Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Mar 26;24(7):6250.
doi: 10.3390/ijms24076250.

Integrating AI/ML Models for Patient Stratification Leveraging Omics Dataset and Clinical Biomarkers from COVID-19 Patients: A Promising Approach to Personalized Medicine

Babatunde Bello  1 , Yogesh N Bundey  1 , Roshan Bhave  1 , Maksim Khotimchenko  1 , Szczepan W Baran  1 , Kaushik Chakravarty  1 , Jyotika Varshney  1
Affiliations

Integrating AI/ML Models for Patient Stratification Leveraging Omics Dataset and Clinical Biomarkers from COVID-19 Patients: A Promising Approach to Personalized Medicine

Babatunde Bello et al. Int J Mol Sci. .

Abstract

The COVID-19 pandemic has presented an unprecedented challenge to the healthcare system. Identifying the genomics and clinical biomarkers for effective patient stratification and management is critical to controlling the spread of the disease. Omics datasets provide a wealth of information that can aid in understanding the underlying molecular mechanisms of COVID-19 and identifying potential biomarkers for patient stratification. Artificial intelligence (AI) and machine learning (ML) algorithms have been increasingly used to analyze large-scale omics and clinical datasets for patient stratification. In this manuscript, we demonstrate the recent advances and predictive accuracies in AI- and ML-based patient stratification modeling linking omics and clinical biomarker datasets, focusing on COVID-19 patients. Our ML model not only demonstrates that clinical features are enough of an indicator of COVID-19 severity and survival, but also infers what clinical features are more impactful, which makes our approach a useful guide for clinicians for prioritization best-fit therapeutics for a given cohort of patients. Moreover, with weighted gene network analysis, we are able to provide insights into gene networks that have a significant association with COVID-19 severity and clinical features. Finally, we have demonstrated the importance of clinical biomarkers in identifying high-risk patients and predicting disease progression.

Keywords: COVID-19; SARS-CoV-2; artificial intelligence; omics; patient stratification; risk management.

PubMed Disclaimer

Conflict of interest statement

The authors are all employees of VeriSIM Life and used a proprietary AI/ML-driven platform to generate the outcomes for the manuscript.

Figures

Figure 1
Figure 1
Overview of the workflow for an AI/ML-driven model that predicts the outcome of COVID-19 infection in patients, utilizing a set of clinical biomarkers and genomics dataset.
Figure 2
Figure 2
ROC curve on test set for COVID-19 severity prediction model.
Figure 3
Figure 3
ROC curve on test set for COVID-19 survival prediction model.
Figure 4
Figure 4
ROC curve on test set for COVID-19 severity prediction model with comorbidities re-moved from training set.
Figure 5
Figure 5
ROC curve on the test set for COVID-19 survival prediction model with comorbidities removed from the training set.
Figure 6
Figure 6
Top 20 most impactful features for COVID-19 severity predictive model. The red data points indicate higher values for the particular feature, while blue data points indicate lower values for the particular feature. A higher feature impact score indicates a higher contribution to a “severe” case prediction, while a lower feature impact score indicates a higher contribution to a “moderate” case prediction value.
Figure 7
Figure 7
Top 20 most impactful features for COVID-19 survival predictive model. The red data points indicate higher values for the particular feature, while blue data points indicate lower values for the particular feature. A higher feature impact score indicates a higher contribution to a “no survival from COVID-19” prediction, while a lower feature impact score indicates a higher contribution to a “survival from COVID-19” prediction value.
Figure 8
Figure 8
Range of impactful and validated biomarker values for COVID-19 severity prediction.
Figure 8
Figure 8
Range of impactful and validated biomarker values for COVID-19 severity prediction.
Figure 9
Figure 9
Range of impactful and validated biomarker values for COVID-19 survival prediction.
Figure 9
Figure 9
Range of impactful and validated biomarker values for COVID-19 survival prediction.
Figure 10
Figure 10
Heatmaps of module traits for severity and clinical biomarkers. The boxes indicate the correlation based on module eigengenes in the rows and traits in the column. The color legend—blue (negative correlation) and red (positive correlation). p-values represented by asterisks indicated significance. Asterix signs indicate the order of significance where ***—very significant, **—significant, and *—significant.
Figure 11
Figure 11
Heatmaps of module traits for severity and disease comorbidity. The boxes indicate the correlation based on module eigengenes in the rows and traits in the column. The color legend—blue (negative correlation) and red (positive correlation). p-values represented by asterisks indicated significance. Asterix signs indicate the order of significance where ***—very significant, **—significant, and *—significant.
Figure 12
Figure 12
Gene network for top 20 enriched pathways for genes module, MEcyan based on GO biological process annotation.
Figure 13
Figure 13
Dot plot for top 20 enriched pathways based for gene module, MEcyan based on KEGG.
Figure 14
Figure 14
Gene network for top 20 enriched pathways for genes module, MEdarkred based on GO biological process annotation.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Lee B., Lewis G., Agyei-Manu E., Atkins N., Bhattacharyya U., Dozier M., Rostron J., Sheikh A., McQuillan R., Theodoratou E., et al. Risk of serious COVID-19 outcomes among adults and children with moderate-to-severe asthma: A systematic review and meta-analysis. Eur. Respir. Rev. 2022;31:220066. doi: 10.1183/16000617.0066-2022. - DOI - PMC - PubMed
    1. Song C.Y., Xu J., He J.Q., Lu Y.Q. Immune dysfunction following COVID-19, especially in severe patients. Sci. Rep. 2020;10:15838. doi: 10.1038/s41598-020-72718-9. - DOI - PMC - PubMed
    1. Zhang J., Yang W., Huang F., Zhang K. Effectiveness of mRNA and viral-vector vaccines in epidemic period led by different SARS-CoV-2 variants: A systematic review and meta-analysis. J. Med. Virol. 2023;95:e28623. doi: 10.1002/jmv.28623. - DOI - PubMed
    1. Whiteside T., Kane E., Aljohani B., Alsamman M., Pourmand A. Redesigning emergency department operations amidst a viral pandemic. Am. J. Emerg. Med. 2020;38:1448–1453. doi: 10.1016/j.ajem.2020.04.032. - DOI - PMC - PubMed
    1. Sipior J.C. Considerations for development and use of AI in response to COVID-19. Int. J. Inf. Manag. 2020;55:102170. doi: 10.1016/j.ijinfomgt.2020.102170. - DOI - PMC - PubMed