📌 Pinned story
We are excited to announce that our brand new #AI online course, commissioned by @NIHRresearch Clinical Research Network and co-developed with @Imperial_IGHI, is now available on the #NIHRLearn platform!

This is a self-paced training course aimed towards colleagues involved in #clinical #research & interested in AI. The course covers a. Background & context of AI in healthcare in UK, b. Demystifying technical AI concepts , c. Practical considerations, ethics & regulation of AI.


📢Ariane Duverdier #AI4Health PhD researcher gave a poster talk at the BSID conference in April 2022. This was a great opportunity to introduce a clinical audience to AI methods she is researching on, to design personalised prognostic tools for eczema skin conditions. She networked with clinicians to discuss some of her results and how they may impact clinical practice. @TanakaGroup #SystemsMeds


Kai Zhang attended the 6th AAAI International Workshop on Health Intelligence which showcased a broad range of research from the health intelligence community ranging from machine learning applications, to drug discovery and genome analysis, to theoretical works on algorithms. He presented his work on federated Cox models. The workshop not only forged new connections with other labs but also inspired ideas for future work.


Journey into the Brain, event in March 2022. It was fantastic to hear more about mind-controlled technology, about how utilising AI can improve surgical expertise and what a day of a brain surgeon’s life looks like.


Do you treat non-small cell lung cancer (NSCLC)? Interested in AI & Machine Learning approaches for improved prognostication? Check out the journal paper by Sumeet Hindocha, MD, et al., in The Lancet’s EBioMedicine .


At our AI Seminar on 9 March 2022, Annalu Waller PhD OBE is Professor of Human Communication Technologies and the Academic Lead for Computing at Dundee University, Scotland, talks about Supporting extended conversation for people using Augmentative and alternative communication (AAC). AAC attempts to augment natural speech, or to provide alternative ways to communicate for people with limited or no functional speech. Technology provides access to voice output and plays an important role in AAC. At the simplest level, people with complex communication needs (CCN) can cause a prestored message to be spoken by activating a single switch. At the most sophisticated level, literate users can generate novel text.


Alistair Weld’s PhD research is about automated robotic solution for intraoperative ultrasound for brain resection, in collaboration with the German Technical University Munich (TUM) and the German Aerospace Center (DLR). In January 2022 he went to meet the Teams in Germany and push forward their work on navigating the robotic arm control algorithms, gaining insight into how DLR operate and into the breadth of application and potential of robotic arms.


Neophytos Polydorou had 2 papers accepted to the 2021 IEEE CogMI – International Conference on Cognitive Machine Intelligence, one called An Empathetic AI Coach for Self-Attachment Therapy and the other Valence/Arousal Estimation of Occluded Faces from VR Headsets. At the research workshops he enjoyed the discussions with the audience about his papers. He said it was great to learn about various fields and technologies, such as Edge-AI, IoT, NLP, Quantum Computing, Networks, Cyber-security, and a platform for trustworthy, secure and reliable machine learning in a hospital setting.


Dimitar Georgiev, who started his PhD in October 2021, also attended NeurIPS 2021 and found that this conference was a major success for AI that showcased novel research in both fundamental and translational research. He explained that the conferences allowed him to explore various sessions which are and will be very useful for his own research on deep learning for cerebral organoid phenotyping and drug discovery. Talks and workshops on graph representation learning, graph neural networks, geometric deep learning, (variational) autoencoder architectures, transformers and Gaussian processes were also very interesting and inspiring for him.


NeurIPS is a popular leading AI conference our CDT students attend. Hadrien Reynaud attended NeurIPS 2021 to present a workshop paper on using Style Transfer and Reinforcement Learning to automatically find the 4-chamber view of the heart in CT volumes. The workshop also provided insights into early work in the medical imaging field. Hadrien was also keen to learn more about causality which is a growing topic in the field of machine learning and has not yet been applied to medical imaging.


Dimitar Georgiev attended the 2021 Nvidia GTC, a developer conference for the era of AI. He thought that this was a fantastic opportunity to join sessions where researchers, creators, IT and business leaders explored AI and HPC approaches in the field of healthcare. For example, he participated in talks on how to accelerate deep learning pipelines with HPC and distributed computing which are now easier than ever to apply with the development of packages and libraries, such as NVIDIA’s CUDA, CuNumeric, Triton, etc. He also joined discussions on how such methods can be used to accelerate AI-driven healthcare applications, most notably with NVIDIA’s new frameworks called Clara Discovery and Clara Holoscan, which help speed up drug discovery processes and medical imaging.


Doga Basaran, 2nd year PhD researcher, attended the MICCAI conference 2021 and its MSSEG-2 Challenge, the Educational Challenge of the conference, presenting a talk on ‘Cascaded Networks for new MS lesion detection’. The main goal of this challenge -which is led by the MICCAI Student Board- is to provide educational material for researchers in the field of medical image computing and computer-assisted interventions. Doga said it was exciting to see latest state-of-the-art methods and what different methods other teams developed.


Clinical PhD fellow Sumeet Hindocha, MD, participated in the Industry Xchange Retreat, which was combined with an Entrepreneurial Masterclass for the London Tech Week. This retreat was themed according to the government’s Industrial Strategy challenges and offered a unique opportunity for UKRI-funded researchers for enhanced professional development through interaction with industry.  As an AI for Healthcare fellow, Sumeet found that the workshops of the Forum HealthTech Summit were super useful as they could practice effective pitching of healthcare tech ideas. He said that he would recommend the retreat to other students.


