<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>For cancer patients, accurate estimation of patients’ survival time facilitates medical personnel to formulate treatment measures while avoiding the negative impact of poor treatment programs. In this paper, in order to overcome the limitations of the single-factor survival analysis model, the Cox proportional risk analysis method is proposed and combined with Lasson regression analysis at the same time to establish a cancer prognosis risk assessment model. After extracting and preliminary screening cancer gene features and then analyzing the correlation between the features, the most important indicators used for cancer prognosis assessment were summarized. In the correlation analysis of cancer indicators, the number of factors showing a positive correlation with lncRAN in cancer cells was 136, and the number of negative correlations was 92. The correlation coefficients ranged from −2.65 to 0.52. After the ROC curve evaluated the predictive risk model, the areas of 1-, 3-, and 5-year OS curves of cancer patients were 0.7854, 0.8462, and 0.7855, respectively. The number of deaths increased gradually along with the increase in the risk scores, and the model predicted the results more accurately.</p>
</abstract>

For cancer patients, accurate estimation of patients’ survival time facilitates medical personnel to formulate treatment measures while avoiding the negative impact of poor treatment programs. In this paper, in order to overcome the limitations of the single-factor survival analysis model, the Cox proportional risk analysis method is proposed and combined with Lasson regression analysis at the same time to establish a cancer prognosis risk assessment model. After extracting and preliminary screening cancer gene features and then analyzing the correlation between the features, the most important indicators used for cancer prognosis assessment were summarized. In the correlation analysis of cancer indicators, the number of factors showing a positive correlation with lncRAN in cancer cells was 136, and the number of negative correlations was 92. The correlation coefficients ranged from −2.65 to 0.52. After the ROC curve evaluated the predictive risk model, the areas of 1-, 3-, and 5-year OS curves of cancer patients were 0.7854, 0.8462, and 0.7855, respectively. The number of deaths increased gradually along with the increase in the risk scores, and the model predicted the results more accurately.

Application of statistical models based on survival analysis in the assessment of cancer prognosis

Shandong University of Finance and Economics

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

For cancer patients, accurate estimation of patients’ survival time facilitates medical personnel to formulate treatment measures while avoiding the negative...

Application of statistical models based on survival analysis in the assessment of cancer prognosis

Published Online: May 30, 2024

Page range: -

Received: Feb 11, 2024

Accepted: Apr 14, 2024

DOI: https://doi.org/10.2478/amns-2024-1224

Keywords
One-way survival analysis, Cox proportional risk analysis, Lasson regression, Cancer prognostic assessment

© 2024 Xinren Zhang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Application of statistical models based on survival analysis in the assessment of cancer prognosis

Published Online: May 30, 2024

Page range: -

Received: Feb 11, 2024

Accepted: Apr 14, 2024

DOI: https://doi.org/10.2478/amns-2024-1224

KeywordsOne-way survival analysis, Cox proportional risk analysis, Lasson regression, Cancer prognostic assessment

© 2024 Xinren Zhang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
One-way survival analysis, Cox proportional risk analysis, Lasson regression, Cancer prognostic assessment