Improving Multi-Tumor Biomarker Health Check-Up Tests with Machine Learning Algorithms

Wang, Hsin-Yao; Chen, Chun‐Hsien; Shi, Steve; Chung, Chia-Ru; Wen, Yu‐Chuan; Wu, Min‐Hsien; Lebowitz, Michael; Zhou, Jiaqi; Lu, Jang‐Jih

doi:10.3390/cancers12061442

Cited by 21 publications

(21 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is not accurate to use individual gene to judge the prognosis of cancer patients owing to the differences in race, region and organ specificity [33,34]. In recent years, with the development of computer technology, researchers have begun to use regression models, artificial intelligence and other computational methods to screen and analyze gene expression data in cancer databases on a large scale to explore prognostic biomarkers associated with specific cancers, such as this study [35]. LRFN4, ADAMTS12, MCEMP1, HP and MUC15 are all cancerrelated biomarkers, and all of them can also be used in judgment of cancer incidence in an independent manner.…”

Section: Discussionmentioning

confidence: 99%

Establishment of a 5-gene risk model related to regulatory T cells for predicting gastric cancer prognosis

Sun

Jiang

et al. 2020

Cancer Cell Int

View full text Add to dashboard Cite

Background: Gastric cancer (GC) is one of the high-risk cancers that lacks effective methods for prognosis prediction. Therefore, we searched for immune cells related to the prognosis of GC and studied the role of related genes in GC prognosis. Methods: In this study, we collected the mRNA data of GC from The Cancer Genome Atlas (TCGA) database and studied the immune cells that were closely related to the prognosis of GC. Spearman correlation analysis was performed to show the association between immune cell-related genes and the differentially expressed genes (DEGs) of GC. Univariate and multivariate Cox regression analyses were conducted on the immune cell-related genes with a high correlation with GC. A prognostic risk score model was constructed and the most significant feature genes were identified. Kaplan-Meier method was then used to compare the overall survival (OS) of patients with high-risk and low-risk, and receiver operating characteristic (ROC) analysis was used to assess the accuracy of the risk model. In addition, GC patients were grouped according to the median expression of the features genes, and survival analysis was further carried out. Results: It was noted that regulatory T cells (Tregs) were significantly correlated with the prognosis of GC, and 172 genes related to Tregs were found to be closely associated with GC. An optimal prognostic risk model was constructed, and a 5-gene (including LRFN4, ADAMTS12, MCEMP1, HP and MUC15) signature-based risk score was established. Survival analysis showed significant difference in OS between low-risk and high-risk samples. ROC analysis results indicated that the risk model had a high accuracy for the prognosis prediction of samples (AUC = 0.717). The results of survival analysis on each feature gene based on expression levels were consistent with the results of multivariate Cox analysis for predicting the risk rate of the 5 genes. Conclusion: These results proved that the 5-gene signature-based risk score could be used to predict the survival of GC patients, and these 5 genes were closely related to Tregs. These findings are of great significance for studying the role of immune cells and related immune factors in regulating the prognosis of GC.

show abstract

Section: Discussionmentioning

confidence: 99%

Establishment of a 5-gene risk model related to regulatory T cells for predicting gastric cancer prognosis

Sun

Jiang

et al. 2020

Cancer Cell Int

View full text Add to dashboard Cite

show abstract

“…Many studies have aimed at improving the diagnostic accuracy in bladder cancer. Wang et al utilized machine learning algorithms to improve the tumor marker-based screening for multiple cancers and yielded a sensitivity of 0.81 and a specificity of 0.64 [ 82 ]. Shao et al applied ultra-performance liquid chromatography coupled with time-of-flight mass spectrometry to acquire metabolites profiles in 152 samples from patients with bladder cancer and hernia; furthermore, the decision tree model embedded in this study obtained an accuracy of 76.6%, a sensitivity of 71.88%, and a specificity of 86.67% [ 83 ].…”

