This course is part of the Informed Clinical Decision Making using Deep Learning Specialization

Fani Deligianni

Offered By

Informed Clinical Decision Making using Deep Learning SpecializationUniversity of Glasgow

About this Course

1,673 recent views

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.

Flexible deadlines

Reset deadlines in accordance to your schedule.

Shareable Certificate

Earn a Certificate upon completion

100% online

Start instantly and learn at your own schedule.

Course 3 of 5 in the

Informed Clinical Decision Making using Deep Learning Specialization

Intermediate Level

Python programming and experience with basic packages such as numpy, scipy and matplotlib

Approx. 30 hours to complete

English

Subtitles: English

What you will learn

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

Skills you will gain

attention mechanisms
explainable machine learning models
model-agnostic and model specific models
global and local explanations
interpretability vs explainability

Flexible deadlines

Reset deadlines in accordance to your schedule.

Shareable Certificate

Earn a Certificate upon completion

100% online

Start instantly and learn at your own schedule.

Course 3 of 5 in the

Informed Clinical Decision Making using Deep Learning Specialization

Intermediate Level

Python programming and experience with basic packages such as numpy, scipy and matplotlib

Approx. 30 hours to complete

English

Subtitles: English

Instructor

Fani Deligianni

School of Computing Science

451 Learners

5 Courses

Offered by

University of Glasgow

The University of Glasgow has been changing the world since 1451. It is a world top 100 university (THE, QS) with one of the largest research bases in the UK.

We are a member of the prestigious Russell Group of leading UK Universities with annual research income of more than £179m. The University’s #TeamUofG community is truly international with over 8000 staff and 28,0000 students from from more than 140 countries. A 2019 Time Out survey placed Glasgow in the top ten cities in the world. Ranked between Berlin and Paris, Glasgow was voted number one for both friendliness and affordability. Right now our dedicated community of staff, students and alumni is working to address the challenges of Covid-19 and understand how we can make life safer for everyone.

Syllabus - What you will learn from this course

Week

Week 1

9 hours to complete

Interpretable vs Explainable Machine Learning Models in Healthcare

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.

9 hours to complete

6 videos (Total 72 min), 8 readings, 1 quiz

6 videos

Welcome video - Explainable Deep Learning Models for Healthcare55s

Interpretability vs Explainability16m

'Explainability' in Healthcare Applications17m

Taxonomy of Explainability Methods17m

Model Agnostic Explainability Methods10m

Permutation Feature Importance in Time Series Data8m

8 readings

The importance of explainable prediction models in healthcare10m

Explainable Artificial Intelligence - Taxonomy1h

Model Agnostic Explainability30m

Permutation Feature Importance1h

Practical Exercise: Interpretability of the MLP model using Permutation Feature Importanceg1h

Practical Exercise: Interpretability of the CNN model using Permutation Feature Importance1h

Practical Exercise: Interpretability of the LSTM model using Permutation Feature Importance1h

Explainability models in ECG50m

1 practice exercise

End of week 1 quiz30m

Week

Week 2

8 hours to complete

Local Explainability Methods for Deep Learning Models

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.

8 hours to complete

5 videos (Total 48 min), 7 readings, 1 quiz

5 videos

Local Interpretable Model Agnostic Explanations (LIME)9m

LIME in Time-Series Classification9m

Shapley Additive Explanations11m

Model-Specific Explanations: Visualisation Methods12m

CAM in Time-Series Classification4m

7 readings

Why Should I Trust You?30m

Practical Exercise: Interpretability of heartbeat classification using LIME and an NNMLP model1h

Practical Exercise: Interpretability of heartbeat classification using LIME and a CNN model1h

Practical Exercise: Interpretability of heartbeat classification using LIME and an LSTM model1h

A Unified Approach to Interpreting Model Predictions40m

Practical Exercise: Interpretability of CNN models using Class Activation Maps10m

Class Activation Mapping1h

1 practice exercise

End of week 2 quiz30m

Week

Week 3

8 hours to complete

Gradient-weighted Class Activation Mapping and Integrated Gradients

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.

8 hours to complete

4 videos (Total 37 min), 6 readings, 1 quiz

4 videos

Gradient Weighted Class Activation Maps11m

Grad-CAM in Time-Series Classification8m

Integrated Gradients12m

Integrated Gradients in Time Series Classification3m

6 readings

GRAD - Class Activation Mapping 30m

Practical Exercise: Interpretability of the CNN model using Gradient-weighted Class Activation Mapping1h

Practical Exercise: Interpretability of the LSTM model using Gradient-weighted Class Activation Mapping1h

Axiomatic Attribution for Deep Networks40m

Practical Exercise: Interpretability of the CNN model using Integrated Gradients1h

Practical Exercise: Interpretability of the LSTM model using Integrated Gradients1h

1 practice exercise

End of week 3 quiz30m

Week

Week 4

5 hours to complete

Attention mechanisms in Deep Learning

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.

5 hours to complete

3 videos (Total 34 min), 3 readings, 2 quizzes

About the Informed Clinical Decision Making using Deep Learning Specialization

This specialisation is for learners with experience in programming that are interested in expanding their skills in applying deep learning in Electronic Health Records and with a focus on how to translate their models into Clinical Decision Support Systems.

The main areas that would explore are: Data mining of Clinical Databases: Ethics, MIMIC III database, International Classification of Disease System and definition of common clinical outcomes. Deep learning in Electronic Health Records: From descriptive analytics to predictive analytics Explainable deep learning models for healthcare applications: What it is and why it is needed Clinical Decision Support Systems: Generalisation, bias, ‘fairness’, clinical usefulness and privacy of artificial intelligence algorithms.

Informed Clinical Decision Making using Deep Learning

Frequently Asked Questions

When will I have access to the lectures and assignments?
Access to lectures and assignments depends on your type of enrollment. If you take a course in audit mode, you will be able to see most course materials for free. To access graded assignments and to earn a Certificate, you will need to purchase the Certificate experience, during or after your audit. If you don't see the audit option:
The course may not offer an audit option. 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.
What will I get if I subscribe to this Specialization?
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. If you only want to read and view the course content, you can audit the course for free.
Is financial aid available?
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.

More questions? Visit the Learner Help Center.

Explainable deep learning models for healthcare - CDSS 3