The ethical considerations including inclusion and biases, data protection, and proper implementation among AI in radiology and potential implications

Abstract

Artificial intelligence (AI) has tremendous potential to improve diagnostics, efficiency, and consistency in radiology, but implementation of generalized regulations is crucial to prevent violation of ethical principles in patient privacy, data management, and diagnostic accuracy. Key terms searched on PubMed included “Artificial Intelligence or Machine Learning”, “Ethics” and “COVID-19 or Coronavirus” and “Radiology or Imaging”. Twenty Articles published between June 2016 and November 2020 were included for synthesis, after exclusion of 53 of the non-duplicate studies following screening for original information and relevance to clinical AI imaging and ethics. Clear themes emerged, allowing for these four key recommendations: 1. Implementation of protocols for providers to explain how patient data will be used in treatment and research. 2. Differences in data regulations among industry partners must be considered for successful implementation of AI in radiology, and protections against entities with a financial interest in patient data must be included. 3. Providers must understand AI algorithm methodology to explain care plans to patients, obtain informed consent, explain potential risks, and avoid propagating current disparities in care through AI. 4. Radiologists must work with AI developers to share their expertise. Our goal is to introduce ethical considerations in AI imaging without requiring a background in mathematics so that they may not fall for potential pitfalls that may be encountered as AI progresses into clinical settings. We conclude in recommending increased focus on the use of existing AI platforms or collaborations with existing industries rather than developing new algorithms. Future challenges involve addressing implicit bias and managing data privacy in industry.

Ähnliche Arbeiten