SCH: Enabling Data Outsourcing and Sharing for AI-powered Parkinson's Research

SCH：为人工智能驱动的帕金森病研究提供数据外包和共享

• 批准号: 10480884
• 负责人: Shigang Chen
• 金额: $ 29.43万
• 依托单位: UNIVERSITY OF FLORIDA
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-03 至 2025-05-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10480884
• 关键词: Artificial Intelligence; Biomedical Computing; Biomedical Research; Classification; Cloud Computing; Complex; Consumption; Data; Data Set; Diagnosis; Disease; Foundations; Image; Institutes; Joint Prosthesis; Laws; Lead; Machine Learning; Masks; Mathematics; Medical; Methods; Modeling; Modernization; Neural Network Simulation; Non-linear Models; Outcome; Outsourcing; Parkinson Disease; Parkinsonian Disorders; Patient Data Privacy; Patients; Performance; Privacy; Public Health Informatics; Quality of life; Randomized; Regulation; Research; Risk; Series; Source; Technology; Theoretical Studies; Time; Training; University Hospitals; Work; accurate diagnosis; artificial neural network; base; biomedical informatics; cloud based; cloud storage; cost; data privacy; data sharing; deep learning; deep neural network; digital; distributed data; encryption; experimental study; imaging genetics; improved; individualized medicine; learning network; mobile computing; novel; operation; privacy preservation; privacy protection; theories

Artificial intelligence holds the promise of transforming data-driven biomedical research and computational health informatics for more accurate diagnosis and better treatment at lower cost. In the meantime, modern digital and mobile technologies make it much easier to collect information from patients in large scale. While “big” medical data offers unprecedented opportunities of building deep-learning artificial neural network (ANN) models to advance the research of complex diseases such as Parkinson’s disease (PD), it also presents unique challenges to patient data privacy. The task of training and continuously refining ANN models with data from tens of thousands of patients, each with numerous attributes and images, is computation-intensive and time-consuming. Outsourcing such computation and its data to the cloud is a viable solution. However, the problem of performing the ANN learning operations in the cloud, without the risk of leaking any patient data from their distributed sources, remains open to date. This application proposes to develop novel data masking technologies based on randomized orthogonal transformation to enable AI-computation outsourcing and data sharing, with the following two specific aims: 1) Perform two experimental studies of training ANN models with data masking in the HiperGator cloud for PD prediction and Parkinsonism diagnosis; 2) establish the theoretical foundation on data privacy, inference accuracy, and training performance of the ANN models used in the experimental studies. The interdisciplinary project team combines the expertise from data privacy, biomedical informatics, machine learning, and cloud computing to develop data outsourcing and sharing technologies for AI-powered PD research. The proposed research will remove a major roadblock that restricts medical data accessibility and hinders cloud-based operations of deep-learning artificial neural networks for biomedical research. The outcome is expected to have a broader impact beyond PD research in advancing the theory and implementation of cloud-based medical studies with data privacy protection.

人工智能有望改变数据驱动的生物医学研究和计算健康信息学，以更低的成本实现更准确的诊断和更好的治疗。与此同时，现代数字和移动的技术使大规模收集患者信息变得更加容易。虽然“大”医疗数据为构建深度学习人工神经网络（ANN）模型以推进帕金森病（PD）等复杂疾病的研究提供了前所未有的机会，但它也对患者数据隐私提出了独特的挑战。使用来自数万名患者的数据（每个患者都有许多属性和图像）训练和不断改进ANN模型的任务是计算密集型和耗时的。将这种计算及其数据外包到云是一个可行的解决方案。然而，在云中执行ANN学习操作的问题，而没有从其分布式源泄漏任何患者数据的风险，迄今为止仍然是开放的。本申请提出开发基于随机正交变换的新型数据掩蔽技术，以实现AI计算外包和数据共享，具体目标有以下两个：1）在HiperGator云中使用数据掩蔽训练ANN模型进行两项实验研究，用于PD预测和帕金森病诊断; 2）为实验研究中使用的ANN模型的数据隐私性、推理精度和训练性能奠定理论基础。跨学科项目团队结合了数据隐私，生物医学信息学，机器学习和云计算的专业知识，为AI驱动的PD研究开发数据外包和共享技术。这项拟议中的研究将消除限制医疗数据可访问性和阻碍生物医学研究深度学习人工神经网络基于云的操作的主要障碍。该成果预计将在PD研究之外产生更广泛的影响，推动基于云的医学研究的理论和实施，并保护数据隐私。