Future Perspectives: Implementation of Artificial Intelligence in Umbilical Cord Biobanking

Biobanking, July 9 2025

Introduction

The integration and implementation of Artificial Intelligence (AI) in umbilical cord biobanking is still in its early stages. However, AI holds great potential to be gradually applied across multiple aspects of cord blood and tissue banking systems. This article presents a detailed overview of the key areas in which AI could play a pivotal role in the field of umbilical cord biobanking.

Key Areas in Umbilical Cord Biobanking Where AI Can Be Applied

1. Informed Consent Management

AI systems—including machine learning (ML) and natural language processing (NLP) algorithms—can be designed to interpret and explain the content of informed consent forms and assist with web-based communication between donors and biobank staff.
If a participant chooses to withdraw consent, the AI system can automatically delete related data and notify biobank administrators to dispose of the associated biological specimen.

2. Sample Management

Integrating AI into automated sample storage systems allows for better utilization of storage space by efficiently managing sample positioning.

AI can also be programmed to generate standard operating procedures (SOPs) tailored to different types of biological specimens and to identify or match suitable samples with specific research studies.

Moreover, AI can generate prospective collection plans for biomedical research based on inventory trends, distribution data, and current research demand.

3. Human Leukocyte Antigen (HLA) Typing

HLA genotyping is a regulatory requirement in umbilical cord blood banking. Deep learning and machine learning algorithms have been developed to predict HLA genotypes and match recipients with suitable cord blood units.
These AI-based methods are faster and more cost-effective than traditional laboratory-based techniques such as direct sequencing, allele-specific amplification, or hybridization.

• HLA*IMP: Uses genome-wide SNP data to impute HLA genotypes with 92–98% accuracy.
• HLA*IMP:02: Improves accuracy by incorporating population and ethnic diversity to handle genetic variability.
• HIBAG: Utilizes “attribute bagging” for enhanced accuracy compared to HLA*IMP.
• DEEP*HLA: A convolutional neural network (CNN) that accurately predicts rare HLA alleles at high speed, ideal for large-scale biobank data.

4. Diagnostic Screening

Screening for infectious diseases from maternal blood is essential to ensure the safety of cord blood samples. AI plays a critical role here.

For example, a 2022 study by Mohammad M. et al. employed a stacked ensemble model—including logistic regression and non-parametric methods—to detect hepatitis C virus (HCV) with 97% precision, 66% higher than standard logistic regression.

AI is also useful in chromosomal analysis (karyotyping):

• Ikaros: A deep learning-based software for accurate and rapid chromosome image analysis.
• ChromoEnhancer: Uses CycleGAN to enhance malignant karyotype images, aiding the detection of hidden chromosomal abnormalities.

5. Genomic Data Provision

In umbilical cord biobanks that provide genomic datasets, methods like Sure Independence Screening (SIS) are valuable for SNP analysis in genome-wide association studies (GWAS) and gene-gene interaction mapping.

• EPISIS: A software tool using SIS to identify epistatic interactions between genes linked to severe forms of Stevens-Johnson syndrome.
• Other methods, such as artificial neural networks (ANNs), lasso regression, support vector machines (SVM), and random forest (RF), are also employed to predict phenotypes or disease risk.
• ExPecto: A modern deep learning model capable of predicting common and rare genetic variants from DNA sequence data.

6. Detection of Data Errors in Biospecimen Records

Advanced machine learning algorithms such as kurPCA (which combines Principal Component Analysis with kurtosis) and RAMP (Regression Adjustment Method for Systematic Missing Patterns) can identify outliers and inconsistencies in biospecimen-related datasets, reducing the need for extensive manual data cleaning.

7. Quality Assessment

AI supports the quality evaluation of umbilical cord-derived mesenchymal stem cells (UC-MSCs), helping streamline the process from storage to cell therapy.

• Marklein et al., 2019: Used viSNE (a machine learning approach) to identify morphological subpopulations of IFN-γ-stimulated UC-MSCs, which were associated with different immunosuppressive capacities.
• Mota et al., 2021: Developed an ML algorithm to classify UC-MSCs by their morphological phenotype and replication rate—slower-replicating cells had impaired differentiation potential and thus lower therapeutic utility.

AI can help identify high-potential cell populations without invasive testing.

8. Predictive Modeling

Data mining techniques such as decision trees, artificial neural networks (ANNs), and SVMs can be used to develop predictive models for post-transplant complications following allogeneic hematopoietic stem cell transplantation (allo-HSCT), assisting in donor and patient selection.

• Giovanni et al.: Found ANN models had 83.3% sensitivity in predicting acute graft-versus-host disease (aGvHD) and 90.1% accuracy in predicting the absence of aGvHD after HSCT.

Conclusion

AI is not yet widely implemented in umbilical cord biobanking, but its potential is enormous. If deployed appropriately, AI could revolutionize the biobanking process—enhancing the use of cord blood stem cells for both clinical applications and biomedical research.

Further studies to validate and tailor AI algorithms specifically for cord blood biobanking are essential to expand their real-world applicability in this field.

