MACHINE LEARNING AND AI - 2023/4

Module Overview

Within biological, health, pharmaceutical, and medical research, now more than ever, there is a need for Big Data manipulation and the employment of Artificial Intelligence (AI) and advanced analytic approaches such as machine learning. The data produced in biomedical research can originate from many different sources (e.g. routinely collected primary and secondary care data, genomics, transcriptomics, proteomics,) and often these data are huge and can be challenging to manipulate and interrogate effectively. Nevertheless, when mined properly it offers exceptional insights into health and disease. Within this module we consider the means of analysing data using machine learning and other techniques in Artificial Intelligence, and the relevance of these to understanding health and disease. The module will provide students with an immersive introduction to modern machine learning approaches and the application of machine learning within the health and biomedical domains. Students will gain hands-on practical experience in the utilisation and application of advanced analytics methods such as machine learning. Students will further be equipped with transferable skills by focusing on state-of-the-art generic AI techniques that will go beyond aspects of healthcare research and which can be applied in other domains. The module will contextualise big data manipulation and analysis methods into the precision medicine landscape, population health, and other areas. The course will include lectures and practical, computational sessions covering topics to address the learning outcomes described below.

Module provider

School of Health Sciences

Module Leader

SPICK Matt (Health Sci.)

Number of Credits: 30

ECTS Credits: 15

Framework: FHEQ Level 7

Module cap (Maximum number of students): 35

Module Availability

Semester 2

Prerequisites / Co-requisites

BMSM033

Module content

While the terms AI and machine learning are often used interchangeably, and both are highly related, they do not mean exactly the same thing. Artificial Intelligence (AI) is the broader concept of machines being able to carry out tasks in a way that we would consider “smart” or “intelligent”; while machine learning is an implementation or approach in AI, that through processing of examples in data learns the patterns and connections that may be used for a variety of applications. The module on Machine Learning and AI will cover a range of fundamental as well as emerging topics related to this field. A grounded theoretical understanding, together with practical hands-on experience, will equip students with valuable skills applicable to Big Data analytics, in health and biomedicine, but also across a range of sectors.

 

This module will cover the following:

• Methods for high dimensional data and feature selection


• Regression and classification


• Supervised machine learning (basic concepts, ensemble methods)


• Neural networks and deep learning


• Clustering methods and their applications (e.g. partitioning-, hierarchical-, density-, grid-based)


• Understanding, evaluating and selecting models


• Multimodal data integration


• Ethics in machine learning and AI

Assessment pattern

Alternative Assessment

Students will be able to submit their presentation as a recorded video / PowerPoint. Allowing students to give their presentation to classmates as a formative assessment can be employed.

Assessment Strategy

The assessment strategy is designed to is allow students to take the knowledge acquired in lectures and guest seminars, with the addition of all taught practical exercises, to address specific questions in health and biomedicine, in relation to AI and machine learning. Students will primarily be assessed though two written assessments, each covering a portion of the topics listed above. Each written assessment will be founded on the development and application of the relevant tools and code, applied to relevant datasets with the aim of answering specific, defined research questions. Students will need to demonstrate their work by providing both the code used, as well as a synthesis of the results obtained. Furthermore, online quizzes will give students the opportunity to demonstrate their understanding of the topics, and through provided feedback, gauge on how they are progressing
Formative assessment goes hand in hand with summative assessment throughout this process.


The summative assessment for this module consists of the following:

• Three online quizzes (each accounting for 10% of the final grade): Students will complete three online quizzes over the course duration. These will be multiple choice, true /false, matching questions and provide feedback to students on how they are progressing (addressing learning outcomes 1, 3, and 4).


• Oral presentation (10% of the final grade): Students will prepare a short presentation on a machine learning method / technique of their choice – this will be presented as part of a dedicated workshop (addressing learning outcomes 3, 4, 5 and 6).


• Two data analysis programming-based assessment: Students will analyse real-world datasets using techniques in machine learning and AI, and present the findings in the form of a written piece of work of approximately 1500 words (each assignment will account for 30% of the final grade). These practical assessments will focus on use and implementation and evaluation of a set of tools for health care data with the focus on the selection of appropriate ML tools for data analysis (addressing learning outcomes 1, 2, 3, and 6).

• During lectures by question and answer sessions


• During supervised practical sessions


• Feedback on progress through the results from online quizzes


• Via feedback comments on assessed coursework and draft powerpoint presentations.

Module aims

• The aim of this module is for students to understand ¿state-of-the-art¿ machine learning, advanced quantitative analysis, and Artificial Intelligence approaches and their application to healthcare, health informatics, omics-based approaches and bioinformatics. Further, students will gain hands-on technical experience in applying such approaches to large health and biomedical datasets (e.g.UK Biobank, Physionet). In particular this module aims to:
• Introduce students to key concepts in machine learning and AI
• Demonstrate how to practically analyse real-world healthcare data using modern machine learning and AI techniques
• Introduce the students to best practices in machine learning and AI
• Discuss real-world ethical considerations of machine learning and AI, as well as ways to assess and evaluate such techniques
• Through research-orientated guest lectures from leading experts, students will develop an understanding of how machine learning technologies are applied to specific discipline areas. Students will have the opportunity to develop the practical skills required to analyse data using machine learning and other advanced analytics techniques in a series of hands-on computer-based workshops. This module will provide students with knowledge and experience of the multiple steps (e.g. experimental definition, data analysis and required interpretation plus quality control procedures) involved in using this suite of techniques in healthcare and biomedical research.

