From Hype to Action: AI series: organization's expectations from AI solutions

Once we have a comprehensive understanding of the organization's current situation, including its structure and level of digitalization, we can begin planning for the future. However, before we invest in any AI solutions, we must ensure that they will address the organization's challenges and opportunities while meeting its goals.

While most technology companies today offer their own AI platforms or modules as part of their solutions, we need to determine whether they are the right fit for our organization. To do this, we must first assess our expectations and how we intend to use the AI tool. Once we have a clear understanding of our needs, we can explore our options more efficiently.

For example, if we manage an internal department in a hospital, we might want to use AI to predict patient deterioration three days before visible symptoms occur. The benefits of early detection are clear: patients can receive treatment sooner, potentially saving lives and reducing healthcare costs. However, we must also consider the resources required to make this a reality. We will need to invest in the solution, implement it, and provide ongoing professional and management attention. Additionally, we must verify that we have all the necessary facilities and data available to use this solution effectively.

To implement a real-time or near-real-time AI solution like this, we must ensure that we are already measuring all the necessary data. This may involve using sensors, medical devices, or other methods to collect information, as well as centralizing data storage and developing interfaces for secure data transmission. Furthermore, we must consider security and privacy concerns when handling sensitive patient data.

Once we have a full understanding of the infrastructure and the effort that is needed, we can evaluate the AI solution itself. We should carefully review the data used to train the AI model, taking into account any potential bias that may influence the results. We should also ask if there are any rule-based components in the solution and whether they align with our clinicians' guidelines.

By asking the right questions and having a clear understanding of what we can expect from an AI solution, we can choose the best option for our organization with maximum benefits.