Training vision

The UKRI Centre for Doctoral Training in Artificial Intelligence for Healthcare at Imperial focusses on healthcare applications of core AI to train AI PhDs and Clinical PhD Fellows. We will deliver training integrating the development of technical skills with an appreciation for approaches to human-in-the-Loop AI design that are socially and ethically acceptable. The term “AI” means for us the development of intelligent systems that embody a practical solution. Amongst contemporary AI approaches, Machine Learning methods have shown to yield powerful solutions which work in purely data-driven manners and link via data science to emerging biomedical research methodologies. However, practical solutions involving AI will require a broader approach and we will drive technical innovation by providing broad training for exploitation of multiple technological strategies within the broader realm of AI, including, Machine Learning, Logic-based, Computer Vision or Natural Language Processing methods.

Expert Mentors

The exposure of our students to our expert mentors in related areas such as computer vision (e.g. for digital pathology and diagnostics) as well as sensing and wearable technologies (e.g. mobile mental health and digital biomarkers) will enhance the depth of training we can offer. Ultimately, no technological progress can help patients unless it is transformed into a (regulated) product or service. The UKRI CDT in AI for Healthcare plays to the established strength of the investigators and the whole supervisory team in leading cross-disciplinary research spanning AI and healthcare. Students will have at least two supervisors one with AI and one with biomedical background, and the two supervisors will be normally from two different Faculties/Departments – thus following our tried-and-tested crossdisciplinary training tradition at Imperial. The Centre will leverage the research excellence of its 100+ participating potential PhD supervisors across Imperial and our partners.


We have partnered closely with patient-focused charities, NHS trusts and a range of industries to ensure that students are trained in how to entrepreneurially realise impact from their work and understand how to navigate the regulatory pathways. One approach to delivering this will be for our PhD students to form co-creation teams including world-leading AI researchers, clinicians across 3 NHS trusts through our Academic Health Science Centre (Imperial Healthcare NHS Trust, Royal Brompton & Harefield NHS Foundation Trust, Royal Marsden NHS Trust), industry stakeholders and patient organisation partners. Moreover, we will embed clinicians as Clinical PhD Fellows into the centre’s PhD cohorts to foster a shared understanding between individuals developing and applying AI approaches and those using the technology offering a unique level of cognitive diversity and complementary in skills that will stimulate students to learn from each other in our co-training groups create a unique environment for research and training.

Integrated Training

Our large and diverse PhD cohorts regular PhD students and clinical PhD Fellows will benefit from an integrated training program and the ability to offer our students a unique sandbox in which to build their PhD success from AI and Healthcare applications. This is enabled through a PhD co-creation process by pairing students with AI and healthcare supervisors as well as industry and healthcare partners. The 4-year long PhD training programme (the programme duration for clinical PhD fellows is 3 years) is split into three phases that provide underpinning skill training (Foundation phase), research training (Research Phase) and finally drive PhD impact training (Impact phase).

The Foundation Phase – Months 1-6

The Foundation Phase starts with a Welcome Week with plenty of cohort-building and social activities. The students and their project teams will start their regular meetings and the student-led co-training teams will define their outline topics in the first month. The underpinning AI training is delivered by taking AI-relevant taught modules and the Research Tutorial. The Foundation Phase also sees the completion of the Research Planning Report. AI Foundation Courses AI foundation courses will be selected from established degree programmes and allocated taking the student‘s background and project aims into account. For example, in 2019-20 the following modules were available to select from: Mathematics for Machine Learning, Introduction to Machine Learning, and Deep Learning. When taking the taught modules, students are required to complete and pass one piece of coursework per module but do not have to take any exams. The Research Tutorial is a bespoke module which is core for all AI4Health students. The aim of this course to evaluate research and to develop critical research thinking. Research Planning Report Student will submit their Research Planning Report within 10-12 weeks of starting. The research plan is an important document to support defining the PhD project. Professional skills development training All PhD students are also required to complete the Professional skills programme of as prescribed by the College’s Graduate School. Details are here:

The Research Phase – Months 6-42 During the Research Phase students will focus on their research project with their co-creation team. Technical training will shift to project-based training, emphasis reflected in the technical masterclasses as well as the activities of their students AI-centred co-training teams (e.g. journal tutorials). The Research phase sees two major milestones which need to be completed and passed, according to Imperial College’s academic milestones’ policy in order to progress the PhD registration. Early Stage Assessment (ESA) All PhD students are required to complete the Early Stage Assessment (ESA), including ethical and regulatory implications of their research, as well as preliminary results of their work. The ESA milestone must be completed within 12 months of the initial registration. It consists of a comprehensive report and a seminar. The ESA report will be structured and formatted as a journal or review journal paper as determined by the student and their co-creation team, to a standard as if it were to be submitted. Late Stage Review (LSR) The LSR must be completed within 20-24 months of the initial registration. Students will be required to present a poster detailing achievements and a plan to finish and give a brief oral presentation. Towards the end of the Research Phase, students will be encouraged to submit three completed PhD chapters.

The Impact Phase – Months 42-48 In this phase, monitoring the write-up progress and activity proposed in the impact plan takes centre stage. During the annual Centre conference, students will have the opportunity to prepare a completion talk, presenting their work to a relevant audience. The maximum thesis submission date within 48 months of registration. Throughout studies, Imperial College Academic and Examination Regulations for Research Degrees apply.

Student-led training will organise the cohort in co-training teams. Cohort cohesion after year 1 is maintained by regular cohort wide meetings, training events and activities. Our Clinical CDT Fellows will receive clinical career training as needed and organised through the Imperial Clinical Academic Training Office (CATO) which allows us to integrate Clinical Fellows seamlessly into the CDT cohort training. Thus, all students will be able to go through training together as one cohort. The unique cohort size enable us to harness economies of scale to develop a bespoke training program involving different domains, such as regulatory training via the NHS Digital Academy, or our start-up building and launch via partner Agorai, which would otherwise not have been possible.