A qualitative research framework for the design of user-centered displays of explanations for machine learning model predictions in healthcare.
A qualitative research framework for the design of user-centered displays of explanations for machine learning model predictions in healthcare.
复制标题
DOI：
10.1186/s12911-020-01276-x
复制
发表时间：
2020-10-08
影响因子：
3.5
通讯作者：
Hochheiser H
中科院分区：
文献类型：
--
作者：
Barda AJ;Horvat CM;Hochheiser H
人们对临床预测工具越来越感兴趣，这些工具可以实现高预测准确性，并解释导致不良结局风险增加的因素。然而，解释复杂机器学习（ML）模型的方法很少被最终用户需求所告知，并且在医疗保健领域缺乏对模型可解释性的用户评估。我们使用了以前出版的理论框架的扩展修订，提出了一个以用户为中心的解释显示设计的框架。这个新的框架作为基础的定性调查和设计审查会议与重症监护护士和医生，通知设计的一个以用户为中心的解释显示为ML为基础的预测工具。我们使用我们的框架提出解释显示从儿科重症监护病房（PICU）住院死亡风险模型的预测。建议的显示是基于一个模型不可知的，实例级的解释方法的基础上的功能的影响，确定的Shapley值。焦点小组会议征求了重症监护提供者对拟议显示器的反馈，然后进行了相应的修改。建议的显示被认为是有用的工具，在评估模型的预测。然而，具体的解释目标和信息需求因临床角色和预测建模知识水平而异。提供者更喜欢解释显示，需要较少的信息处理工作，并可以支持各种用户的信息需求。据认为，提供辅助信息以协助解释对于促进提供者理解和接受预测和解释至关重要。以用户为中心的PICU院内死亡风险模型的解释显示结合了最初显示的元素，沿着了提供者建议的增强功能。我们提出了一个框架，用于设计以用户为中心的ML模型解释显示。我们使用所提出的框架，以激励设计一个以用户为中心的显示解释的预测，从一个PICU在医院的死亡风险模型。来自焦点小组参与者的积极反馈为使用模型不可知的、对特征影响的实例级解释作为理解医疗保健中ML模型预测的方法提供了初步支持，并推动了关于如何有效地将ML模型信息传达给医疗保健提供者的讨论。
There is an increasing interest in clinical prediction tools that can achieve high prediction accuracy and provide explanations of the factors leading to increased risk of adverse outcomes. However, approaches to explaining complex machine learning (ML) models are rarely informed by end-user needs and user evaluations of model interpretability are lacking in the healthcare domain. We used extended revisions of previously-published theoretical frameworks to propose a framework for the design of user-centered displays of explanations. This new framework served as the basis for qualitative inquiries and design review sessions with critical care nurses and physicians that informed the design of a user-centered explanation display for an ML-based prediction tool. We used our framework to propose explanation displays for predictions from a pediatric intensive care unit (PICU) in-hospital mortality risk model. Proposed displays were based on a model-agnostic, instance-level explanation approach based on feature influence, as determined by Shapley values. Focus group sessions solicited critical care provider feedback on the proposed displays, which were then revised accordingly. The proposed displays were perceived as useful tools in assessing model predictions. However, specific explanation goals and information needs varied by clinical role and level of predictive modeling knowledge. Providers preferred explanation displays that required less information processing effort and could support the information needs of a variety of users. Providing supporting information to assist in interpretation was seen as critical for fostering provider understanding and acceptance of the predictions and explanations. The user-centered explanation display for the PICU in-hospital mortality risk model incorporated elements from the initial displays along with enhancements suggested by providers. We proposed a framework for the design of user-centered displays of explanations for ML models. We used the proposed framework to motivate the design of a user-centered display of an explanation for predictions from a PICU in-hospital mortality risk model. Positive feedback from focus group participants provides preliminary support for the use of model-agnostic, instance-level explanations of feature influence as an approach to understand ML model predictions in healthcare and advances the discussion on how to effectively communicate ML model information to healthcare providers.
登录
查看更多内容
影响因子：
14.4
作者：
Miller, Tim
通讯作者：
Miller, Tim
影响因子：
28.1
作者：
Lundberg, Scott M.;Nair, Bala;Lee, Su-In
通讯作者：
Lee, Su-In
影响因子：
2.9
作者：
Holzinger, Andreas;Carrington, Andre;Mueller, Heimo
通讯作者：
Mueller, Heimo
影响因子：
0.9
作者：
Goodman, Bryce;Flaxman, Seth
通讯作者：
Flaxman, Seth
DOI：
10.13063/2327-9214.1181
发表时间：
2015-01-01
期刊：
EGEMS (Washington, DC)
影响因子：
--
作者：
Johnson, Tracy L;Brewer, Daniel;Batal, Holly
通讯作者：
Batal, Holly