Explainable deep learning models for healthcare - CDSS 3

Explainable deep learning models for healthcare - CDSS 3

本课程是 Informed Clinical Decision Making using Deep Learning 专项课程的一部分

位教师：Fani Deligianni

1,798 人已注册

包含在 Coursera Plus 中

了解更多

4个模块

深入了解一个主题并学习基础知识。

4.6

（15 条评论）

中级等级

推荐体验

3 周完成

在 10 小时一周

灵活的计划

自行安排学习进度

4个模块

深入了解一个主题并学习基础知识。

4.6

（15 条评论）

中级等级

推荐体验

3 周完成

在 10 小时一周

灵活的计划

自行安排学习进度

您将学到什么

Program global explainability methods in time-series classification
Program local explainability methods for deep learning such as CAM and GRAD-CAM
Understand axiomatic attributions for deep learning networks
Incorporate attention in Recurrent Neural Networks and visualise the attention weights

您将获得的技能

要了解的详细信息

可分享的证书

添加到您的领英档案

作业

5 项作业

授课语言：英语（English）

了解顶级公司的员工如何掌握热门技能

了解关于 Coursera for Business 的更多信息

Petrobras, TATA, Danone, Capgemini, P&G 和 L'Oreal 的徽标

积累特定领域的专业知识

本课程是 Informed Clinical Decision Making using Deep Learning 专项课程专项课程的一部分

在注册此课程时，您还会同时注册此专项课程。

向行业专家学习新概念
获得对主题或工具的基础理解
通过实践项目培养工作相关技能
获得可共享的职业证书

该课程共有4个模块

This course will introduce the concepts of interpretability and explainability in machine learning applications. The learner will understand the difference between global, local, model-agnostic and model-specific explanations. State-of-the-art explainability methods such as Permutation Feature Importance (PFI), Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanation (SHAP) are explained and applied in time-series classification. Subsequently, model-specific explanations such as Class-Activation Mapping (CAM) and Gradient-Weighted CAM are explained and implemented. The learners will understand axiomatic attributions and why they are important. Finally, attention mechanisms are going to be incorporated after Recurrent Layers and the attention weights will be visualised to produce local explanations of the model.

Deep learning models are complex and it is difficult to understand their decisions. Explainability methods aim to shed light to the deep learning decisions and enhance trust, avoid mistakes and ensure ethical use of AI. Explanations can be categorised as global, local, model-agnostic and model-specific. Permutation feature importance is a global, model agnostic explainabillity method that provide information with relation to which input variables are more related to the output.

涵盖的内容

6个视频8篇阅读材料1个作业1个讨论话题5个非评分实验室

6个视频总计71分钟

Welcome video - Explainable Deep Learning Models for Healthcare0分钟
Interpretability vs Explainability16分钟
'Explainability' in Healthcare Applications17分钟
Taxonomy of Explainability Methods17分钟
Model Agnostic Explainability Methods10分钟
Permutation Feature Importance in Time Series Data8分钟

8篇阅读材料总计390分钟

The importance of explainable prediction models in healthcare10分钟
Explainable Artificial Intelligence - Taxonomy60分钟
Model Agnostic Explainability30分钟
Permutation Feature Importance60分钟
Practical Exercise: Interpretability of the MLP model using Permutation Feature Importanceg60分钟
Practical Exercise: Interpretability of the CNN model using Permutation Feature Importance60分钟
Practical Exercise: Interpretability of the LSTM model using Permutation Feature Importance60分钟
Explainability models in ECG50分钟

1个作业总计30分钟

End of week 1 quiz30分钟

1个讨论话题总计10分钟

Week 1 - Your experience10分钟

5个非评分实验室总计50分钟

Permutation feature importance for classifying heart beats using a CNN10分钟
Light - Permutation feature importance for classifying heart beats using a CNN10分钟
Permutation feature importance for classifying heart beats using an LSTM10分钟
Light - Permutation feature importance for classifying heart beats using an LSTM10分钟
Permutation feature importance for classifying heart beats using a multi-layer perceptron10分钟

Local explainability methods provide explanations on how the model reach a specific decision. LIME approximates the model locally with a simpler, interpretable model. SHAP expands on this and it is also designed to address multi-collinearity of the input features. Both LIME and SHAP are local, model-agnostic explanations. On the other hand, CAM is a class-discriminative visualisation techniques, specifically designed to provide local explanations in deep neural networks.

