Page 44 - AIH-1-1

Artificial Intelligence in Health Digital and AI strategies for pharmaceutical industry

direct customer models and integrator channels. Logistics
enterprises and e-commerce platforms also study the
application of AI technology in various aspects of logistics
by establishing R&D teams and technology subsidiaries,
and there is a relationship of cooperation and potential
competition among the three .
[26]
4.3. Technology, cost, and benefits
AI has great application potential in the transportation
and distribution field of pharmaceutical logistics, but due
to insufficient technological stability and mismatched costs
and benefits, unmanned pharmaceutical delivery is still in
its infancy. Although major domestic R&D platforms use
the order sorting system for immediate medical delivery,
AI software and hardware suppliers are still unable to make
a profit from this technology .
[27]
4.4. Warehousing
The application of pharmaceutical warehousing AMR
(applicable margin reset) is gradually gaining traction,
Figure 4. Flowchart of passing collective knowledge in Industry 4.0. and its market development prospects are huge. In 2019,
the market size of domestic pharmaceutical warehousing
Therefore, AI technology needs more data support in AMR was US$ 32 million. In the coming years, the
medical research. market size of pharmaceutical AMR will rapidly expand.
By 2025, the market size of domestic warehousing
4.1. Self-improvement ability AMR is expected to exceed US$ 300 million. Although
As a form of intelligence reactive only based on programs, pharmaceutical warehousing AMR has advantages such
AI can only judge the input conditions based on their as flexible deployment and independent flexibility, the
agenda and is not capable of independent thinking. AI is technical requirements for AMR products are relatively
based on a large amount of data, and the quality of the data high, and there are relatively few enterprises in Pakistan
will directly affect the judgment results of AI . If doctors that can achieve mass production and promote project
[25]
cannot correctly judge whether patients have suffered from implementation. Nevertheless, the AMR market is still in
eye diseases, the distortion of big data will lead to errors in its infancy and a long-cycle market validation is required .
[28]
the spontaneous judgment of products. Hence, in product
development, it is critical to first screen the quality of the 4.5. Hospitals and pharmacies
data to avoid database errors that would affect the entire We should intensely utilize 5G technology to enhance
product. There is a gradual trend in the adoption of cloud video resolution, increase research efforts on intelligent
platforms and cloud computing data management methods recognition, and minimize the impact of objective
in AI products. At present, the storage and analysis of data conditions on AI as much as possible. Big data platform
are outside the monitoring scope of hospitals. Ensuring data can be adopted to increase data openness, and AI can be
security and avoiding illegal use of data are vital issues that trained to conduct comprehensive analysis and application
AI enterprises need to focus on in future product design. of data in an effective and reasonable manner. AI can be
trained in different scenarios to enhance its autonomy,
4.2. Mature industry enhance medical cognition through core technologies,
The most significant difference between the “AI + optimize AI algorithms, and reduce the time that it
medicine” logistics and the traditional logistics industry takes to continuously explore potential associations with
chains is that the upstream and downstream relationships diseases. A good cloud database is crucial, as scattered
are unclear or that the “AI + logistics” industry chain is data are challenging to work with. The cloud and
not yet mature. AI companies, logistics enterprises, and database – whether from a public cloud or a private cloud –
e-commerce platforms are essential in the industry chain. help hospitals standardize these data. This way, even if the
AI companies provide downstream customers with “AI + data units are dispersed, they can still be used as composite
Industries” related products and technical services through data when needed .
[29]

Volume 1 Issue 1 (2024) 38 https://doi.org/10.36922/aih.1486