References

1. Annaratone L, De Palma G, Bonizzi G, et al. 2021. Basic principles of biobanking: from biological samples to precision medicine for patients. Virchows Archiv. 2021;479(2):233–246.
2. Ballen KK, Verter F, Kurtzberg J. 2012. Umbilical cord blood donation: public or private? Bone Marrow Transplant. 2015;50:1271–1278.
3. Battineni G, Hossain MA, Chintalapudi N, Amenta F. 2022. A survey on the role of artificial intelligence in biobanking studies: a systematic review. Diagnostics. 2022;12:1179.
4. Bokhari Y, Alhareeri A, Aljouie A, et al.2022.ChromoEnhancer: an artificial-intelligence-based tool to enhance neoplastic karyograms as an aid for effective analysis. Cells. 2022;11:2244.
5. Butler MG, Menitove JE.2011. Umbilical cord blood banking: an update. J Assist Reprod Genet. 2011;28:669–676.
6. Caocci G, Baccoli R, Vacca A, et al.2010. Comparison between an artificial neural network and logistic regression in predicting acute graft-vs-host disease after unrelated donor hematopoietic stem cell transplantation in thalassemia patients. Exp Hematol. 2010;38:426–433.
7. Cirillo D, Valencia A.2019. Big data analytics for personalized medicine. Curr Opin Biotechnol. 2019;58:161–167.
8. Committee on Establishing a National Cord Blood Stem Cell Bank Program, et al. 2005. Cord Blood: Establishing a National Hematopoietic Stem Cell Bank Program. National Academies Press; 2005.
9. Dagher G.2022. Quality matters: international standards for biobanking. Cell Prolif. 2022;55.
10. De Antonio M., Timothy J.W.D., James Philip H., O’Regan D.P 2020. Artificial intelligence and the cardiologist: What you need to know for 2020. 2020;106:39
11. Dilthey AT, Moutsianas L, Leslie S, McVean G.2011. HLA*IMP—an integrated framework for imputing classical HLA alleles from SNP genotypes. Bioinformatics. 2011;27:968–972.
12. Dilthey A, Leslie S, Moutsianas L, et al.2013. Multi-population classical HLA type imputation.PLoS Comput Biol. 2013;9:e1002877.
13. Knoppers BM, Isasi R. 2010. Stem cell banking: between traceability and identifiability. Genome Med. 2010;2:73.
14. Kim G, Jeon JH, Park K, et al.2022. High throughput screening of mesenchymal stem cell lines using deep learning. Sci Rep. 2022;12:17507.
15. Lee J-E. 2018.  Artificial intelligence in the future biobanking: current issues in the biobank and future possibilities of artificial intelligence. Biomed J Sci Tech Res. 2018;7:5937–9.
16. Marklein RA, Klinker MW, Drake KA, et al.2019. Morphological profiling using machine learning reveals emergent subpopulations of interferon-γ–stimulated mesenchymal stromal cells that predict immunosuppression. Cytotherapy. 2019;21:17–31.
17. Matsuoka F, Takeuchi I, Agata H, et al.2013. Morphology-based prediction of osteogenic differentiation potential of human mesenchymal stem cells. PLoS One. 2013;8:e55082.
18. Mahajan A, Vaidya T, Gupta A, Rane S, Gupta S.2019. Artificial intelligence in healthcare in developing nations: the beginning of a transformative journey. Cancer Res Stat Treat. 2019;2:182.
19. Mota SM, Rogers RE, Haskell AW, et al.2021. Automated mesenchymal stem cell segmentation and machine learning-based phenotype classification using morphometric and textural analysis. J Med Imaging. 2021;8.
20. Mohammad MM, Sepideh N, Roya NV, Ali AH, Hossein M. 2022. Prediction of Hepatitis disease using ensemble learning methods.J Prev Hyg.2022: 63 (3): E424-E428.
21. Naito T, Suzuki K, Hirata J, et al. 2021. A deep learning method for HLA imputation and trans-ethnic MHC fine-mapping of type 1 diabetes. Nat Commun. 2021;12:1639.
22. Narita A, Ueki M, Tamiya G. 2021. Artificial intelligence powered statistical genetics in biobanks. J Hum Genet. 2021;66:61–65.
23. Sakurai R, Ueki M, Makino S, et al. 2019. Outlier detection for questionnaire data in biobanks. Int J Epidemiol. 2019;48:1305–1315.
24. Shouval R, Bondi O, Mishan H, et al. 2014. Application of machine learning algorithms for clinical predictive modeling: a data-mining approach in SCT. Bone Marrow Transplant. 2014;49:332–337.
25. Yan Q, Jiang Y, Huang H, et al. 2021. Genome-wide association studies-based machine learning for prediction of age-related macular degeneration risk. Transl Vis Sci Technol. 2021;10:29.
26. Zhou J, Theesfeld CL, Yao K, et al.2018. Deep learning sequence-based ab initio prediction of variant effects on expression and disease risk. Nat Genet. 2018;50:1171–1179.

Source: Biobanking

Link: https://www.biobanking.com/future-perspectives-implementation-of-artificial-intelligence-in-umbilical-cord-biobanking/