黑料吃瓜网

Interoperability And Security: Navigating Ai Challenges In Modern Healthcare

Aaron Anthony Anderson | Alexander Dean Garcia | Jose N. Cruz | John Arnold Valentino | Nathan Esperancilla | Warren Motes | Jericho C. Guzman | Emmanuel Jude Salvador | Mark Vincent Dimatulac | Cris B. Yanguas Jr. | Christensen Cervantes | Carl Bugarin

Discipline: medical sciences (non-specific)

 

Abstract:

AI is considered to be very potent in the subject of healthcare and is gradually growing in popularity. But there are also many problems linked to the application of such solutions as AI in healthcare, for example, questions of the interaction between different systems or data protection. The purpose of this study is to report on the concerns and opportunities regarding AI in terms of the obstacles to, and implementation of successfully delivering healthcare in today’s society with significant attention being paid to the issue of data safety and security. It describes the sources of bias and considers literature describing different forms of algorithmic/AI bias in education and in the groups that are underrepresented in the development of EdTech software. To address this issue, in this paper, we put forward a Phase, Guarantee, and Utility (PGU) triad-based model to facilitate the evaluation of various PPML solutions in terms of their decomposed privacy-preserving functionalities. In this study, we reviewed available research on security, privacy, and defense mechanisms and policies to enhance the trustworthiness of ML. The study reveals various risks including the reluctance to adopt novel technologies by the care providers in the hospitals, the concern of patients’ confidentiality, technicalities as faced by the IT employees, limitation of access to broad health care data by the researchers, and the concern of the schools to update the existing material. Based on this, the advocacy of this study is on the academic issues and issues of AI in learning incorporated as AIED, the first entry points for the data center nodes; possible offerings that can enrich scholars’ learning and potentiality for data-driven learning, and future opportunities. This study’s surveys’ outcomes reveal that there was a demand for the guidelines’ development to enable the introduction of AI technologies across the units of the health systems and the necessity of possessing robust privacy provisions to safeguard patients’ information.