The Ethics of AI

Artificial Intelligence (AI) is undoubtedly beneficial. It has improved patient outcomes and made radiology more efficient. However, some great challenges remain and researchers believe that AI cannot be trusted completely.

Global imaging societies and their representatives are collaborating on a   live document outlining a clear set of AI ethics to be used in radiology. In order to prove the point of ethical guidance about AI, major radiology organizations from America, Canada, and Europe have jointly drafted the Ethics of AI in Radiology for the community.

The American College of Radiology published the consensus draft document, which was co-authored by the American Association of Physicists in Medicine (AAPM), American College of Radiology (ACR), Canadian Association of Radiologists (CAR), Radiological Society of North America (RSNA), European Society of Radiology (ESR), European Society of Medical Imaging Informatics (EuSoMII), and Society for Imaging Informatics in Medicine (SIIM).

Key Areas of AI Ethics

The statement released by the team breaks the code of AI ethics into three areas:

  • Ethics of data, which includes informed consent, ownership, data privacy, patient data of transactions, etc along with social and technical issues on the bias
  • Ethics of programs, algorithms, and considerations for verifying the moral use and safety
  • Ethics of practice, which includes practice-level policies to do all everything necessary for patients and minimize chances of inequalities related to potential gain and resources

Geraldine McGinty, the chair of the ACR Board of Chancellors, published a blog post on March 1, 2019, saying, “Importantly, the group included trainees, patients and other stakeholders such as an ethicist from MIT. However, despite the wide-ranging backgrounds and expert input that created this draft, the writing group, and our societies’ leaders are very clear that this is just that: a draft. That’s why, having kept the writing group fairly small to facilitate an accelerated turnaround, it’s now critical to get extensive comments from the broader imaging and healthcare ecosystem.”

The Document Details

The consensus document is a 38-page long draft that was released on March 5. The draft was broken into three sections, as mentioned above, detailing the ethics of AI. Overall, 18 authors address privacy, data sharing, algorithm and automation biases, workforce disruption, ideas of machine medical decisions, and workforce disruption.

The authors of the document also wrote, “AI has noticeably altered our perception of radiology data – their value, how to use them and how they may be misused. Rather than simply understanding AI, radiologists have a moral duty both to understand their data and to use the data they collect to improve the common good, extract more information about patients and their diseases and improve the practice of radiology.”

In addition, the authors also expressed that it is critical for radiologists to start developing these AI codes of ethics and practice them. As the radiology community is trying to learn more about ethical AI while making efforts to invent and utilize the same, McGinty believes that the AI code of ethics shall be a living document.

sepStream® is committed to providing the best services to its patients as well as doctors and healthcare organizations. We have an extensive range of diagnostic services available to deliver a comprehensive range of solutions to our clients using the best tools.