911±¬ÁÏÍø

Aims:

The aims of the module are: to provide a biomedical and healthcare domain specific knowledge of the applications of machine learning and artificial intelligence methodologies; to capacitate students to propose and design ML and AI solutions to solve specific problems in the biomedical and healthcare field and to argue the reasons that justify their proposals; and to capacitate students to be effective team players in interdisciplinary research groups within the biomedical and clinical sciences.

Intended learning outcome:

On successful completion of the module, a student will be able to:

1. Understand biomedical terminology, its relation to clinical pathways and the challenges in clinical decision support.
2. Understand how AI and ML algorithms can be employed to solve real-world clinical problems.
3. Apply data science tools to interpret healthcare datasets.
4. Implement and evaluate current state-of-the art ML and AI approaches on multi-modal biomedical and healthcare datasets.
5. Skilfully approach data challenges spanning different application domains and core learning tasks.
6. Develop and validate state-of-the art ML and AI approaches.

Indicative content:

The following is indicative of the topics the module will typically cover:

The module introduces students with the current trends in ML and AI applications across the large sector of subspecialties within the biomedical and clinical domains. With the continuous advances in medical imaging, genomics, wearable medical devices and countless other sources of clinical data, ML and AI have the potential to transform healthcare in profound and drastic ways. From association discovery, diagnosis, prognosis, treatment selection, to clinical services workflow delivery and financial outcome prediction; ML and AI are revolutionizing the way we think about diseases and well-being. The module will introduce application-specific knowledge necessary for successful hands-on interaction and critical thinking within interdisciplinary biomedical and clinical sciences teams. The module will highlight how the application domain has become a rich source of theoretical and practical challenges in ML and AI. Examples include the use of convolutional neural networks on digital pathology for diagnostic and outcome prediction and personalised treatment schemes. Other examples include how ML and AI approaches can provide relevant clinical decision support within well-established clinical pathways in high and low-resource settings.

Requisites:

To be eligible to select this module an optional or elective, a student must be registered on a programme and year of study for which it is formally available.