Learning outcomes

Methods of Teaching / Learning

The learning and teaching strategy is designed to equip students with the skills needed to become a successful data scientist by providing training in cutting-edge advanced analysis methodology, supported by an understanding of the theoretical underpinning of these. Furthermore, the module will inform students of the opportunities that machine learning and AI can offer in the biomedical and health research and practice. The objectives and learning outcomes of this module are designed to enable the students in implementing learning from other modules delivered through the MSc programme, namely Statistics and Modelling for Health Data, Big Data in Health and Biomedicine, and Reporting and Visualisation.

The module will guide students through the variety of machine learning and AI techniques being used in health, medical and biological research as well as in application to clinical practice.

Students will be encouraged to explore and demonstrate their gained theoretical and practical knowledge of machine learning and AI, within the contexts of health and biomedicine. Thus, the method is to make students aware of the opportunities available via machine learning analysis, giving them a training in the available tools and novel, cutting-edge developments, coupled with providing students with the critical thinking required to examine when machine learning is appropriately used and applied (and when not).

This module will employ lectures, practical computer-based session, workshops, and tutorial sessions as the main tools of teaching and learning. The lectures will provide a formal and intuitive overview of the modern machine learning and AI approaches with particular emphases on those relevant in healthcare research. Lectures will also cover best practices in scientific reporting and presentation, as well as ethical consideration of the work contributing to the development of global and cultural intelligence. 

Each of the lectures will be accompanied by hands-on practical sessions that will focus on applying and evaluating various machine learning techniques, including those typically used in health, medical and biological research. These practical sessions will provide students with training how to wrangle and analyse real health data using appropriate software commonly used in industry and research institutions (Python and R together with specific packages), which will contribute to enhancing student’s employability and digital capabilities. As an integral part of the weekly coursework, students will be asked to provide a short overview of the methods and results of their work emphasising the limitations of the method discussed. Students will be required to complete three online quizzes (described below) over the course duration providing them with a sense of progress on the fundamental understanding the core material. The hands-on practical sessions will also allow students to learn from each other providing space for collaboration, often imitating real work settings, which will contribute to the students’ resourcefulness and resilience.

The module will also include a workshop, during which students will give a short presentation on an advanced analytics approach, machine learning or AI method of choice. This workshop aims to enhance students’ critical thinking as well as presentation skills, confidence and self-assurance.
 

 The learning and teaching methods include:

• Lectures on core material combined with open discussion


• Hands on practical sessions (computer-based labs)


• A workshop, focused on student presentations and peer-to-peer feedback


• Online quizzes


• Written assessments (described in more detail below)

Indicated Lecture Hours (which may also include seminars, tutorials, workshops and other contact time) are approximate and may include in-class tests where one or more of these are an assessment on the module. In-class tests are scheduled/organised separately to taught content and will be published on to student personal timetables, where they apply to taken modules, as soon as they are finalised by central administration. This will usually be after the initial publication of the teaching timetable for the relevant semester.

Reading list

https://readinglists.surrey.ac.uk
Upon accessing the reading list, please search for the module using the module code: BMSM035

Other information

The MSc Health and Biomedical Informatics programme is committed to developing graduates with strengths in Employability, Digital Capabilities, Global and Cultural Capabilities, Sustainability, and Resourcefulness and Resilience. This module is designed to allow students to develop knowledge, skills, and capabilities in the following areas:

• Digital Capabilities: Enhancing digital capabilities is an integral element of this module. The ability to employ computer coding languages such as Python and R, which will be used throughout this module, and the development of skills in data analysis through hands-on computer-based workshops, students will gain a high level of digital capabilities.
• Employability: The knowledge of generic machine learning and other advanced data analysis techniques, coupled with the ability to apply these to a real-world data, is currently one of the most sought-after skillsets on the market, and across a range of sectors including the pharmaceutical industry, high-tech, biotech, and academia. Through lectures, workshops, and preparation of data analysis reports, students will gain transferrable data science skills that will support them in their future career.
• Global and Cultural Capacities: These will be enhanced by reflecting on the considerations and issues related to the ethical use of AI and machine learning, and in particular when it comes to human health - a highly debated topic in recent times. The module will provide students with knowledge related to the impact of high computational demand that modern machine learning methods require and their ramification for the society.
• Resourcefulness and Resilience: Teamworking, in multidisciplinary settings, is essential in this field, and will be fostered during the workshop and in practical sessions. The hands-on practical tools that will be covered by the core material will enhance students’ resourcefulness in their ability to apply data analytics to a wide range of problems. Further, a better grasp of data methods, will enhance confidence and self-assurance in those students who may initially find this field to be intimidating.

Please note that the information detailed within this record is accurate at the time of publishing and may be subject to change. This record contains information for the most up to date version of the programme / module for the 2023/4 academic year.