The last decade has brought the greatest advances in Artificial Intelligence, thanks to which new models of diagnosis, prevention and treatment of diseases have been developed.
Researchers in the United States conducted a national survey of the population's perception of the use of Artificial Intelligence applied in health. The results were presented in a article published in JAMA Network. The survey was conducted between December 3 and 18, 2019, and included various demographic groups. In total, 926 people were interviewed, of which 471 were women and 455 men.
The results show that most of the patients (55.4%) believe that AI could make medical care better or much better, however, 4.3% and 1-9% answered that the implementation of AI could worsen medical care. .
On the other hand, 66% of the respondents considered that AI played a very important role in their diagnosis or treatment, however, AI generates a certain feeling of discomfort in the vast majority of respondents, since 31% said they felt very uncomfortable when receiving a diagnosis through an AI algorithm that was accurate while another 40% said he felt somewhat uncomfortable.
Thus, despite the effectiveness of AI models, their use in diagnosis can make patients uncomfortable. Also, this is a general feeling among various ages and ethnicities. However, the answers depended on the type of tools and clinical applications of AI, for example, in reading chest X-rays by AI, 42,7% responded that they felt very comfortable with it. On the contrary, the AI applied in the diagnosis of cancer generates discomfort, according to the participants, since only the 6% felt very comfortable.
Regarding the unintended consequences of adopting and applying AI tools in clinical care, respondents expressed concern about misdiagnosis, privacy violations, less time with specialists, and higher health care costs.
“Physicians, policymakers and developers need to know what patients are saying about AI. Patients can benefit from education about how AI is being incorporated into care and the extent to which clinicians rely on AI to aid decision-making," the authors explain.
While there are further advances in AI, people's opinions will continue to change and in the same way, patients will be able to become more familiar with these new models of care.
