Course

Deep learning in Electronic Health Records - CDSS 2

University of Glasgow

Delve into the principles of Deep Learning and its application in Electronic Health Records with the Deep learning in Electronic Health Records - CDSS 2 course. This comprehensive program explores the utilization of deep learning architectures, including Multi-layer perceptron, Convolutional Neural Networks, and Recurrent Neural Networks for time-series classification, particularly focusing on vital signals like ECG.

Throughout the course, you will tackle the complexities of EHR such as missing values and data heterogeneity, and learn effective imputation techniques and data encoding strategies. With a special emphasis on the MIMIC-III database, you will gain practical experience in formulating clinical prediction benchmarks from EHR data.

Gain insights into deep learning architectures and their application in EHR
Explore time-series classification for vital signals like ECG
Tackle challenges of missing values and data heterogeneity in EHR
Learn effective imputation techniques and data encoding strategies
Practical experience in formulating clinical prediction benchmarks from MIMIC-III database

Certificate Available ✔

Get Started / More Info

Deep learning in Electronic Health Records - CDSS 2

The course unfolds in four modules covering Artificial Intelligence and Multi-Layer Perceptron, Convolutional and Recurrent Neural Networks, Preprocessing and imputation of MIMIC III data, and EHR Encodings for machine learning models. Each module offers in-depth insights and practical exercises to enhance your understanding and skills in applying deep learning to EHR data.

Artificial Intelligence and Multi-Layer Perceptron

Module 1 - Artificial Intelligence and Multi-Layer Perceptron

Gain insights into Artificial Intelligence and Multi-Layer Perceptron
Explore the training and optimization of Multi-Layer Perceptron
Practical exercises for pre-processing and classification of ECG data

Convolutional and Recurrent Neural Networks.

Module 2 - Convolutional and Recurrent Neural Networks

Learn about Convolutional Neural Networks and Recurrent Neural Networks
Evaluate learning algorithms and practical exercises for classification using different models
Dimensionality reduction techniques to visualize ECG data

Preprocessing and imputation of MIMIC III data

Module 3 - Preprocessing and imputation of MIMIC III data

Benchmark Deep Learning Models with EHR and practical exercises in data extraction and preprocessing
Imputation strategies and practical exercises for in-hospital mortality prediction

EHR Encodings for machine learning models

Module 4 - EHR Encodings for machine learning models

Understand different encodings for categorical and continuous variables
Practical exercises in representation learning and similarity encodings
End of course summative quiz

Course

Deep learning in Electronic Health Records - CDSS 2

Course Modules

Artificial Intelligence and Multi-Layer Perceptron

Convolutional and Recurrent Neural Networks.

Preprocessing and imputation of MIMIC III data

EHR Encodings for machine learning models

More Machine Learning Courses

Preparing for Google Cloud Certification: Cloud Data Engineer

Language Classification with Naive Bayes in Python

Save, Load and Export Models with Keras

Procesamiento de Lenguaje Natural