Explainable, Fair, Reproducible and Collaborative Surgical Artificial Intelligence: Integrating data, algorithms and clinical reasoning for surgical risk assessment (XAI-IDEALIST)

可解释、公平、可重复和协作的手术人工智能：整合数据、算法和临床推理以进行手术风险评估（XAI-IDEALIST）

• 批准号: 10681418
• 负责人: Azra Bihorac
• 金额: $ 54.2万
• 依托单位: UNIVERSITY OF FLORIDA
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-03-01 至 2026-05-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10681418
• 关键词: Acceleration; Address; Adopted; Algorithms; American; Artificial Intelligence; Artificial Intelligence platform; Behavioral; Benchmarking; Bridge to Artificial Intelligence; Clinical; Clinical Research; Clinical Trials; Cognitive; Collaborations; Complication; Computing Methodologies; Critical Care; Data; Data Pooling; Data Set; Early Intervention; Environment; Ethics; Evaluation; Florida; Foundations; Funding; Generations; Health; Hospital Costs; Hospitalization; Hospitals; Human; Informatics; Infrastructure; Institution; Intelligence; Investments; Legal patent; Machine Learning; Medical; Mind; Mission; Modeling; Operative Surgical Procedures; Patient Care; Patient-Focused Outcomes; Patients; Performance; Perioperative Care; Physicians; Physiological; Population Heterogeneity; Postoperative Complications; Prevention strategy; Privacy; Process; Productivity; Psyche structure; Public Health; Publications; Reproducibility; Research; Risk; Risk Assessment; Science; System; Technology; Testing; Time; Training; Trust; United States; United States National Institutes of Health; Universities; Validation; Work; advanced disease; clinical implementation; collaborative approach; computerized tools; data integration; data modeling; data sharing; data streams; disease diagnosis; distributed data; effectiveness evaluation; federated learning; high risk; human centered computing; improved; innovation; interoperability; machine learning algorithm; multimodal data; multimodality; novel; operation; preference; privacy preservation; programs; prospective; prospective test; risk mitigation; social health determinants; success; surgical risk; theories; tool; trustworthiness; usability

Project Summary In the United States, the average American can expect to undergo seven surgical operations during a lifetime. Each year 150,000 surgical patients die, and 1.5 million develop a complication after surgery. Progress in medical Artificial Intelligence (AI) remains halted by limited datasets and models with insufficient interpretability, transparency, fairness, and reproducibility that are difficult to implement and share across institutions. In the previous funding period, in addition to 98 publications and 3 patents, a real-time intelligent surgical risk assessment system was successfully implemented at University of Florida. The overall objective of this renewal application is to develop a new conceptual framework for “Explainable, Fair, Reproducible, and Collaborative Medical AI” to provide a foundation for clinical implementation at scale. It will leverage the OneFlorida, a large clinical consortium of 22 hospitals serving 10 million patients in Florida, the nation’s third largest state. The overall objective will be achieved by pursuing three specific aims. (1) External and prospective validation of novel interpretable, dynamic, actionable, fair and reproducible algorithmic toolkit for real-time surgical risk surveillance. (2) Developing and evaluating explainable AI platform (XAI-IDEALIST) for real-time surgical risk surveillance using human-grounded benchmarks. (3) Implementing and evaluating a federated learning approach with advanced privacy features for collaborative surgical risk model training. The approach is innovative, because it represents the first attempt to (1) build the first surgical FAIR (Findable, Accessible, Interoperable, Reproducible) AI-ready, large multicenter multimodal dataset, (2) Novel computational approaches accompanied by assessing fairness and reproducibility, (3) a multifaceted and full-stack explainable AI framework, and (4) federated learning capacity for privacy-preserving model trainingacross institutions. The proposed research is significant since it will address several key problems and critical barriers, including (1) lack of AI-ready large surgical datasets, (2) lack of interpretable, dynamic, actionable, fair and reproducible surgical risk algorithms, (2) lack of a medical AI explainability platform, and (4) lack of a systematic approach for collaborative model training and sharing across institutions. Ultimately, the results are expected to improve patient outcomes and decrease hospitalization costs, as well as lifelong complications.

