Achieve Fairness in AI-Assisted Mobile Healthcare Apps through Unsupervised Federated Learning

通过无监督联合学习实现人工智能辅助移动医疗应用程序的公平性

• 批准号: 10504193
• 负责人: Jingtong Hu
• 金额: $ 47.98万
• 依托单位: UNIVERSITY OF PITTSBURGH AT PITTSBURGH
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-08-15 至 2026-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10504193
• 关键词: Address; Age; Algorithms; Architecture; Artificial Intelligence; Asian; Awareness; Case Study; Cellular Phone; Code; Communities; Computer software; Data; Data Collection; Data Set; Data Sources; Databases; Dermatology; Detection; Devices; Diagnosis; Emotions; Ensure; Exhibits; Face; Gender; Goals; Healthcare; Heterogeneity; Hispanic; Human; Image; Inequality; Interdisciplinary Study; Label; Lead; Learning; Life; Light; Machine Learning; Medical; Minority; Minority Groups; Modeling; Monitor; Outcome; Patients; Performance; Persons; Population; Population Heterogeneity; Race; Reporting; Research; Safety; Skin; Socioeconomic Status; Supervision; Techniques; Underrepresented Populations; Visual; Woman; artificial intelligence algorithm; base; deep learning model; federated learning; handheld mobile device; health care disparity; improved; innovation; learning progression; machine learning framework; malignant neoplasm of eye; marginalized community; men; mobile application; neural network; new technology; privacy preservation; racial disparity; relating to nervous system; sex; skin color; socioeconomic disparity; socioeconomics

Deep learning models have been deployed in an increasing number of edge and mobile devices to provide healthcare in our life, from mobile dermatology assistant, mobile eye cancer (leukoria) detection, emotion detection, to comprehensive vital signs monitoring. All these techniques rely on visual assistance of the cameras that come with mobile devices and inevitably lead to different levels of fairness concerns, due to the inherent gender, race and/or socioeconomic bias in existing AI models. Compounding contributing factors include a lack of medical professionals from marginalized communities, inadequate information about those communities, and socioeconomic barriers to participating in data collection and research. In the absence of a diverse population that reflects that of the U.S. population, potential safety or efficacy considerations could be missed. What is worse, with inadequate data, AI algorithms could misdiagnose underrepresented people, leading to increasing health care disparities. Therefore, there is a critical need to address racial, skin color, and socioeconomic inequities in AI-assisted mobile diagnosis. This project will address the fairness issue in mobile AI assistants, using dermatology diagnosis and skin color inequity as the study case. Instead of collecting equitable demographic dataset in a centralized way, it will develop a federated on-device learning framework for participation inclusion, selective data contribution, and continuous personalization. The framework can continuously learn from new users’ data as they use the mobile apps with little human supervision. An unsupervised federated learning (FL) framework will be developed with heterogeneous hardware (high-end and low-end) and models such that users from all socioeconomic status can participate in the research. While various FL techniques have been developed, how to implement unsupervised FL with both hardware and model heterogeneity is not clear. It is also essential to achieve this goal with as little human supervision as possible since it is impractical to have a doctor constantly label the images when users are using these AI-based apps. In addition, even with FL, data from predominating population will still dominate the data collected. Non-uniform data selection techniques will be developed to automatically weigh the importance of different data for maximum fairness. Finally, not all neural networks exhibit the same inherent fairness even with the same biased data. A fairness-aware neural architecture search framework will be developed to find the networks that can achieve the most fairness. The expected outcome of this project is a holistic framework to mitigate the impacts of inequity by improving the inference performance for minorities. The developed techniques will be implemented as mobile apps with heterogeneous smart phones and evaluated with both public dataset and patients at UPMC. Data and code will be made available for public research. The developed techniques can be easily extended to all AI-assisted diagnosis and account for the inequity in various aspects such as age, sex, racial, etc.

深度学习模型已经部署在越来越多的边缘和移动的设备中，以提供 我们生活中医疗保健，从移动的皮肤科助理，移动的眼癌（白内障）检测，情感 检测，到全面的生命体征监测。所有这些技术都依赖于 由于移动的设备附带的摄像头，不可避免地会导致不同程度的公平性问题 现有人工智能模型中固有的性别、种族和/或社会经济偏见。复合促成因素 包括缺乏来自边缘化社区医疗专业人员，有关这些社区的信息不足， 参与数据收集和研究的社会经济障碍。在没有 反映美国人群的多样化人群，潜在的安全性或有效性考虑可能是 错过了.更糟糕的是，由于数据不足，人工智能算法可能会误诊代表性不足的人， 导致医疗保健差距扩大。因此，迫切需要解决种族、肤色和 AI辅助移动的诊断中的社会经济不平等。 该项目将解决移动的AI助手的公平性问题，使用皮肤科诊断和皮肤 颜色不平等作为研究案例。它不是以集中的方式收集公平的人口数据集，而是将 开发一个联合的设备上学习框架，用于参与包容、选择性数据贡献，以及 持续的个性化。框架可以在新用户使用 移动的应用程序几乎没有人监督。一个无监督的联邦学习（FL）框架将在 使用异构硬件（高端和低端）和模型开发， 社会经济地位可以参与研究。虽然已经开发了各种FL技术，但如何 如何实现具有硬件和模型异质性的无监督FL尚不清楚。还必须 在尽可能少的人力监督下实现这一目标，因为不断地有医生是不切实际的 在用户使用这些基于AI的应用程序时标记图像。此外，即使使用FL， 占主导地位的人口仍将主导收集的数据。非均匀数据选择技术将 开发自动权衡不同数据的重要性，以实现最大的公平性。最后，并非所有神经 网络即使在具有相同的偏置数据的情况下也表现出相同的固有公平性。一个公平意识神经系统 架构搜索框架将被开发，以找到网络，可以实现最公平的。 这一项目的预期成果是一个减轻不平等影响的整体框架， 提高少数民族的推理能力。所开发的技术将作为移动的 应用程序与异构智能手机，并与公共数据集和患者在UPMC评估。数据 代码将被公开供公众研究。开发的技术可以很容易地扩展到所有 人工智能辅助诊断，并解释了年龄、性别、种族等各方面的不公平。