Yang Michael, El-Attar Abd-Allah, Chaspari Theodora
Computer Science & Engineering, Texas A&M University, College Station, TX, United States.
Computer Science & Engineering, Texas A&M University Qatar, Al Rayyan, Qatar.
Front Digit Health. 2024 Jul 25;6:1351637. doi: 10.3389/fdgth.2024.1351637. eCollection 2024.
Machine learning (ML) algorithms have been heralded as promising solutions to the realization of assistive systems in digital healthcare, due to their ability to detect fine-grain patterns that are not easily perceived by humans. Yet, ML algorithms have also been critiqued for treating individuals differently based on their demography, thus propagating existing disparities. This paper explores gender and race bias in speech-based ML algorithms that detect behavioral and mental health outcomes.
This paper examines potential sources of bias in the data used to train the ML, encompassing acoustic features extracted from speech signals and associated labels, as well as in the ML decisions. The paper further examines approaches to reduce existing bias via using the features that are the least informative of one's demographic information as the ML input, and transforming the feature space in an adversarial manner to diminish the evidence of the demographic information while retaining information about the focal behavioral and mental health state.
Results are presented in two domains, the first pertaining to gender and race bias when estimating levels of anxiety, and the second pertaining to gender bias in detecting depression. Findings indicate the presence of statistically significant differences in both acoustic features and labels among demographic groups, as well as differential ML performance among groups. The statistically significant differences present in the label space are partially preserved in the ML decisions. Although variations in ML performance across demographic groups were noted, results are mixed regarding the models' ability to accurately estimate healthcare outcomes for the sensitive groups.
These findings underscore the necessity for careful and thoughtful design in developing ML models that are capable of maintaining crucial aspects of the data and perform effectively across all populations in digital healthcare applications.
机器学习（ML）算法被誉为实现数字医疗辅助系统的有前景的解决方案，因为它们能够检测人类不易察觉的细粒度模式。然而，ML算法也因根据人口统计学特征区别对待个体而受到批评，从而加剧了现有的差异。本文探讨了用于检测行为和心理健康结果的基于语音的ML算法中的性别和种族偏见。
本文研究了用于训练ML的数据中潜在的偏见来源，包括从语音信号中提取的声学特征和相关标签，以及ML决策中的偏见来源。本文还研究了通过使用对个人人口统计信息最缺乏信息量的特征作为ML输入，并以对抗方式变换特征空间来减少现有偏见的方法，以减少人口统计信息的证据，同时保留有关重点行为和心理健康状态的信息。
结果在两个领域呈现，第一个领域涉及估计焦虑水平时的性别和种族偏见，第二个领域涉及检测抑郁症时的性别偏见。研究结果表明，不同人口群体在声学特征和标签方面存在统计学上的显著差异，并且群体之间的ML性能也存在差异。标签空间中存在的统计学显著差异在ML决策中部分保留。尽管注意到不同人口群体的ML性能存在差异，但关于模型准确估计敏感群体医疗结果的能力，结果好坏参半。
这些发现强调了在开发ML模型时进行谨慎和深思熟虑设计的必要性，这些模型能够保留数据的关键方面，并在数字医疗应用中对所有人群有效发挥作用。
