2024/25 Taught Postgraduate Module Catalogue

YCHI5087M Artificial Intelligence and Machine Learning in Health

15 creditsClass Size: 40

Module manager: Dr Samuel D. Relton
Email: s.d.relton@leeds.ac.uk

Taught: Semester 2 (Jan to Jun) View Timetable

Year running 2024/25

Pre-requisite qualifications

First degree in a relevant subject e.g. Social Sciences, STEMM, Nursing (or equivalent) 2:1 OR previous work experience (minimum 2 years) of handling and/or analysing data

IELTS 7 – minimum of 6.5 in each component

Pre-requisites

NONE

Co-requisites

NONE

This module is mutually exclusive with

NONE

Module replaces

YCHI5065M

This module is not approved as an Elective

Module summary

In healthcare and other areas, there are increasingly vast quantities of complex, heterogeneous data produced. This includes medical images, electronic healthcare records, and genetic data. An increasingly popular method of analysing (and combining) such datasets is using machine learning to identify significant trends and patterns. This module will introduce students to a variety of different machine learning algorithms for supervised and unsupervised learning problems. These include random forest, support vector machines, k-means clustering, and neural networks with use-cases identified from across the healthcare domain. Students will also be introduced to techniques for feature selection, dimensionality reduction, and in avoiding overfitting. This builds upon knowledge gained in the core module on statistical modelling. By the end of the module, students will be familiar with a variety of alternative approaches to traditional statistical modelling and will have gained experience in using them within Python.

Objectives

Introduce students to methods for the appropriate training and testing of various machine learning models using healthcare data, whereupon a student could run an analysis independently and critique the work of others. Fundamental concepts of machine learning theory will also be covered, allowing students to expand their knowledge by reading the current literature.

Learning outcomes
1. Explain the use-case for machine learning within healthcare applications, and how this compares with more traditional statistical models.
2. Critically appraise scenarios where supervised and unsupervised learning approaches may be used.
3. Critically appraise health data analysis of others, identifying potential sources of bias and modelling issues.
4. Demonstrate a critical understanding by selecting and applying a range of machine learning approaches to data using Python and analyse the results appropriately.
5. Apply the technology underlying cutting-edge healthcare applications (i.e. automated diagnosis of disease from MRI imaging) and demonstrate a sophisticated understanding of the resources and tools available to advance their knowledge.

Syllabus

1. Definitions of machine learning in health and difference from traditional statistical models.
2. Supervised and unsupervised learning approaches (e.g. support vector machines, random forests, k-means clustering).
3. Feature engineering (e.g. one-hot encoding), dimensionality reduction, optimizers, avoiding overfitting.
4. Testing and comparing machine learning approaches.
5. Neural networks and deep learning.
6. Machine learning for image and signal analysis.

Teaching methods

Delivery type	Number	Length hours	Student hours
Group learning	5	1.00	5.00
Lecture	5	1.00	5.00
Practical	5	3.00	15.00
Seminar	5	1.00	5.00
Private study hours			120.00
Total Contact hours			30.00
Total hours (100hr per 10 credits)			150.00

Private study

120 hours.
Students are expected to read further detail on the topics discussed from the reading list and current literature on the subject.

Opportunities for Formative Feedback

The course will be run over a 1 week period with lectures in mornings and computer-based practical sessions in the afternoon. As such, students will have ample chance to discuss their progress with staff members each day. There will also be a short essay (500 words max) where students will explain their analysis plan for the summative assessment. This allows them to demonstrate their knowledge of material and obtain detailed feedback on their plans before commencing with the summative assignment.

Methods of assessment

Coursework

Assessment type	Notes	% of formal assessment
Essay	500 words. Formative assignment to create a statistical analysis plan explaining the way students intend to perform feature engineering, model development, and testing.	0.00
Report	3000 words. Project report. Analysis of a given dataset, which will require minimal data cleaning. Students asked to produce and compare models for the prediction of patient outcomes based on cross-sectional data using the best practice we have discussed in class.	100.00
Total percentage (Assessment Coursework)		100.00

Normally resits will be assessed by the same methodology as the first attempt, unless otherwise stated

Reading list

There is no reading list for this module

Last updated: 29/04/2024 16:15:43

Disclaimer

Browse Other Catalogues

• Undergraduate module catalogue
• Taught Postgraduate module catalogue
• Undergraduate programme catalogue
• Taught Postgraduate programme catalogue

Errors, omissions, failed links etc should be notified to the Catalogue Team.PROD