With the widespread application of artificial intelligence (AI) in healthcare, algorithmic bias has become a significant issue affecting medical fairness. Health data, as the foundation for training AI models, can reduce accuracy and exacerbate health inequalities if biased. When health data fails to represent all social groups, including minorities, the elderly, and people with disabilities, AI models trained on such data may produce unfair decisions. To address this issue, the STANDING Together consensus has proposed a series of recommendations aimed at enhancing the transparency of health data and actively assessing the fairness of datasets. By analyzing relevant literature, this paper explores the issue of bias in health data and discusses the potential impacts of the STANDING Together consensus recommendations. The paper argues that improving the diversity and transparency of health data, as well as proactively evaluating dataset fairness, will help eliminate algorithmic bias and promote fairness in healthcare, thereby providing more equitable medical services to all groups.