Meet: Dr Liane dos Santos Canas

Dr Liane dos Santos Canas is an AI+ Fellow at King’s College London. These interdisciplinary fellowships are central to King’s £18 million strategic investment in academic excellence and demonstrate King's commitment to transforming AI research and innovation across all disciplines.

Liane grew up in Portugal, where she studied Biomedical Engineering and Biophysics at the University of Lisbon. She was drawn early to machine learning, though at the time it was far from the fashionable choice it is today. "It was not a trend to do it at the time," she recalls. "It was more linked to maths." She was particularly drawn to the brain, the most complex and least understood organ in the body, and for her, the most compelling research challenge. She pursued a master’s degree at the University of Cambridge before moving to London for a PhD in Medical Imaging at University College London. Her doctoral research focused on predicting the onset of prion disease, a rare and rapidly progressing neurological condition, before any symptoms appeared.

When her PhD supervisors moved to King’s College London, they invited her to join them. She accepted. Then the COVID-19 pandemic arrived. Rather than the deep learning work she’d planned, Liane found herself contributing to the COVID Symptom Study, a large-scale international project tracking how symptoms varied across different groups of people throughout the pandemic.

That experience led her to her current focus: ageing. Liane now works with TwinsUK, a large longitudinal study tracking how genetics, lifestyle and clinical factors interact as people get older. From this data, she’s developing computational models that build a picture of an individual and how their health is likely to evolve over time, tracking changes across organs and predicting future risks before symptoms emerge.

She describes these models as a flight simulator for health. "When a pilot is learning to fly, you don’t try right away on the real plane," she explains. "If you build a digital twin and you try to simulate how you are ageing over time, it gives you the chance to learn all of the possibilities, and to pick what is the best path to land the flight in the most gracious and efficient way." That might mean testing the impact of a dietary change, or identifying that someone’s trajectory suggests they should be screened sooner than planned. Rather than a snapshot at the moment of diagnosis, the approach tracks the full journey.

There’s a paradox at the heart of personalised medicine that Liane is direct about: to build a model that’s truly specific to you, you first need everyone else’s data. "We cannot just build a model for yourself," she says. "We don’t have anything as a comparison, or something we can use as a starting point. The starting point comes from the population that are similar to you." From that baseline, the model can begin to trace what makes your trajectory distinct.

For Liane, the answer to the current NHS crisis lies in prevention rather than treatment. Her models would help clinicians identify which patients need to be seen and when, so the most urgent cases are prioritised and others are supported to stay well for longer. However, she’s clear that none of this replaces clinicians.

The models process complex data and surface options. Clinicians bring the knowledge, judgement and human connection to decide what’s right for each person. What changes is how much of their time gets spent on analysis rather than on the people in front of them.

For Liane, what makes King’s the right environment is its direct connection to clinical practice. Its integration within hospital settings means that tools developed here can move from research into real use, tested and refined in the settings where they’ll actually matter. The AI+ Fellowship gives her the time, infrastructure and partnerships to make that happen.

Building the first brain ageing simulation could take around two years. Extending that to model how multiple organs interact over time will be a longer undertaking. But she’s clear about her goal.