Robust, Generalizable, and Fair Machine Learning Models for Biomedicine

稳健、可推广且公平的生物医学机器学习模型

• 批准号: 10688028
• 负责人: Kun-Hsing Yu
• 金额: $ 42.38万
• 依托单位: HARVARD MEDICAL SCHOOL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2026-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10688028
• 关键词: Address; Adverse effects; Algorithms; Biological; Computing Methodologies; Data; Data Analyses; Development; Disease; Ensure; Goals; Informatics; Knowledge; Machine Learning; Methodology; Methods; Mission; Modeling; Modernization; Molecular; Molecular Biology; National Institute of General Medical Sciences; Pathologic; Pathology; Pattern Recognition; Pharmaceutical Preparations; Pharmacology; Phenotype; Population Heterogeneity; Public Health; Research; Research Activity; Signal Transduction; Techniques; Toxic effect; Vision; advanced disease; analytical method; drug response prediction; improved; innovation; machine learning algorithm; machine learning model; molecular modeling; multidimensional data; multiple omics; novel; predictive modeling; programs; response; success

Project Summary Modern machine learning approaches have attained substantial success in pattern recognition and high-dimensional data analyses. However, these algorithms heavily rely on association discovery, which cannot elucidate the mechanisms underpinning the observed correlations and suffers from limited generalizability. To address this challenge, the Yu Lab focuses on the development of robust and generalizable machine learning approaches to integrate various types of biomedical data, including multi-omics, pathology, and phenotypic information. The goal of the next five years is to develop novel computational methods that connect machine learning algorithms with causal inference methodologies to better understand the molecular mechanisms underpinning disease pathology and enable fair and robust predictions of drug response and toxicity. The overall vision of the proposed research program is to establish generalizable data-driven methods to transform biomedical data into robust prediction and mechanistic models. The proposed approach will systematically connect diverse biomedical signals to extract previously unknown knowledge on the molecular mechanisms and derive reliable prediction models for the effects of medications. The proposed approaches are innovative because they depart from the status quo by incorporating advanced causal inference techniques with data-driven algorithms to enhance mechanistic and predictive modeling. This research program is significant because it is expected to improve our understanding of disease pathology and provide a fair and generalizable informatics framework for drug response and adverse effects prediction in diverse populations. The proposed research activities will open new research horizons by establishing a new machine learning platform for generating reliable predictions, which will vertically advance molecular biology, pharmacology, and computational research in biomedicine.

项目摘要 现代机器学习方法在以下方面取得了巨大成功： 模式识别和高维数据分析。然而，这些算法 严重依赖关联发现，不能阐明机制 支持所观察到的相关性，并具有有限的普遍性。到 为了应对这一挑战，Yu Lab专注于开发强大的， 可推广的机器学习方法来整合各种类型的生物医学 数据，包括多组学、病理学和表型信息。下一个目标 五年的时间是开发新的计算方法， 算法与因果推理方法，以更好地了解分子 支持疾病病理学的机制，并能够公平和可靠地预测 药物反应和毒性。拟议研究计划的总体愿景是 建立可推广的数据驱动方法，将生物医学数据转换为 预测和机械模型。拟议的方法将系统地连接 不同的生物医学信号，以提取以前未知的知识的分子 机制，并得出可靠的预测模型的药物的影响。的 所提出的方法是创新的，因为它们通过以下方式脱离了现状： 将先进的因果推理技术与数据驱动算法相结合， 加强机械和预测建模。这项研究计划意义重大 因为它有望提高我们对疾病病理学的理解， 药物反应和不良反应的公平和可推广的信息学框架 在不同的人群中进行预测。研究活动将开启新的 通过建立一个新的机器学习平台， 可靠的预测，这将垂直推进分子生物学，药理学， 生物医学中的计算研究

项目成果

A survival guide for interdisciplinary PhD students.

跨学科博士生的生存指南。

• DOI: 10.1038/nbt.3671
• 发表时间: 2016
• 期刊: Nature biotechnology
• 影响因子: 46.9
• 作者: Yu,Kun-Hsing

Survival Prediction After Neurosurgical Resection of Brain Metastases: A Machine Learning Approach.

神经外科切除脑转移瘤后的生存预测：机器学习方法。

• DOI: 10.1227/neu.0000000000002037
• 发表时间: 2022
• 期刊: Neurosurgery
• 影响因子: 4.8
• 作者: Hulsbergen,AlexanderFC;Lo,YuTung;Awakimjan,Ilia;Kavouridis,VasileiosK;Phillips,JohnG;Smith,TimothyR;Verhoeff,JoostJC;Yu,Kun-Hsing;Broekman,MarikeLD;Arnaout,Omar
• 通讯作者: Arnaout,Omar

Prioritization of cancer marker candidates based on the immunohistochemistry staining images deposited in the human protein atlas.

• DOI: 10.1371/journal.pone.0081079
• 发表时间: 2013
• 期刊: PloS one
• 影响因子: 3.7
• 作者: Chiang SC;Han CL;Yu KH;Chen YJ;Wu KP
• 通讯作者: Wu KP

Real-world data analyses unveiled the immune-related adverse effects of immune checkpoint inhibitors across cancer types.

• DOI: 10.1038/s41698-021-00223-x
• 发表时间: 2021-09-10
• 期刊: NPJ precision oncology
• 影响因子: 7.9
• 作者: Wang F;Yang S;Palmer N;Fox K;Kohane IS;Liao KP;Yu KH;Kou SC
• 通讯作者: Kou SC

Epidemiology and risk factors for the development of cutaneous toxicities in patients treated with immune-checkpoint inhibitors: A United States population-level analysis.

• DOI: 10.1016/j.jaad.2021.03.094
• 发表时间: 2022-03
• 期刊: JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY
• 影响因子: 13.8
• 作者: Wongvibulsin, Shannon;Pahalyants, Vartan;Kalinich, Mark;Murphy, William;Yu, Kun-Hsing;Wang, Feicheng;Chen, Steven T.;Reynolds, Kerry;Kwatra, Shawn G.;Semenov, Yevgeniy R.
• 通讯作者: Semenov, Yevgeniy R.