RetroSearch Browse

Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://pubmed.ncbi.nlm.nih.gov/38586474/ below:

Enhancing cancer stage prediction through hybrid deep neural networks: a comparative study

doi: 10.3389/fdata.2024.1359703. eCollection 2024. Enhancing cancer stage prediction through hybrid deep neural networks: a comparative study

Affiliations

¹ Graduate School of Natural and Applied Sciences, Department of Computer Engineering, Gazi University, Ankara, Türkiye.
² Khoja Akhmet Yassawi International Kazakh-Turkish University, Faculty of Engineering, Department of Computer Engineering, Turkistan, Kazakhstan.
³ Department of Computer Engineering, Faculty of Technology, Gazi University, Ankara, Türkiye.

PMID: 38586474
PMCID: PMC10995364
DOI: 10.3389/fdata.2024.1359703

Item in Clipboard

Enhancing cancer stage prediction through hybrid deep neural networks: a comparative study

Alina Amanzholova et al. Front Big Data. 2024.

doi: 10.3389/fdata.2024.1359703. eCollection 2024. Affiliations

¹ Graduate School of Natural and Applied Sciences, Department of Computer Engineering, Gazi University, Ankara, Türkiye.
² Khoja Akhmet Yassawi International Kazakh-Turkish University, Faculty of Engineering, Department of Computer Engineering, Turkistan, Kazakhstan.
³ Department of Computer Engineering, Faculty of Technology, Gazi University, Ankara, Türkiye.

PMID: 38586474
PMCID: PMC10995364
DOI: 10.3389/fdata.2024.1359703

Item in Clipboard

Abstract

Efficiently detecting and treating cancer at an early stage is crucial to improve the overall treatment process and mitigate the risk of disease progression. In the realm of research, the utilization of artificial intelligence technologies holds significant promise for enhancing advanced cancer diagnosis. Nonetheless, a notable hurdle arises when striving for precise cancer-stage diagnoses through the analysis of gene sets. Issues such as limited sample volumes, data dispersion, overfitting, and the use of linear classifiers with simple parameters hinder prediction performance. This study introduces an innovative approach for predicting early and late-stage cancers by integrating hybrid deep neural networks. A deep neural network classifier, developed using the open-source TensorFlow library and Keras network, incorporates a novel method that combines genetic algorithms, Extreme Learning Machines (ELM), and Deep Belief Networks (DBN). Specifically, two evolutionary techniques, DBN-ELM-BP and DBN-ELM-ELM, are proposed and evaluated using data from The Cancer Genome Atlas (TCGA), encompassing mRNA expression, miRNA levels, DNA methylation, and clinical information. The models demonstrate outstanding prediction accuracy (89.35%-98.75%) in distinguishing between early- and late-stage cancers. Comparative analysis against existing methods in the literature using the same cancer dataset reveals the superiority of the proposed hybrid method, highlighting its enhanced accuracy in cancer stage prediction.

Keywords: DNA methylation; artificial intelligence; cancer stage prediction; deep belief network; mRNA expression.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures
Figure 1

Quantification of TCGA repository samples…

Figure 1

Quantification of TCGA repository samples by tumor type and biotechnological analysis (Liñares-Blanco et…

Figure 1

Quantification of TCGA repository samples by tumor type and biotechnological analysis (Liñares-Blanco et al., 2021).

Figure 2

Heatmap of the expression levels…

Figure 2

Heatmap of the expression levels of DNA methylation hyperparameter for a selected gene…

Figure 2

Heatmap of the expression levels of DNA methylation hyperparameter for a selected gene across TCGA samples, specifically focusing on the KIRP, KIRC, LUSC, and HNSC cancer types.

Figure 3

Structure of DBN-ELM and DBN-ELM-BP.

Figure 3

Structure of DBN-ELM and DBN-ELM-BP.

Figure 3

Structure of DBN-ELM and DBN-ELM-BP.

Figure 4

The confusion matrix for binary…

Figure 4

The confusion matrix for binary prediction obtained from various CNN approaches for the…

Figure 4

The confusion matrix for binary prediction obtained from various CNN approaches for the KIRC cancer type. The subfigures depict different models: (A) DBN, (B) WE-DBN, (C) DBN-ELM, and (D) DBN-ELM-BP.

Figure 5

Validation and training loss profiles…

Figure 5

Validation and training loss profiles for the KIRP, LUSC, and HNSC datasets based…

Figure 5

Validation and training loss profiles for the KIRP, LUSC, and HNSC datasets based on the proposed DBN-ELM-BP network for multi-omics data.

Figure 6

Comparative analysis of the accuracy…

Figure 6

Comparative analysis of the accuracy of the proposed DBN-ELM-BP model for multi-omics data…

Figure 6

Comparative analysis of the accuracy of the proposed DBN-ELM-BP model for multi-omics data about other networks across three distinct datasets: (A) KIRC, (B) LUSC, and (C) HNSC.

Figure 7

Two distinct aspects related to…

Figure 7

Two distinct aspects related to the KIRP datasets: (A) accuracy at each iteration…

Figure 7

Two distinct aspects related to the KIRP datasets: (A) accuracy at each iteration and (B) the ROC curve.