涵盖的内容

5个视频7篇阅读材料1个作业1个讨论话题7个非评分实验室

5个视频总计47分钟

Local Interpretable Model Agnostic Explanations (LIME)9分钟
LIME in Time-Series Classification9分钟
Shapley Additive Explanations11分钟
Model-Specific Explanations: Visualisation Methods12分钟
CAM in Time-Series Classification4分钟

7篇阅读材料总计320分钟

Why Should I Trust You?30分钟
Practical Exercise: Interpretability of heartbeat classification using LIME and an NNMLP model60分钟
Practical Exercise: Interpretability of heartbeat classification using LIME and a CNN model60分钟
Practical Exercise: Interpretability of heartbeat classification using LIME and an LSTM model60分钟
A Unified Approach to Interpreting Model Predictions40分钟
Practical Exercise: Interpretability of CNN models using Class Activation Maps10分钟
Class Activation Mapping60分钟

1个作业总计30分钟

End of week 2 quiz30分钟

1个讨论话题总计10分钟

Week 2 - Your experience10分钟

7个非评分实验室总计65分钟

Interpretability of heartbeat classification using a CNN model and Class Activation Maps10分钟
Interpretability of heartbeat classification using a CNN model and CAM10分钟
LIME interpretability for heartbeat classification with a convolutional neural network10分钟
Interpretability of heartbeat classification using an LSTM model and CAM10分钟
LIME interpretability for heartbeat classification with a long short-term memory network10分钟
Light - LIME interpretability for heartbeat classification with a long short-term memory network5分钟
LIME interpretability for heartbeat classification with a multi-layer perceptron10分钟

GRAD-CAM is an extension of CAM, which aims to a broader application of the architecture in deep neural networks. Although, it is one of the most popular methods in explaining deep neural network decisions, it violates key axiomatic properties, such as sensitivity and completeness. Integrated gradients is an axiomatic attribution method that aims to cover this gap.

涵盖的内容

4个视频6篇阅读材料1个作业1个讨论话题7个非评分实验室

4个视频总计36分钟

Gradient Weighted Class Activation Maps11分钟
Grad-CAM in Time-Series Classification8分钟
Integrated Gradients12分钟
Integrated Gradients in Time Series Classification3分钟

6篇阅读材料总计310分钟

GRAD - Class Activation Mapping 30分钟
Practical Exercise: Interpretability of the CNN model using Gradient-weighted Class Activation Mapping60分钟
Practical Exercise: Interpretability of the LSTM model using Gradient-weighted Class Activation Mapping60分钟
Axiomatic Attribution for Deep Networks40分钟
Practical Exercise: Interpretability of the CNN model using Integrated Gradients60分钟
Practical Exercise: Interpretability of the LSTM model using Integrated Gradients60分钟

1个作业总计30分钟

End of week 3 quiz30分钟

1个讨论话题总计10分钟

Week 3 - Your experience10分钟

7个非评分实验室总计70分钟

Interpretability of heartbeat classification using a CNN model and Grad-CAM10分钟
Interpretability of heartbeat classification using integrated gradients and a CNN model10分钟
Light - Interpretability of heartbeat classification using integrated gradients and a CNN model10分钟
Interpretability of heartbeat classification using an LSTM model and Grad-CAM10分钟
Light - Interpretability of heartbeat classification using an LSTM model and Grad-CAM10分钟
Interpretability of heartbeat classification using integrated gradients and an LSTM model10分钟
Light - Interpretability of heartbeat classification using integrated gradients and an LSTM model10分钟

Attention in deep neural networks mimics human attention that allocates computational resources to a small range of sensory input in order to process specific information with limited processing power. In this week, we discuss how to incorporate attention in Recurrent Neural Networks and autoencoders. Furthermore, we visualise attention weights in order to provide a form of inherent explanation for the decision making process.

