.webp)
Artificial Intelligence (AI) is here to stay, but we decided how to receive it—and how to implement it. For smart solution providers like Eden, it's crucial to think about how radiologists interact with these tools in their workspaces. This can make the difference between an AI that works with the doctor, and one that works against them.
The field of radiology employs a variety of intelligent tools for image analysis and the generation of diagnostic impressions. These programs are capable of analyzing large amounts of information in seconds, reducing the doctor's workload, expediting emergency diagnosis, and can even provide findings previously invisible to the radiologist.
But like any other technology, AI is fallible. According to a 2023 study, based on diagnoses of knee osteoarthritis, the error rate of an AI is generally equivalent to that of a human radiologist. However, depending on the interaction with the radiologist, these incidences can increase significantly and lead to the wrong decisions. Risk factors for AI medical error include:
- Lack of knowledge about how the algorithm works (also called the “black box” effect)
- Absence of ROI or other visual indicators from the AI to indicate its findings in the study image
- Automatic insertion of the findings made by AI in the study report
These results point to an implementation of AI in which the radiologist's agency is not prioritized. Raymundo González, CTO at Eden, believes that this should be a crucial consideration for any intelligent tool in the sector: “everything must be designed to give the doctor complete control”.
Raymundo is the leader in creating Eden Creator, a language intelligence model that provides accurate diagnostic conclusions based on the findings of a study. The purpose of this tool is not to replace the doctor's interpretation, but to complement their written report, a task that “does not always require the skill and knowledge of the radiologist”.
Creator is designed to empower the user, who has the ability to observe the suggested conclusions, edit them, accept them or simply reject them. With each instance, the algorithm learns from these expert decisions to fine-tune its most accurate predictions. This avoids the risk of a wrong conclusion due to the use of AI. In addition, this tool can save the radiologist up to an hour of work, reducing fatigue and eye strain that can also lead to medical error.
As intelligent tools become increasingly accessible globally, it is essential to explore forms of implementation that consider the radiologist's working conditions and mitigate the risk of medical error. As established by the American College of Radiology (ACR), “AI is never “an alternative” to the radiologist”, but rather an ally at their service.
.jpg)
