| Unit name | Data for Healthcare and AI |
|---|---|
| Unit code | BRMSM0096 |
| Credit points | 20 |
| Level of study | M/7 |
| Teaching block(s) |
Teaching Block 4 (weeks 1-24) |
| Unit director | Dr. Jon Lees |
| Open unit status | Not open |
| Units you must take before you take this one (pre-requisite units) |
None |
| Units you must take alongside this one (co-requisite units) |
None |
| Units you may not take alongside this one |
None |
| School/department | Bristol Medical School |
| Faculty | Faculty of Health and Life Sciences |
Why is this unit important?
Data is the fuel for AI, with many of the breakthrough coming from generation of publicly accessible, large scale annotated datasets. For example, the rise of ImageNet (a large, labeled image dataset) was pivotal for breakthroughs in computer vision, enabling models like AlexNet to outperform traditional approaches. AI holds great promise form medicine and health, but there are many issues with the associated datasets (e.g. privacy, complexity, bias, noise etc.) that need to be addressed for building effective tools. The unit will give you a deeper understanding of the source medical datasets and the practical considerations for making use of them for successful AI development for patients.
You will explore the main data types currently driving AI innovations in medicine and health. We will examine some core technologies used to generate these data types and the unique challenges associated with each. The unit will give you more confidence for sourcing and using data for AI and critically evaluate the appropriateness for a given medical AI/health tool. We will cover various practical aspects for successful AI innovation from medical/healthcare data looking at limitations and common pitfalls to avoid. You’ll gain insights into career pathways in AI for different Health datasets through mentoring and talks.
How does this unit fit into your programme of study
Starting from the healthcare data itself you will look at how it can be obtained and used successfully for successfully developing healthcare/medicine AI models. The unit allows you to apply more general skills and gain key insights about real world applications of AI in medicine and healthcare and progress towards your planned career path.
Overview of Content
In this unit, you will be introduced to examples of medical and healthcare datasets and how these has been successfully used to power AI methods. For example, the ongoing digitisation of histopathology is leading to new datasets that are powering a step change in the downstream AI capabilities, producing rapid advances for clinicians and patients. You will learn how to independently source relevant healthcare datasets, through platforms such as Kaggle, PhysioNet, or open NHS repositories, and understand the specific ongoing challenges (ethical, governance, infrastructure etc.) in accessing and harnessing medical / healthcare data.
Through hands-on practical sessions, you will learn how to identify and overcome various issues with medical/health data for AI. We will cover the importance of data quality and the many dataset pitfalls that you need to be aware of to successfully deliver value to patients. Through notebook-based practical’s you will identify dataset issues across a range of scenarios. Actively engaging with the medical datasets will give you an appreciation of how it forms the foundation for all the other downstream steps of AI including model quality, robustness and confidence. You will see how increasingly large datasets are becoming available and being combined to produce rich and complex features for AI. This complexity in turn is producing a great need for methods to explain the source features in the data powering the AI’s clinical suggestions and we explore the importance of explainable AI in healthcare.
Additionally, you will benefit from additional talks providing a wider background of experts, across both public and private sectors for a wide range of subjects around data for AI including careers and legal aspects.
How Will You Be Different After This Unit?
By the end of this unit, you will be familiar with key datasets and how they have been successfully translated from hospital data silos to power new AI tools helping patient care (e.g. radiology and histopathology datasets). You will be able to identify and apply appropriate methods for different healthcare datatypes. You will have improved your ability to source and manage real-world healthcare datasets and gained greater awareness of the ethical, technical, and contextual issues unique to the field including costs and benefits of these systems. You will have developed your critical appraisal skills, helping you to assess existing medical-AI tools or build your own. Through internal and external experts speakers, you will have a greater awareness of important factors (e.g. commercialisation and legal aspects) to help support your future career.
Learning Outcomes:
At the end of the unit, a successful student will be able to:
Teaching will be delivered through a combination of synchronous and asynchronous sessions, including lectures, lectorials, practical activities and self-directed exercises.
Tasks which help you learn and prepare you for summative tasks (formative):
The practical sessions will prepare you for the summative tasks. Tutors provide support during the tutorials to help you complete the tasks and provide feedback. In lectures, you will be encouraged to be an active learner through MCQ’s associated activities. In class reflections, with feedback from tutors and peers, on datasets and papers will help you cement the core critical evaluation skills needed when working with AI tools.
Summative assessment
Report (100%):
The assessment will consist of a report that critically evaluates an application for AI in medicine and health identifying the works strengths, weaknesses and potential errors. Through a notebook-based challenge you will suggest solutions that improve the proposed work, leading to improvements in the reliability and performance of the method for patients (LO 1-3).
When assessment does not go to plan
If you do not pass the unit, you will normally be given the opportunity to take a reassessment as per the Regulations and Code of Practice for Taught Programmes. Decisions on the award of reassessment will normally be taken after all taught units of the year have been completed. Reassessment will be in a similar format to the original assessment that has been failed.
If this unit has a Resource List, you will normally find a link to it in the Blackboard area for the unit. Sometimes there will be a separate link for each weekly topic.
If you are unable to access a list through Blackboard, you can also find it via the Resource Lists homepage. Search for the list by the unit name or code (e.g. BRMSM0096).
How much time the unit requires
Each credit equates to 10 hours of total student input. For example a 20 credit unit will take you 200 hours
of study to complete. Your total learning time is made up of contact time, directed learning tasks,
independent learning and assessment activity.
See the University Workload statement relating to this unit for more information.
Assessment
The assessment methods listed in this unit specification are designed to enable students to demonstrate the named learning outcomes (LOs). Where a disability prevents a student from undertaking a specific method of assessment, schools will make reasonable adjustments to support a student to demonstrate the LO by an alternative method or with additional resources.
The Board of Examiners will consider all cases where students have failed or not completed the assessments required for credit.
The Board considers each student's outcomes across all the units which contribute to each year's programme of study. For appropriate assessments, if you have self-certificated your absence, you will normally be required to complete it the next time it runs (for assessments at the end of TB1 and TB2 this is usually in the next re-assessment period).
The Board of Examiners will take into account any exceptional circumstances and operates
within the Regulations and Code of Practice for Taught Programmes.
