The potential of artificial intelligence in the future depends on the "consensus" of human society.

2019-12-26

The National Institutes of Health (NIH) has published a special report on the application of artificial intelligence (AI) in the field of healthcare, stating that the role of AI and machine learning will be crucial as the industry deals with massive amounts of data that have the potential to improve but also disrupt health conditions and cost priorities. However, the explosive growth in investment and development currently lacks a responsible and transparent consensus foundation, which may limit its potential. Dr. Michael Martini, MD, MS, PhD, from Vanderbilt University Medical Center, stated that this new report aims to provide comprehensive references for organizational leaders, healthcare professionals, data analysts, model developers, and those dedicated to integrating machine learning into healthcare. (He is an associate professor in the Department of Biomedical Informatics and co-editor of "Artificial Intelligence in Healthcare: Hope, Hype, Promise, Peril".) "For the healthcare industry, it is important to learn from successful experiences, challenges, and recent failures in using these tools. We begin by listing important examples in the field of medical AI, emphasizing best practices surrounding the development and implementation of AI, and providing key points for discussion to reach a consensus on how the AI community and society can address these issues," wrote Michael in an article. He emphasized that the popular image of autonomous vehicles often equates to machine learning or technology-driven systems, which is completely different in the healthcare field. In the near future, artificial intelligence should be seen as a tool in the healthcare field to support highly trained healthcare professionals in reducing their workload. In recent years, there has been significant success in the interpretation of medical images, such as radiology and retinal examinations, due to the development of deep learning and related technologies. This has stimulated the development of artificial intelligence, first with venture capital investments and then with industry giants taking notice. However, some tools have biases from the populations they were developed from or biases from selecting inappropriate targets. Data analysts and developers need to focus on increasing data access and standardization, as well as thoughtful development, so that algorithms do not favor marginalized patients. The above highlights the importance of "consensus".

RECENT AI NEWS

RECENT AI TOOLS