Enhancing fairness in AI-enabled medical systems with the attribute neutral framework

Lianting Hu, Dantong Li, Huazhang Liu, Xuanhui Chen, Yunfei Gao, Shuai Huang, Xiaoting Peng, Xueli Zhang, Xiaohe Bai, Huan Yang, Lingcong Kong, Jiajie Tang, Peixin Lu, Chao Xiong () and Huiying Liang ()
Additional contact information
Lianting Hu: Huazhong University of Science and Technology
Dantong Li: Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences)
Huazhang Liu: Southern Medical University
Xuanhui Chen: Southern Medical University
Yunfei Gao: Southern Medical University
Shuai Huang: Southern Medical University
Xiaoting Peng: Southern Medical University
Xueli Zhang: Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences)
Xiaohe Bai: La Jolla
Huan Yang: Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences)
Lingcong Kong: Southern Medical University
Jiajie Tang: Army Medical University
Peixin Lu: Cincinnati Children’s Hospital Medical Center
Chao Xiong: Huazhong University of Science and Technology
Huiying Liang: Huazhong University of Science and Technology

Nature Communications, 2024, vol. 15, issue 1, 1-16

Abstract: Abstract Questions of unfairness and inequity pose critical challenges to the successful deployment of artificial intelligence (AI) in healthcare settings. In AI models, unequal performance across protected groups may be partially attributable to the learning of spurious or otherwise undesirable correlations between sensitive attributes and disease-related information. Here, we introduce the Attribute Neutral Framework, designed to disentangle biased attributes from disease-relevant information and subsequently neutralize them to improve representation across diverse subgroups. Within the framework, we develop the Attribute Neutralizer (AttrNzr) to generate neutralized data, for which protected attributes can no longer be easily predicted by humans or by machine learning classifiers. We then utilize these data to train the disease diagnosis model (DDM). Comparative analysis with other unfairness mitigation algorithms demonstrates that AttrNzr outperforms in reducing the unfairness of the DDM while maintaining DDM’s overall disease diagnosis performance. Furthermore, AttrNzr supports the simultaneous neutralization of multiple attributes and demonstrates utility even when applied solely during the training phase, without being used in the test phase. Moreover, instead of introducing additional constraints to the DDM, the AttrNzr directly addresses a root cause of unfairness, providing a model-independent solution. Our results with AttrNzr highlight the potential of data-centered and model-independent solutions for fairness challenges in AI-enabled medical systems.

Date: 2024
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-024-52930-1 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:15:y:2024:i:1:d:10.1038_s41467-024-52930-1

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-024-52930-1

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().