2018/19 Taught Postgraduate Module Catalogue

YCHI5065M Machine Learning for Health Data

15 creditsClass Size: 40

Module manager: David Wong
Email: d.c.wong@leeds.ac.uk

Taught: Semesters 1 & 2 (Sep to Jun) View Timetable

Year running 2018/19

Pre-requisite qualifications

Identical to student's parent taught postgraduate programme or PhD

Pre-requisites

NONE

Co-requisites

NONE

This module is mutually exclusive with

NONE

Module replaces

None

This module is not approved as an Elective

Module summary

Modern healthcare produces vast quantities of heterogenous data. These datasets are too large to be analysed using traditional means. One popular approach to this problem is machine learning to identify significant trends and patterns. This module is designed to help students understand the core processes for machine learning in healthcare, including the analysis of big data sets, feature selection, model selection, and generalisation. Students will also learn how specific machine learning algorithms work, and how they are used in the context of health data analytics.The module will equip students with the knowledge and skills necessary to analyse large datasets using machine learning algorithms and to critically evaluate their output and communicate the results to a range of audiences.

Objectives

The purpose of the module is to:
- Introduce students to key concepts that underpin machine learning for health data such that they can understand, select and proficiently use machine learning techniques for analysing data;
- Introduce students to a representative selection of the most common machine learning algorithms;
- Enable students to appropriately select and apply machine learning algorithms for the analysis of health data;
- Enable students to be able to critically evaluate the results of an analysis and present the results clearly and coherently.

Learning outcomes
By the end of the module students will be able to:
1. Understand and describe how machine learning techniques can be used for a wide range of healthcare applications
2. Critically understand and explain the differences between supervised and unsupervised learning
3. Create simple programs in a scientific programming language (Python)
4. Apply machine learning algorithms in a scientific programming language using appropriate toolkits
5. Devise a strategy for analysing a large dataset, undertake and present the results coherently, including the selection of appropriate machine learning techniques and critically interpret the results

Skills outcomes
Students will develop a working knowledge of scientific programming that is applicable to health data analytics

Syllabus

1. What is machine learning and why is it useful for healthcare?
2. Introduction to scientific programming
3. Supervised classification (including Decision Trees, K-nearest neighbour, and Support Vector Machines)
4. Unsupervised classification and novelty detection (including K-means clustering, Kernel Density Estimation, Gaussian Mixture models)
5. Feature Selection and dimensionality reduction
6. Training a machine learning model: how to deal with model complexity and results generalisation
7. Artificial neural networks and deep learning
8. Machine learning for image analysis

Teaching methods

Delivery type	Number	Length hours	Student hours
On-line Learning	4	2.00	8.00
Class tests, exams and assessment	1	2.00	2.00
Class tests, exams and assessment	4	0.25	1.00
Group learning	2	3.00	6.00
Lecture	10	1.00	10.00
Practical	3	1.00	3.00
Tutorial	10	1.00	10.00
Private study hours			110.00
Total Contact hours			40.00
Total hours (100hr per 10 credits)			150.00

Private study

Module pre-reading and directed exercises (20 hours)
- Basic statistics and data visualisation refresher
- Background reading on programming
- Installation of scientific programming language and development environment
- Preliminary programming exercises

During contact week (12 hours)
- Directed reading and exercises
- Formative quizzes to consolidate learning

After contact week (78 hours)
- Formative assessment
- Summative assessment

Opportunities for Formative Feedback

Group feedback on computer exercises during the contact week (individual, written and verbal)
Seminar discussions and short exercises (group feedback, verbal)
A formative peer and tutor assessed group verbal presentation on the final day of the contact week, with immediate feedback (individual feedback, verbal)
A formative assignment to be completed following the contact week before summative coursework due (individual feedback, written)

Methods of assessment

Coursework

Assessment type	Notes	% of formal assessment
Report	Written report on the analysis of a given set of datea (3,000 words)	100.00
Oral Presentation	Formative group verbal presentation	0.00
In-course MCQ	Formative quizzes during the contact week with immediate feedback	0.00
Computer Exercise	Computer exercises during the contact week with immediate feedback	0.00
Total percentage (Assessment Coursework)		100.00

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

Reading list

The reading list is available from the Library website

Last updated: 12/12/2018 10:48:54

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