Section: Discussionmentioning

confidence: 99%

Machine Learning in Prediction of Bladder Cancer on Clinical Laboratory Data

et al. 2022

View full text Add to dashboard Cite

Bladder cancer has been increasing globally. Urinary cytology is considered a major screening method for bladder cancer, but it has poor sensitivity. This study aimed to utilize clinical laboratory data and machine learning methods to build predictive models of bladder cancer. A total of 1336 patients with cystitis, bladder cancer, kidney cancer, uterus cancer, and prostate cancer were enrolled in this study. Two-step feature selection combined with WEKA and forward selection was performed. Furthermore, five machine learning models, including decision tree, random forest, support vector machine, extreme gradient boosting (XGBoost), and light gradient boosting machine (GBM) were applied. Features, including calcium, alkaline phosphatase (ALP), albumin, urine ketone, urine occult blood, creatinine, alanine aminotransferase (ALT), and diabetes were selected. The lightGBM model obtained an accuracy of 84.8% to 86.9%, a sensitivity 84% to 87.8%, a specificity of 82.9% to 86.7%, and an area under the curve (AUC) of 0.88 to 0.92 in discriminating bladder cancer from cystitis and other cancers. Our study provides a demonstration of utilizing clinical laboratory data to predict bladder cancer.

show abstract

“…This application of liquid biopsy to cancer screening in the general population raises concerns about the frequency of false-positive results, which can lead to unnecessary further testing or invasive procedures [24] . The clinical utility of liquid biopsy for cancer screening could be improved through the application of machine learning-based approaches to optimize the interpretation of multi-analyte screening results [25,26] . The current literature lacks objective published data supporting the effectiveness or potential economic impact of including multi-analyte cancer biomarker panels in wellness checks; more studies are needed to determine the best way to integrate liquid biopsy into general health screening programs.…”

Section: Application Of Liquid Biopsy To Population Cancer Screeningmentioning

confidence: 99%

Current and future opportunities for liquid biopsy of circulating biomarkers to aid in early cancer detection

Gawel¹,

Jackson²,

Jeanblanc³

et al. 2022

J Cancer Metastasis Treat

View full text Add to dashboard Cite

Early diagnosis of cancer can significantly improve treatment and survival outcomes. Imaging and tissue biopsy are the gold standard diagnostic approaches but are costly, invasive, and often unable to detect early-stage tumors. The past decade has marked an acceleration in the discovery and development of liquid biopsy tests for aiding in the detection of various types of tumor markers in non-tissue samples, such as blood. Liquid biopsy markers include circulating tumor cells, as well as tumor cell fragments, nucleic acids, and proteins. Liquid biopsy may be useful in screening patients considered to be at high risk of developing cancer, for refining diagnosis when combined with other test results, and for early detection of recurrence. Advances in big data analytics, informatics, and artificial intelligence will make it possible to combine patient history, clinical data, and liquid biopsy marker profiles to achieve more accurate and earlier diagnosis. In this review, we summarize the current use of liquid biopsy in cancer care, including the development of multi-analyte panels to improve diagnostic accuracy and detect several cancer types in a single assay. We highlight recent advances for potential future applications of liquid biopsy to aid in the diagnosis of early-stage lung cancer. We also discuss the opportunities and challenges of integrating liquid biopsy into current algorithms for cancer screening and diagnosis.

show abstract

Improving Multi-Tumor Biomarker Health Check-Up Tests with Machine Learning Algorithms

Cited by 21 publications

References 36 publications

Establishment of a 5-gene risk model related to regulatory T cells for predicting gastric cancer prognosis

Establishment of a 5-gene risk model related to regulatory T cells for predicting gastric cancer prognosis

Machine Learning in Prediction of Bladder Cancer on Clinical Laboratory Data

Current and future opportunities for liquid biopsy of circulating biomarkers to aid in early cancer detection

Contact Info

Product

Resources

About