涵盖的内容

3个视频3篇阅读材料2个作业1个讨论话题4个非评分实验室

3个视频总计34分钟

Attention in Deep Learning13分钟
Taxonomy of Attention12分钟
Attention and Explainability8分钟

3篇阅读材料总计190分钟

Survey on Attention Mechanisms10分钟
Practical Exercise: Classification of heartbeats using an LSTM with attention mechanism90分钟
Practical Exercise: Interpretability of the LSTM model with attention mechanism90分钟

2个作业总计60分钟

End of week 4 quiz30分钟
End of course summative quiz30分钟

1个讨论话题总计10分钟

Week 4 - Your experience10分钟

4个非评分实验室总计30分钟

Interpretability of heartbeat classification using an LSTM model with attention mechanism10分钟
Light - Interpretability of heartbeat classification using an LSTM model with attention mechanism5分钟
Heartbeat classification using an LSTM model with attention mechanism10分钟
Light - Heartbeat classification using an LSTM model with attention mechanism5分钟

获得职业证书

将此证书添加到您的 LinkedIn 个人资料、简历或履历中。在社交媒体和绩效考核中分享。

位教师

Fani Deligianni

University of Glasgow

5 门课程5,880 名学生

提供方

University of Glasgow

从 Machine Learning 浏览更多内容

状态：免费试用
Duke University
Explainable Machine Learning (XAI)
课程
状态：免费试用
University of Illinois Urbana-Champaign
Deep Learning Methods for Healthcare
课程
状态：免费试用
University of Illinois Urbana-Champaign
Deep Learning for Healthcare
专项课程
状态：免费试用
University of Glasgow
Deep learning in Electronic Health Records - CDSS 2
课程

人们为什么选择 Coursera 来帮助自己实现职业发展

Felipe M.

自 2018开始学习的学生

''能够按照自己的速度和节奏学习课程是一次很棒的经历。只要符合自己的时间表和心情，我就可以学习。'

Jennifer J.

自 2020开始学习的学生

''我直接将从课程中学到的概念和技能应用到一个令人兴奋的新工作项目中。'

Larry W.

自 2021开始学习的学生

''如果我的大学不提供我需要的主题课程，Coursera 便是最好的去处之一。'

Chaitanya A.

''学习不仅仅是在工作中做的更好：它远不止于此。Coursera 让我无限制地学习。'

通过 Coursera Plus 开启新生涯

无限制访问 10,000+ 世界一流的课程、实践项目和就业就绪证书课程 - 所有这些都包含在您的订阅中

了解更多

通过在线学位推动您的职业生涯

获取世界一流大学的学位 - 100% 在线

探索学位

加入超过 3400 家选择 Coursera for Business 的全球公司

提升员工的技能，使其在数字经济中脱颖而出

了解更多

常见问题

To access the course materials, assignments and to earn a Certificate, you will need to purchase the Certificate experience when you enroll in a course. You can try a Free Trial instead, or apply for Financial Aid. The course may offer 'Full Course, No Certificate' instead. This option lets you see all course materials, submit required assessments, and get a final grade. This also means that you will not be able to purchase a Certificate experience.

When you enroll in the course, you get access to all of the courses in the Specialization, and you earn a certificate when you complete the work. Your electronic Certificate will be added to your Accomplishments page - from there, you can print your Certificate or add it to your LinkedIn profile.

Yes. In select learning programs, you can apply for financial aid or a scholarship if you can’t afford the enrollment fee. If fin aid or scholarship is available for your learning program selection, you’ll find a link to apply on the description page.