项目摘要 在美国，普通美国人可以期望在一生中进行七项手术手术。 每年有15万名手术患者死亡，手术后有150万例并发症。进步 医疗人工智能（AI）仍然被有限的数据集和模型停止，解释性不足， 透明度，公平性和可重复性难以在机构中实施和共享。在 以前的资金期，除了98份出版物和3项专利外，还具有实时智能手术风险 评估系统已在佛罗里达大学成功实施。总体目标 更新应用是为“可解释，公平，可重现和 协作医学AI”为大规模临床实施提供基础。它将利用 Oneflorida是一家大型临床财团，由22家医院组成，在佛罗里达州为1000万患者提供服务，这是美国第三 最大的状态。总体目标将通过追求三个具体目标来实现。 （1）新颖的可解释，动态，可操作，公平和可再现的外部和前瞻性验证 实时手术风险监视的算法工具包。 （2）开发和评估可解释的AI平台 （xai-Idealist）使用人体基准的基准进行实时手术风险监视。 （3）实施 并评估具有先进隐私功能的联合学习方法，以实现协作手术风险 模型培训。该方法具有创新性，因为它代表了（1）建立第一个手术的第一次尝试 公平（可找到，可访问，可互操作，可重现）AI-Ready，大型多中心多模式数据集，（2） 通过评估公平性和可重复性来完成的新型计算方法，（3）多方面 以及全堆栈可解释的AI框架，以及（4）保护隐私模型的联合学习能力 培训越野机构。拟议的研究很重要，因为它将解决几个关键问题，并且 关键障碍，包括（1）缺乏AI-Ready大型手术数据集，（2）缺乏可解释的，动态的， 可操作，公平和可重复的手术风险算法，（2）缺乏医学AI解释平台，（4） 缺乏用于协作模型培训和跨机构共享的系统方法。最终， 预计结果将改善患者预后并降低住院成本，并终身 并发症。

项目成果

Cardiac and Vascular Surgery-Associated Acute Kidney Injury: The 20th International Consensus Conference of the ADQI (Acute Disease Quality Initiative) Group.

• DOI: 10.1161/jaha.118.008834
• 发表时间: 2018-06-01
• 期刊: Journal of the American Heart Association
• 影响因子: 5.4
• 作者: Nadim MK;Forni LG;Bihorac A;Hobson C;Koyner JL;Shaw A;Arnaoutakis GJ;Ding X;Engelman DT;Gasparovic H;Gasparovic V;Herzog CA;Kashani K;Katz N;Liu KD;Mehta RL;Ostermann M;Pannu N;Pickkers P;Price S;Ricci Z;Rich JB;Sajja LR;Weaver FA;Zarbock A;Ronco C;Kellum JA
• 通讯作者: Kellum JA

Computable Phenotypes to Characterize Changing Patient Brain Dysfunction in the Intensive Care Unit.

可计算表型来表征重症监护病房中不断变化的患者脑功能障碍。

• 发表时间: 2023
• 期刊: ArXiv
• 作者: Ren,Yuanfang;Loftus,TylerJ;Guan,Ziyuan;Uddin,Rayon;Shickel,Benjamin;Maciel,CarolinaB;Busl,Katharina;Rashidi,Parisa;Bihorac,Azra;Ozrazgat-Baslanti,Tezcan
• 通讯作者: Ozrazgat-Baslanti,Tezcan

The Pattern of Longitudinal Change in Serum Creatinine and 90-Day Mortality After Major Surgery.

大手术后血清肌酐和 90 天死亡率的纵向变化模式。

• DOI: 10.1097/sla.0000000000001362
• 发表时间: 2016-06
• 期刊: Annals of surgery
• 影响因子: 9
• 作者: Korenkevych D;Ozrazgat-Baslanti T;Thottakkara P;Hobson CE;Pardalos P;Momcilovic P;Bihorac A
• 通讯作者: Bihorac A

Postoperative acute kidney injury in adult non-cardiac surgery: joint consensus report of the Acute Disease Quality Initiative and PeriOperative Quality Initiative.

• DOI: 10.1038/s41581-021-00418-2
• 发表时间: 2021-09
• 期刊: Nature reviews. Nephrology
• 作者: Prowle JR;Forni LG;Bell M;Chew MS;Edwards M;Grams ME;Grocott MPW;Liu KD;McIlroy D;Murray PT;Ostermann M;Zarbock A;Bagshaw SM;Bartz R;Bell S;Bihorac A;Gan TJ;Hobson CE;Joannidis M;Koyner JL;Levett DZH;Mehta RL;Miller TE;Mythen MG;Nadim MK;Pearse RM;Rimmele T;Ronco C;Shaw AD;Kellum JA
• 通讯作者: Kellum JA

Intelligent Perioperative System: Towards Real-time Big Data Analytics in Surgery Risk Assessment.

• DOI: 10.1109/dasc-picom-datacom-cyberscitec.2017.201
• 发表时间: 2017-11
• 期刊: DASC-PICom-DataCom-CyberSciTech 2017 : 2017 IEEE 15th International Conference on Dependable, Autonomic and Secure Computing ; 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing ; 2017 IEEE 3rd International...
• 作者: Feng Z;Bhat RR;Yuan X;Freeman D;Baslanti T;Bihorac A;Li X
• 通讯作者: Li X