Artificial intelligence in transfusion medicine and its impact on the quality concept

摘要

In the blood supply and clinical consumption there are currently two main directions for the future of manufacture or procurement, and patient care. One is determined by the state of the art in the advanced world with by and large strong and stable infrastructures, where systems have developed and are in the process of refining and strengthening the customer relationships, quality of care and related marketing principles. These include social marketing to create a solid and sustained access to the unique source material - human blood and blood components, and clinical advocacy and marketing to advance the quality and specificity of stochastic dynamic programming of transfusion practices at the bedside. To enable this strategy to be implemented, the process of manufacturing the source material in new and specific cellular blood products is certainly a challenge. However, the infrastructure on which the system is built needs to be solid and complete, and to progress in its structural elements might need introduction of artificial intelligence and stochastic dynamic programming. The aim is to compute a policy and strategy prescribing how to act and perform optimally in the face of uncertainty. The second direction is the acceleration of synergised efforts to bridge the gaps existing to various degrees in the large developing part of the world. This involves the infrastructure and governance at the national level to economy of scale, leadership, competence of professionals, education and appropriate environmental conditions at the local and regional level. Although in principle the same developmental stages have to be followed, the pace of development could be accelerated to allow a faster narrowing and shallowing of the distinguished gaps [[1]Smit Sibinga C.Th Abdella Y.E. Seghatchian J. Poor economics – transforming challenges in transfusion medicine and science into opportunities.Transfus Apher Sci. 2020; https://doi.org/10.1016/j.transci.2020.102752Abstract Full Text Full Text PDF Scopus (1) Google Scholar,[2]Action framework to advance universal access to safe, effective and quality-assured blood products, 2020-2023. World Health Organization, Geneva2020Google Scholar]. Here artificial intelligence and in particular stochastic dynamic programming seem far in this instance but might turn out to be a blessing in disguise! A wider introduction and application of stochastic dynamic programming [[3]Bellman R. Dynamic programming. 1957 Princeton University Press, Princeton, USA2003Google Scholar] and artificial intelligence will play a major role through the creation of reliable quality data bases (big data), and the use of these ‘big data’ bases through a network of algorithms, deep learning and machine learning processes. This would allow better management of the vein-to-vein manufacture and clinical application of blood and blood components, based on stochastic dynamic programming, computer simulation and algorithmic consumption and production predictions that would allow an optimal use of source material, equipment, consumables, and above all, professional staff [[4]Haijema R. van Dijk N. van der Wal J. Smit Sibinga C.Th. Blood platelet production with breaks: optimization by SDP and simulation.Int J Prod Econ. 2009; 121: 464-473Crossref Scopus (78) Google Scholar,[5]van Dijk N.M. Haijema R. van der Wal J. Smit Sibinga C.Th. Blood platelet production. A novel approach for practical optimization.Transfusion. 2009; 49: 411-420Crossref PubMed Scopus (88) Google Scholar]. More advanced and sophisticated robotics will be introduced to standardise a growing number of production processes and procedures, controlled by an advanced system of communication through information technology including automated alert systems to detect early trends in deviation from the standards. As a consequence waterproof documentation is paramount. The human work force has to be educated in such environment and quality culture to be able to handle and act