Exploring AI Bias in Key Healthcare: The Role of Data Science
TL;DR
The study highlights the need of data sciences for addressing AI-related biases in healthcare to ensure the successful and ethical implementation of AI technologies and develops strategies to mitigate these biases.
Abstract
Artificial Intelligence (AI) is increasingly integrated into healthcare to enhance decision-making; improve patient outcomes; and optimize healthcare operations. However, the adoption of AI in this sector presents challenges, particularly regarding ethical considerations, data privacy, and bias. The purpose of this study is to explore the types of AI bias reported in healthcare literature from 2018 to 2024 and to highlight the implications for AI implementation in medical practice. the paper discusses Data sciences strategies to mitigate these biases, highlighting the critical role for responsible and ethical AI development. It roles in role in mitigating and enhancing the fairness in ML and LLM. It then delves into the process of fine-tuning, outlining the steps taken to adapt the model to specific datasets and tasks. This study is a systematic review of papers published between 2018 and 2024. The literature was analyzed to identify various types of AI bias; including algorithmic bias; data bias; and ethical-human-related bias. Relevant articles were selected based on their contribution to understanding AI bias in healthcare and its effects on healthcare delivery. The analysis identified several types of AI bias prevalent in healthcare. Algorithmic bias; data bias; and ethical-human-related bias were the primary categories. These biases can distort the decision-making process; impact patient outcomes; and challenge the fundamental principles of medical ethics. The review also found that human and data biases are significant factors contributing to AI bias in healthcare systems. The study highlights the need of data sciences for addressing AI-related biases in healthcare to ensure the successful and ethical implementation of AI technologies. Special attention is given to avoiding overfitting and optimizing hyperparameters during the fine-tuning process. Future research should focus on developing strategies to mitigate these biases; ensuring that AI systems are fair; reliable; and trustworthy. Identifying and addressing these biases is essential for the ethical integration of AI in healthcare and for protecting the integrity of the medical profession.