.

Comparison and analysis among expression of CRP and tumor markers in advanced non-small cell lung cancer patients

2006 ◽  
Vol 24 (18_suppl) ◽  
pp. 17126-17126
Author(s):  
W. Dong ◽  
X. Ren ◽  
P. Liu
Keyword(s):  
Lung Cancer ◽  
Small Cell Lung Cancer ◽  
Cancer Patients ◽  
Tumor Markers ◽  
Cell Lung Cancer ◽  
Squamous Carcinoma ◽  
Small Cell ◽  
Small Cell Lung ◽  
Lung Cancer Patients ◽  
Significant Difference

17126 Background: To investigate the correlation of C-reactive protein (CRP) and tumor markers such as carcinoembryonic antigen (CEA), CA125, cytokeratin 19 antibody (Cyfra 21–1) in non-small cell lung cancer (NSCLC) patients. Methods: CRP, CEA, CA125, Cyfra 21–1 were measured in the serum of 79 patients with advanced non-small cell lung cancer (stage III and IV ) by independent samples t test. Results: It showed statistically significant difference (P < 0.05) in the level of CRP, CEA, CA125, Cyfra211 between lung cancer patients (25.21 ± 19.12 mg/L, 62.89 ± 53.96 ng/L, 46.36 ± 30.03 U/L, 6.85 ± 2.42 ng/L, respectively) and the control subjects. CRP, CA125 showed no significant difference between squamous carcinoma and adenocarcinoma (P = 0.832, 0.406, respectively). CEA was higher in adenocarcinoma than in squamous carcinoma (P = 0.002). Cyfra211 was lower in adenocarcinoma than in squamous carcinoma (P = 0.039). The increase of CRP was accompanied with the increase of CEA, CA125 and Cyfra211 (p < 0.05). At the same time, CRP elevated while CEA, CA125, Cyfra211 increased (p < 0.05). Conclusions: All of CRP, CEA, CA125 and Cyfra211 increased in advanced non-small lung cancer. Combined monitoring on CRP with CEA, CA125, Cyfra211 may be a complementary method in advanced non-small cell lung cancer patients. No significant financial relationships to disclose.

The diagnostic value of serum tumor markers in newly diagnosed lung cancer patients

2017 ◽  
Author(s):  
Elif Tugce Korkmaz ◽  
Deniz Koksal ◽  
Funda Aksu ◽  
Z. Gunnur Dikmen ◽  
Duygu Icen ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Cancer Patients ◽  
Tumor Markers ◽  
Diagnostic Value ◽  
Newly Diagnosed ◽  
Lung Cancer Patients ◽  
Serum Tumor Markers
Sign in / Sign up

Export Citation Format

Share Document