Our student Reneira Seeamber won the MSK Accelerator Award 2021 for her PlayBack project!

PlayBack is an innovative technology geared towards answering an unmet clinical need: Improving self-management of lower back pain with a core activity detecting belt.

This award is worth £10,000 and offers industry-inspired product development support and management for the design and development of medical devices.


Our Centre, in collaboration with the AI Network, hosted the Symposium on AI in Future Health & Care in Autumn 2021. The goal of this policy symposium was to provide a forum for discussion on topics around the regulation for safe, effective and trusted use of AI.

Topics discussed included evidence based principles and standards, performance and evaluation pre-launch and in use, managing algorithm updates and evolution, whole system assessment and verification, bias and human factors, as well security and privacy risks. More details can be found on our Resources page.


In October 2021, UK’s Medicines and Healthcare products Regulatory Agency (MHRA) together with the U.S. Food and Drug Administration (FDA) and Health Canada have jointly identified 10 guiding principles that can inform the development of Good Machine Learning Practice (GMLP). These principles will help promote safe, effective, and high-quality medical devices that use artificial intelligence and machine learning (AI/ML). Details here.


As part of our research training programme, the Centre launched an educational course on AI Innovation & Regulation in July 2021. This was a three day event that covered the following topics:

• How regulatory processes work for medical devices
• Clinical regulatory landscape & adoption for AI
• Overcoming obstacles innovators face
• Pioneering in the health AI space
• How to build AI with highest levels of safety
• Behavioural differences in adoption of innovation

The course was delivered in collaboration with experts from academia and the industry, including Imperial Business School, Digital Health London and Hardian Health.

The final day of the course featured a Dragons’ Den style exercise, where students worked in teams to prepare and deliver a product pitch for a ‘Dragons’ panel.


This is a fantastic example of how our CDT students move fluidly across the disciplinary AI and healthcare boundaries.

Aizaan Anwar – one of our clinically trained CDT students – was invited to the British Neuro-Oncology Society (BNOS) Annual Meeting 2021 to give a talk on her work entitled: “BrainApp: Using near-patient sensing through a mobile app and machine learning in brain tumour patients.”

Read the abstract here. Great work, Aizaan.


Will Bolton – one of our 1st-year doctoral researchers – visited AIME Conference 2021. A large part of the conference focused on the importance of explainability, which will be considered more rigorously in his research moving forward. He said he heard interesting talks on this which included using decision trees to explain a model’s predictions in a hierarchical manner. Interesting papers were also presented related to timeseries models, leading to the suggestion of using attention ahead of an RNN
for a potential performance jump


Giannis Afentakis – 1st Year CDT student – attended the ATTD Virtual Meeting 2021 (Advanced Technologies & Treatment for Diabetes) and thought it was a great opportunity to learn about the state-of-the-art research and innovation in the field of Diabetes Technology. Clinicians and Industry leaders talked about AI-enabled applications. This session provided lots of insight and updates on recent research on AI and Decision Support Systems in Diabetes Management which is the topic of his PhD.


Student Alina-Irina Serban had a paper accepted in the JMIR Aging journal: “Smart Home Sensing and Monitoring in Households With Dementia: User-Centered Design Approach”.

The study proposes AI-powered technologies to help make living environment safe and appropriate for people living with dementia, taking into account medical and psychological needs of the people.


Clinical PhD Fellow Kavitha Vimalesvaran, MD, had a paper accepted at the European Heart Journal:

Patterns of myocardial injury in recovered troponin-positive COVID-19 patients assessed by cardiovascular magnetic resonance.”

Congratulations Kavitha!


Reneira Seeamber won a place on the MedTech SuperConnector accelerator programme for her novel idea of an AI-assisted wearable belt and interactive gaming interface to enable real-time correction of posture to ameliorate chronic back pain.

Her proposed concept centres on a patented, practical and low-cost muscle activity detection suite.

This is a fantastic example of how AI technologies hold enormous potential in health and care!


In response to the Covid-19 emergency, clinical PhD Fellow Sumeet Hindocha and his team initiated a new study on how to use AI to identify to what extent changes seen on chest scans are due to coronavirus or the side effects of cancer treatment.

They said that a lot of the work they are doing is using AI on CT scan images to identify subtle changes in patients, which might help understand why one group of patients behave in a certain way. Read the story here.


In Spring 2020, Clinical PhD Fellow Dr Sameer Zaman volunteered to treat and look after patients at Imperial College NHS Trust hospitals. He quickly realised that there was a need for mobile devices to help patient better contact with their families in times when they were isolated. Sameer then launched the #Virtual Visiting initiative to call for donations of tablets of phones to help patients who may be isolated, at a time when they may be feeling unwell.


Ruby Sedgwick had a paper accepted at the Machine Learning for Molecules Workshop at NeurIPS 2020: Design of Experiments for Verifying Biomolecular Networks“.

This work looks into the design biomolecular networks using a Bayesian optimisation strategy.


The UKRI Centre in AI for Healthcare opened its doors in October 2019 when the first cohort of researchers started their PhD. UKRI awarded over
£7 million to train leaders in artificial intelligence. We are delighted that we are training PhD students in both disciplines Artificial Intelligence and Healthcare and push this ground-breaking research further. Read the introductory story here.